+ OUTPUT_DIR=./upload-gemma_4_31B_it + LLAMA_CPP=./llama.cpp + DISPLAY_NAME=gemma-4-31B-it + QUANTIZE=./llama.cpp/build/bin/llama-quantize + python3 ./llama.cpp/convert_hf_to_gguf.py ./model-temp-gemma_4_31B_it-PRIMARY --outtype bf16 --outfile ./upload-gemma_4_31B_it/gemma-4-31B-it-BF16.gguf --model-name gemma-4-31B-it INFO:hf-to-gguf:Loading model: model-temp-gemma_4_31B_it-PRIMARY INFO:hf-to-gguf:Model architecture: Gemma4ForConditionalGeneration INFO:hf-to-gguf:gguf: loading model weight map from 'model.safetensors.index.json' INFO:hf-to-gguf:gguf: indexing model part 'model-00001-of-00002.safetensors' INFO:hf-to-gguf:gguf: indexing model part 'model-00002-of-00002.safetensors' INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only INFO:hf-to-gguf:Exporting model... INFO:hf-to-gguf:rope_freqs.weight, torch.float32 --> F32, shape = {256} INFO:hf-to-gguf:token_embd.weight, torch.bfloat16 --> BF16, shape = {5376, 262144} INFO:hf-to-gguf:blk.0.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.0.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.0.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.0.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.0.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.0.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.0.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.0.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.0.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.1.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.1.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.1.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.1.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.1.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.1.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.1.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.1.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.1.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.10.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.10.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.10.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.10.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.10.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.10.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.10.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.10.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.10.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.12.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.12.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.12.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.12.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.12.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.12.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.12.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.12.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.12.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.12.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.13.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.13.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.13.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.13.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.13.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.13.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.13.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.13.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.13.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.13.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.14.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.14.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.14.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.14.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.14.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.14.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.14.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.14.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.14.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.14.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.15.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.15.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.15.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.15.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.15.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.15.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.15.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.15.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.15.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.15.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.16.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.16.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.16.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.16.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.16.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.16.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.16.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.16.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.16.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.18.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.18.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.18.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.18.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.18.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.18.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.18.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.18.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.18.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.18.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.19.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.19.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.19.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.19.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.19.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.19.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.19.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.19.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.19.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.19.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.2.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.2.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.2.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.2.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.2.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.2.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.2.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.2.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.2.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.20.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.20.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.20.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.20.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.20.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.20.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.20.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.20.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.20.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.20.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.21.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.21.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.21.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.21.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.21.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.21.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.21.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.21.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.21.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.21.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.22.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.22.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.22.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.22.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.22.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.22.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.22.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.22.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.22.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.24.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.24.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.24.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.24.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.24.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.24.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.24.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.24.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.24.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.24.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.25.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.25.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.25.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.25.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.25.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.25.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.25.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.25.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.25.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.25.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.26.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.26.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.26.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.26.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.26.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.26.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.26.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.26.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.26.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.26.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.27.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.27.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.27.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.27.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.27.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.27.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.27.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.27.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.27.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.27.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.28.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.28.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.28.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.28.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.28.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.28.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.28.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.28.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.28.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.3.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.3.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.3.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.3.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.3.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.3.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.3.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.3.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.3.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.30.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.30.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.30.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.30.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.30.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.30.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.30.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.30.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.30.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.30.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.31.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.31.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.31.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.31.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.31.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.31.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.31.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.31.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.31.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.31.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.32.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.32.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.32.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.32.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.32.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.32.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.32.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.32.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.32.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.32.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.33.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.33.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.33.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.33.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.33.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.33.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.33.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.33.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.33.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.33.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.34.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.34.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.34.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.34.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.34.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.34.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.34.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.34.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.34.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.36.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.36.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.36.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.36.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.36.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.36.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.36.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.36.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.36.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.36.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.37.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.37.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.37.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.37.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.37.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.37.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.37.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.37.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.37.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.37.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.38.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.38.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.38.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.38.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.38.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.38.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.38.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.38.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.38.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.38.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.39.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.39.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.39.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.39.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.39.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.39.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.39.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.39.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.39.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.39.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.4.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.4.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.4.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.4.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.4.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.4.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.4.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.4.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.4.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.4.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.40.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.40.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.40.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.40.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.40.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.40.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.40.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.40.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.40.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.42.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.42.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.42.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.42.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.42.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.42.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.42.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.42.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.42.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.42.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.43.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.43.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.43.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.43.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.43.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.43.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.43.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.43.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.43.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.43.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.44.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.44.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.44.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.44.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.44.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.44.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.44.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.44.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.44.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.44.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.45.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.45.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.45.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.45.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.45.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.45.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.45.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.45.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.45.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.45.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.46.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.46.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.46.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.46.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.46.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.46.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.46.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.46.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.46.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.48.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.48.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.48.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.48.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.48.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.48.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.48.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.48.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.48.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.48.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.49.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.49.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.49.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.49.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.49.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.49.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.49.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.49.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.49.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.49.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.50.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.50.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.50.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.50.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.50.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.50.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.50.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.50.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.50.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.50.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.51.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.51.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.51.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.51.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.51.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.51.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.51.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.51.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.51.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.51.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.52.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.52.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.52.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.52.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.52.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.52.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.52.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.52.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.52.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.54.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.54.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.54.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.54.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.54.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.54.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.54.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.54.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.54.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.54.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.55.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.55.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.55.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.55.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.55.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.55.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.55.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.55.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.55.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.55.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.56.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.56.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.56.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.56.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.56.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.56.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.56.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.56.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.56.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.56.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.57.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.57.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.57.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.57.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.57.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.57.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.57.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.57.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.57.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.57.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.58.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.58.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.58.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.58.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.58.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.58.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.58.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.58.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.58.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.6.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.6.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.6.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.6.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.6.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.6.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.6.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.6.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.6.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.6.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.7.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.7.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.7.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.7.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.7.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.7.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.7.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.7.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.7.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.7.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.8.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.8.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.8.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.8.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.8.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.8.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.8.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.8.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.8.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.8.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.9.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.9.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.9.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.9.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.9.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.9.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.9.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.9.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.9.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.9.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:output_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.10.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.11.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.11.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.11.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.11.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.11.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.11.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.11.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.11.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.16.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.16.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.16.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.16.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.16.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.17.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.17.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.17.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.17.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.17.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.17.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.17.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.17.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.22.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.22.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.22.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.22.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.22.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.23.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.23.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.23.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.23.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.23.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.23.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.23.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.23.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.28.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.28.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.28.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.28.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.28.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.29.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.29.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.29.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.29.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.29.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.29.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.29.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.29.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.34.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.34.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.34.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.34.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.34.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.35.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.35.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.35.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.35.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.35.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.35.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.35.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.35.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.4.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.40.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.41.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.41.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.41.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.41.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.41.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.41.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.41.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.41.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.46.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.46.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.46.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.46.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.46.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.47.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.47.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.47.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.47.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.47.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.47.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.47.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.47.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.5.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.5.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.5.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.5.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.5.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.5.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.5.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.5.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.5.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.52.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.52.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.52.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.52.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.52.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.53.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.53.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.53.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.53.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.53.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.53.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.53.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.53.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.58.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.58.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.58.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.58.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.58.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.59.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.59.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.59.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.59.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.59.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.59.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.59.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.59.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:gguf: context length = 262144 INFO:hf-to-gguf:gguf: embedding length = 5376 INFO:hf-to-gguf:gguf: feed forward length = 21504 INFO:hf-to-gguf:gguf: head count = 32 INFO:hf-to-gguf:gguf: key-value head count = 16 WARNING:hf-to-gguf:Unknown RoPE type: proportional INFO:hf-to-gguf:gguf: rope scaling type = NONE INFO:hf-to-gguf:gguf: rope theta = 1000000.0 INFO:hf-to-gguf:gguf: rope theta swa = 10000.0 INFO:hf-to-gguf:gguf: rms norm epsilon = 1e-06 INFO:hf-to-gguf:gguf: file type = 32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.context_length', overwriting it with new value 262144 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.head_count', overwriting it with new value 32 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.layer_norm_rms_epsilon', overwriting it with new value 1e-06 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.key_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.value_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.rope.freq_base', overwriting it with new value 1000000.0 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.head_count_kv', overwriting it with new value 16 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.key_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.value_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.head_count_kv', overwriting it with new value [16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4] of type ARRAY INFO:hf-to-gguf:Set model quantization version INFO:hf-to-gguf:Set model tokenizer INFO:hf-to-gguf:Token '<|tool_call>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|tool_response>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|"|>' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|channel>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:gguf.vocab:Adding 514906 merge(s). INFO:gguf.vocab:Setting special token type bos to 2 INFO:gguf.vocab:Setting special token type eos to 1 INFO:gguf.vocab:Setting special token type unk to 3 INFO:gguf.vocab:Setting special token type pad to 0 INFO:gguf.vocab:Setting special token type mask to 4 INFO:gguf.vocab:Setting chat_template to {# Template: Google Gemma 4 Canonical Chat Template Author: Google Gemma Engineering Team Published: 2026-07-09 Context: Fixed tool-calling loops, turn closures, and thinking content-ordering. #} {%- macro format_parameters(properties, required, filter_keys=false) -%} {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%} {%- set ns = namespace(found_first=false) -%} {%- for key, value in properties | dictsort -%} {%- set add_comma = false -%} {%- if not filter_keys or key not in standard_keys -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {{ key }}:{ {%- if value['description'] -%} description:<|"|>{{ value['description'] }}<|"|> {%- set add_comma = true -%} {%- endif -%} {%- if value['type'] | upper == 'STRING' -%} {%- if value['enum'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} enum:{{ format_argument(value['enum']) }} {%- endif -%} {%- elif value['type'] | upper == 'ARRAY' -%} {%- if value['items'] is mapping and value['items'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} items:{ {%- set ns_items = namespace(found_first=false) -%} {%- for item_key, item_value in value['items'] | dictsort -%} {%- if item_value is not none -%} {%- if ns_items.found_first %},{% endif -%} {%- set ns_items.found_first = true -%} {%- if item_key == 'properties' -%} properties:{ {%- if item_value is mapping -%} {{- format_parameters(item_value, value['items']['required'] | default([])) -}} {%- endif -%} } {%- elif item_key == 'required' -%} required:[ {%- for req_item in item_value -%} <|"|>{{- req_item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- elif item_key == 'type' -%} {%- if item_value is string -%} type:{{ format_argument(item_value | upper) }} {%- else -%} type:{{ format_argument(item_value | map('upper') | list) }} {%- endif -%} {%- else -%} {{ item_key }}:{{ format_argument(item_value) }} {%- endif -%} {%- endif -%} {%- endfor -%} } {%- endif -%} {%- endif -%} {%- if value['nullable'] %} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} nullable:true {%- endif -%} {%- if value['type'] | upper == 'OBJECT' -%} {%- if value['properties'] is defined and value['properties'] is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value['properties'], value['required'] | default([])) -}} } {%- elif value is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value, value['required'] | default([]), filter_keys=true) -}} } {%- endif -%} {%- if value['required'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} required:[ {%- for item in value['required'] | default([]) -%} <|"|>{{- item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- endif -%} {%- endif -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} type:<|"|>{{ value['type'] | upper }}<|"|>} {%- endif -%} {%- endfor -%} {%- endmacro -%} {%- macro format_function_declaration(tool_data) -%} declaration:{{- tool_data['function']['name'] -}}{description:<|"|>{{- tool_data['function']['description'] -}}<|"|> {%- set params = tool_data['function']['parameters'] -%} {%- if params -%} ,parameters:{ {%- if params['properties'] -%} properties:{ {{- format_parameters(params['properties'], params['required']) -}} }, {%- endif -%} {%- if params['required'] -%} required:[ {%- for item in params['required'] -%} <|"|>{{- item -}}<|"|> {{- ',' if not loop.last -}} {%- endfor -%} ], {%- endif -%} {%- if params['type'] -%} type:<|"|>{{- params['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} {%- if 'response' in tool_data['function'] -%} {%- set response_declaration = tool_data['function']['response'] -%} ,response:{ {%- if response_declaration['description'] -%} description:<|"|>{{- response_declaration['description'] -}}<|"|>, {%- endif -%} {%- if response_declaration['type'] | upper == 'OBJECT' -%} type:<|"|>{{- response_declaration['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} } {%- endmacro -%} {%- macro format_argument(argument, escape_keys=True) -%} {%- if argument is none -%} {{- 'null' -}} {%- elif argument is string -%} {{- '<|"|>' + argument + '<|"|>' -}} {%- elif argument is boolean -%} {{- 'true' if argument else 'false' -}} {%- elif argument is mapping -%} {{- '{' -}} {%- set ns = namespace(found_first=false) -%} {%- for key, value in argument | dictsort -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {%- if escape_keys -%} {{- '<|"|>' + key + '<|"|>' -}} {%- else -%} {{- key -}} {%- endif -%} :{{- format_argument(value, escape_keys=escape_keys) -}} {%- endfor -%} {{- '}' -}} {%- elif argument is sequence -%} {{- '[' -}} {%- for item in argument -%} {{- format_argument(item, escape_keys=escape_keys) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- ']' -}} {%- else -%} {{- argument -}} {%- endif -%} {%- endmacro -%} {%- macro strip_thinking(text) -%} {%- set ns = namespace(result='') -%} {%- for part in text.split('') -%} {%- if '<|channel>' in part -%} {%- set ns.result = ns.result + part.split('<|channel>')[0] -%} {%- else -%} {%- set ns.result = ns.result + part -%} {%- endif -%} {%- endfor -%} {{- ns.result | trim -}} {%- endmacro -%} {%- macro format_tool_response_block(tool_name, response) -%} {{- '<|tool_response>' -}} {%- if response is mapping -%} {{- 'response:' + tool_name + '{' -}} {%- for key, value in response | dictsort -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- '}' -}} {%- else -%} {{- 'response:' + tool_name + '{value:' + format_argument(response, escape_keys=False) + '}' -}} {%- endif -%} {{- '' -}} {%- endmacro -%} {#- ===== SETUP ===== -#} {%- set ns = namespace(prev_message_type=None, prev_non_tool_role=None) -%} {%- set loop_messages = messages -%} {%- set enable_thinking = enable_thinking | default(false) -%} {%- set preserve_thinking = preserve_thinking | default(false) -%} {{- bos_token -}} {#- Handle System/Tool Definitions Block -#} {%- if enable_thinking or tools or (messages and messages[0]['role'] in ['system', 'developer']) -%} {{- '<|turn>system\n' -}} {#- Inject Thinking token at the very top of the FIRST system turn -#} {%- if enable_thinking -%} {{- '<|think|>\n' -}} {%- set ns.prev_message_type = 'think' -%} {%- endif -%} {%- if messages and messages[0]['role'] in ['system', 'developer'] -%} {%- if messages[0]['content'] is string -%} {{- messages[0]['content'] | trim -}} {%- elif messages[0]['content'] is sequence -%} {%- for item in messages[0]['content'] -%} {{- item['text'] | trim + ' '-}} {%- endfor -%} {%- endif -%} {%- set loop_messages = messages[1:] -%} {%- endif -%} {%- if tools -%} {%- for tool in tools %} {{- '<|tool>' -}} {{- format_function_declaration(tool) | trim -}} {{- '' -}} {%- endfor %} {%- set ns.prev_message_type = 'tool' -%} {%- endif -%} {{- '\n' -}} {%- endif %} {#- Pre-scan: find last user message index for reasoning guard -#} {%- set ns_turn = namespace(last_user_idx=-1) -%} {%- for i in range(loop_messages | length) -%} {%- if loop_messages[i]['role'] == 'user' -%} {%- set ns_turn.last_user_idx = i -%} {%- endif -%} {%- endfor -%} {#- Loop through messages -#} {%- for message in loop_messages -%} {%- if message['role'] != 'tool' -%} {%- set ns.prev_message_type = None -%} {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%} {#- Detect continuation using tracked state — O(1) instead of O(n) backward scan -#} {%- set continue_same_model_turn = (role == 'model' and ns.prev_non_tool_role == 'assistant') -%} {%- if not continue_same_model_turn -%} {{- '<|turn>' + role + '\n' }} {%- endif -%} {#- Render reasoning/reasoning_content as thinking channel -#} {%- set thinking_text = message.get('reasoning') or message.get('reasoning_content') -%} {%- set thinking_gate = (loop.index0 > ns_turn.last_user_idx) or (preserve_thinking and message.get('tool_calls')) -%} {%- if thinking_text and thinking_gate -%} {{- '<|channel>thought\n' + thinking_text + '\n' -}} {%- endif -%} {%- if message.get('tool_calls') -%} {%- for tool_call in message.get('tool_calls') -%} {%- set function = tool_call['function'] -%} {{- '<|tool_call>call:' + function['name'] + '{' -}} {%- if function['arguments'] is mapping -%} {%- set ns_args = namespace(found_first=false) -%} {%- for key, value in function['arguments'] | dictsort -%} {%- if ns_args.found_first %},{% endif -%} {%- set ns_args.found_first = true -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- endfor -%} {%- elif function['arguments'] is none -%} {%- else -%} {{- raise_exception( "chat_template: tool_calls[].function.arguments must be a " "JSON object (mapping), not a string. Deserialize arguments " "before passing to the template." ) -}} {%- endif -%} {{- '}' -}} {%- endfor -%} {%- set ns.prev_message_type = 'tool_call' -%} {%- endif -%} {%- set ns_tr_out = namespace(flag=false) -%} {%- if message.get('tool_responses') -%} {#- Legacy: tool_responses embedded on the assistant message (Google/Gemma native) -#} {%- for tool_response in message.get('tool_responses') -%} {{- format_tool_response_block(tool_response['name'] | default('unknown', true), tool_response['response']) -}} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endfor -%} {%- elif message.get('tool_calls') -%} {#- OpenAI Chat Completions: forward-scan consecutive role:tool messages -#} {%- set ns_tool_scan = namespace(stopped=false) -%} {%- for k in range(loop.index0 + 1, loop_messages | length) -%} {%- if ns_tool_scan.stopped -%} {%- elif loop_messages[k]['role'] != 'tool' -%} {%- set ns_tool_scan.stopped = true -%} {%- else -%} {%- set follow = loop_messages[k] -%} {#- Resolve tool_call_id to function name -#} {%- set ns_tname = namespace(name=follow.get('name') or 'unknown') -%} {%- for tc in message.get('tool_calls') -%} {%- if tc.get('id') == follow.get('tool_call_id') -%} {%- set ns_tname.name = tc['function']['name'] -%} {%- endif -%} {%- endfor -%} {#- Handle content as string or content-parts array -#} {%- set tool_body = follow.get('content') -%} {%- if tool_body is string -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- elif tool_body is sequence and tool_body is not string -%} {%- set ns_txt = namespace(s='') -%} {%- for part in tool_body -%} {%- if part.get('type') == 'text' -%} {%- set ns_txt.s = ns_txt.s + (part.get('text') | default('')) -%} {%- endif -%} {%- endfor -%} {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}} {%- for part in tool_body -%} {%- if part.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif part.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif part.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- else -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- endif -%} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endif -%} {%- endfor -%} {%- endif -%} {%- set captured_content -%} {%- if message.get('content') is string -%} {%- if role == 'model' -%} {{- strip_thinking(message['content']) -}} {%- else -%} {{- message['content'] | trim -}} {%- endif -%} {%- elif message.get('content') is sequence -%} {%- for item in message['content'] -%} {%- if item.get('type') == 'text' -%} {%- if role == 'model' -%} {{- strip_thinking(item['text']) -}} {%- else -%} {{- item['text'] | trim -}} {%- endif -%} {%- elif item.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif item.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif item.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- endif -%} {%- endset -%} {{- captured_content -}} {%- set has_content = captured_content | trim | length > 0 -%} {#- Forward-scan: find next non-tool message role for continuation detection -#} {%- set next_nt = namespace(role=None, found=false) -%} {%- for j in range(loop.index0 + 1, loop_messages | length) -%} {%- if not next_nt.found -%} {%- if loop_messages[j]['role'] != 'tool' -%} {%- set next_nt.role = loop_messages[j]['role'] -%} {%- set next_nt.found = true -%} {%- endif -%} {%- endif -%} {%- endfor -%} {%- set continues_into_next = ( role == 'model' and next_nt.role == 'assistant' and (not message.get('tool_calls') or ns_tr_out.flag) ) -%} {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%} {{- '<|tool_response>' -}} {%- elif continues_into_next -%} {%- elif not (ns_tr_out.flag and not has_content and not next_nt.found) -%} {{- '\n' -}} {%- endif -%} {#- Track previous non-tool role for next iteration (avoids O(n) backward scan) -#} {%- set ns.prev_non_tool_role = message['role'] -%} {%- endif -%} {%- endfor -%} {%- if add_generation_prompt -%} {%- if ns.prev_message_type != 'tool_response' and ns.prev_message_type != 'tool_call' -%} {{- '<|turn>model\n' -}} {%- if not enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- elif ns.prev_message_type == 'tool_response' and enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- endif -%} INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/gemma-4-31B-it-BF16.gguf: n_tensors = 833, total_size = 61.4G Writing: 0%| | 0.00/61.4G [00:00 BF16, shape = {1152, 5376} INFO:hf-to-gguf:v.blk.0.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.0.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.0.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.0.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.0.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.0.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.0.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.0.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.1.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.1.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.1.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.1.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.1.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.10.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.10.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.10.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.10.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.10.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.11.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.11.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.11.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.11.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.11.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.12.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.12.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.12.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.12.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.12.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.13.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.13.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.13.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.13.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.13.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.14.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.14.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.14.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.14.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.14.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.15.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.15.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.15.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.15.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.15.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.16.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.16.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.16.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.16.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.16.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.17.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.17.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.17.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.17.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.17.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.18.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.18.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.18.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.18.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.18.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.19.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.19.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.19.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.19.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.19.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.2.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.2.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.2.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.2.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.2.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.20.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.20.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.20.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.20.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.20.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.21.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.21.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.21.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.21.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.21.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.22.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.22.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.22.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.22.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.22.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.23.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.23.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.23.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.23.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.23.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.24.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.24.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.24.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.24.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.24.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.25.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.25.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.25.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.25.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.25.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.26.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.26.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.26.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.26.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.26.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.3.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.3.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.3.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.3.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.3.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.4.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.4.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.4.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.4.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.4.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.5.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.5.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.5.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.5.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.5.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.6.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.6.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.6.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.6.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.6.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.7.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.7.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.7.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.7.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.7.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.8.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.8.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.8.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.8.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.8.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.ln1.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.ffn_down.weight, torch.bfloat16 --> BF16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.9.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.9.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.9.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.9.attn_k.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.attn_out.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.9.attn_q.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.attn_v.weight, torch.bfloat16 --> BF16, shape = {1152, 1152} INFO:hf-to-gguf:v.patch_embd.weight, torch.bfloat16 --> F32, shape = {16, 16, 3, 1152} INFO:hf-to-gguf:v.position_embd.weight, torch.bfloat16 --> F32, shape = {1152, 10240, 2} INFO:hf-to-gguf:v.std_bias, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.std_scale, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:Set model quantization version INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/mmproj-gemma-4-31B-it-BF16.gguf: n_tensors = 356, total_size = 1.2G Writing: 0%| | 0.00/1.20G [00:00 F32, shape = {256} INFO:hf-to-gguf:token_embd.weight, torch.bfloat16 --> BF16, shape = {1024, 262144} INFO:hf-to-gguf:blk.0.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.0.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.0.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.0.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.0.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.0.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.1.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.1.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.1.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.1.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.1.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.1.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.2.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.2.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.2.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.2.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.2.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.2.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.3.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.3.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.3.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.3.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.3.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 1024} INFO:hf-to-gguf:blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.3.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 16384} INFO:hf-to-gguf:output_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:nextn.post_projection.weight, torch.bfloat16 --> BF16, shape = {1024, 5376} INFO:hf-to-gguf:nextn.pre_projection.weight, torch.bfloat16 --> BF16, shape = {10752, 1024} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:gguf: context length = 262144 INFO:hf-to-gguf:gguf: embedding length = 1024 INFO:hf-to-gguf:gguf: feed forward length = 8192 INFO:hf-to-gguf:gguf: head count = 32 INFO:hf-to-gguf:gguf: key-value head count = 16 WARNING:hf-to-gguf:Unknown RoPE type: proportional INFO:hf-to-gguf:gguf: rope scaling type = NONE INFO:hf-to-gguf:gguf: rope theta = 1000000.0 INFO:hf-to-gguf:gguf: rope theta swa = 10000.0 INFO:hf-to-gguf:gguf: rms norm epsilon = 1e-06 INFO:hf-to-gguf:gguf: file type = 32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.context_length', overwriting it with new value 262144 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.head_count', overwriting it with new value 32 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.layer_norm_rms_epsilon', overwriting it with new value 1e-06 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.key_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.value_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.rope.freq_base', overwriting it with new value 1000000.0 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.head_count_kv', overwriting it with new value 16 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.key_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.value_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.head_count_kv', overwriting it with new value [16, 16, 16, 4] of type ARRAY INFO:hf-to-gguf:Set model quantization version INFO:hf-to-gguf:Set model tokenizer INFO:hf-to-gguf:Token '<|tool_call>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|tool_response>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|"|>' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|channel>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:gguf.vocab:Adding 514906 merge(s). INFO:gguf.vocab:Setting special token type bos to 2 INFO:gguf.vocab:Setting special token type eos to 1 INFO:gguf.vocab:Setting special token type unk to 3 INFO:gguf.vocab:Setting special token type pad to 0 INFO:gguf.vocab:Setting special token type mask to 4 INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/mtp-gemma-4-31B-it-BF16.gguf: n_tensors = 49, total_size = 939.1M Writing: 0%| | 0.00/939M [00:00", "", "", "", ... llama_model_loader: - kv 38: tokenizer.ggml.scores arr[f32,262144] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 39: tokenizer.ggml.token_type arr[i32,262144] = [3, 3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 40: tokenizer.ggml.merges arr[str,514906] = ["\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n", ... llama_model_loader: - kv 41: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 42: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 43: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 44: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 45: tokenizer.ggml.mask_token_id u32 = 4 llama_model_loader: - kv 46: tokenizer.chat_template str = {#\n Template: Google Gemma 4 Canonic... llama_model_loader: - kv 47: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 48: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - type f32: 422 tensors llama_model_loader: - type bf16: 411 tensors [ 1/ 833] output_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 2/ 833] rope_freqs.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 3/ 833] token_embd.weight - [ 5376, 262144, 1, 1], type = bf16, converting to q8_0 .. size = 2688.00 MiB -> 1428.00 MiB [ 4/ 833] blk.0.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 5/ 833] blk.0.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 6/ 833] blk.0.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 7/ 833] blk.0.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 8/ 833] blk.0.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 9/ 833] blk.0.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 10/ 833] blk.0.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 11/ 833] blk.0.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 12/ 833] blk.0.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 13/ 833] blk.0.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 14/ 833] blk.0.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 15/ 833] blk.0.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 16/ 833] blk.0.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 17/ 833] blk.0.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 18/ 833] blk.1.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 19/ 833] blk.1.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 20/ 833] blk.1.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 21/ 833] blk.1.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 22/ 833] blk.1.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 23/ 833] blk.1.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 24/ 833] blk.1.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 25/ 833] blk.1.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 26/ 833] blk.1.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 27/ 833] blk.1.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 28/ 833] blk.1.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 29/ 833] blk.1.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 30/ 833] blk.1.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 31/ 833] blk.1.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 32/ 833] blk.2.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 33/ 833] blk.2.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 34/ 833] blk.2.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 35/ 833] blk.2.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 36/ 833] blk.2.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 37/ 833] blk.2.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 38/ 833] blk.2.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 39/ 833] blk.2.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 40/ 833] blk.2.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 41/ 833] blk.2.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 42/ 833] blk.2.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 43/ 833] blk.2.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 44/ 833] blk.2.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 45/ 833] blk.2.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 46/ 833] blk.3.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 47/ 833] blk.3.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 48/ 833] blk.3.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 49/ 833] blk.3.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 50/ 833] blk.3.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 51/ 833] blk.3.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 52/ 833] blk.3.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 53/ 833] blk.3.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 54/ 833] blk.3.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 55/ 833] blk.3.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 56/ 833] blk.3.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 57/ 833] blk.3.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 58/ 833] blk.3.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 59/ 833] blk.3.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 60/ 833] blk.4.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 61/ 833] blk.4.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 62/ 833] blk.4.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 63/ 833] blk.4.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 64/ 833] blk.4.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 65/ 833] blk.4.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 66/ 833] blk.4.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 67/ 833] blk.4.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 68/ 833] blk.4.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 69/ 833] blk.4.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 70/ 833] blk.4.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 71/ 833] blk.4.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 72/ 833] blk.4.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 73/ 833] blk.4.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 74/ 833] blk.5.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 75/ 833] blk.5.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 76/ 833] blk.5.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 77/ 833] blk.5.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 78/ 833] blk.5.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 79/ 833] blk.5.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 80/ 833] blk.5.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 81/ 833] blk.5.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 82/ 833] blk.5.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 83/ 833] blk.5.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 84/ 833] blk.5.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 85/ 833] blk.5.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 86/ 833] blk.5.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 87/ 833] blk.6.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 88/ 833] blk.6.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 89/ 833] blk.6.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 90/ 833] blk.6.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 91/ 833] blk.6.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 92/ 833] blk.6.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 93/ 833] blk.6.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 94/ 833] blk.6.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 95/ 833] blk.6.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 96/ 833] blk.6.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 97/ 833] blk.6.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 98/ 833] blk.6.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 99/ 833] blk.6.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 100/ 833] blk.6.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 101/ 833] blk.7.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 102/ 833] blk.7.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 103/ 833] blk.7.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 104/ 833] blk.7.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 105/ 833] blk.7.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 106/ 833] blk.7.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 107/ 833] blk.7.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 108/ 833] blk.7.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 109/ 833] blk.7.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 110/ 833] blk.7.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 111/ 833] blk.7.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 112/ 833] blk.7.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 113/ 833] blk.7.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 114/ 833] blk.7.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 115/ 833] blk.8.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 116/ 833] blk.8.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 117/ 833] blk.8.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 118/ 833] blk.8.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 119/ 833] blk.8.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 120/ 833] blk.8.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 121/ 833] blk.8.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 122/ 833] blk.8.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 123/ 833] blk.8.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 124/ 833] blk.8.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 125/ 833] blk.8.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 126/ 833] blk.8.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 127/ 833] blk.8.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 128/ 833] blk.8.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 129/ 833] blk.9.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 130/ 833] blk.9.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 131/ 833] blk.9.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 132/ 833] blk.9.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 133/ 833] blk.9.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 134/ 833] blk.9.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 135/ 833] blk.9.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 136/ 833] blk.9.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 137/ 833] blk.9.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 138/ 833] blk.9.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 139/ 833] blk.9.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 140/ 833] blk.9.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 141/ 833] blk.9.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 142/ 833] blk.9.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 143/ 833] blk.10.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 144/ 833] blk.10.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 145/ 833] blk.10.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 146/ 833] blk.10.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 147/ 833] blk.10.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 148/ 833] blk.10.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 149/ 833] blk.10.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 150/ 833] blk.10.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 151/ 833] blk.10.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 152/ 833] blk.10.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 153/ 833] blk.10.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 154/ 833] blk.10.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 155/ 833] blk.10.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 156/ 833] blk.10.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 157/ 833] blk.11.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 158/ 833] blk.11.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 159/ 833] blk.11.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 160/ 833] blk.11.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 161/ 833] blk.11.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 162/ 833] blk.11.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 163/ 833] blk.11.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 164/ 833] blk.11.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 165/ 833] blk.11.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 166/ 833] blk.11.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 167/ 833] blk.11.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 168/ 833] blk.11.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 169/ 833] blk.11.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 170/ 833] blk.12.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 171/ 833] blk.12.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 172/ 833] blk.12.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 173/ 833] blk.12.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 174/ 833] blk.12.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 175/ 833] blk.12.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 176/ 833] blk.12.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 177/ 833] blk.12.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 178/ 833] blk.12.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 179/ 833] blk.12.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 180/ 833] blk.12.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 181/ 833] blk.12.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 182/ 833] blk.12.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 183/ 833] blk.12.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 184/ 833] blk.13.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 185/ 833] blk.13.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 186/ 833] blk.13.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 187/ 833] blk.13.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 188/ 833] blk.13.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 189/ 833] blk.13.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 190/ 833] blk.13.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 191/ 833] blk.13.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 192/ 833] blk.13.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 193/ 833] blk.13.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 194/ 833] blk.13.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 195/ 833] blk.13.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 196/ 833] blk.13.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 197/ 833] blk.13.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 198/ 833] blk.14.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 199/ 833] blk.14.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 200/ 833] blk.14.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 201/ 833] blk.14.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 202/ 833] blk.14.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 203/ 833] blk.14.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 204/ 833] blk.14.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 205/ 833] blk.14.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 206/ 833] blk.14.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 207/ 833] blk.14.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 208/ 833] blk.14.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 209/ 833] blk.14.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 210/ 833] blk.14.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 211/ 833] blk.14.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 212/ 833] blk.15.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 213/ 833] blk.15.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 214/ 833] blk.15.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 215/ 833] blk.15.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 216/ 833] blk.15.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 217/ 833] blk.15.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 218/ 833] blk.15.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 219/ 833] blk.15.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 220/ 833] blk.15.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 221/ 833] blk.15.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 222/ 833] blk.15.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 223/ 833] blk.15.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 224/ 833] blk.15.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 225/ 833] blk.15.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 226/ 833] blk.16.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 227/ 833] blk.16.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 228/ 833] blk.16.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 229/ 833] blk.16.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 230/ 833] blk.16.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 231/ 833] blk.16.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 232/ 833] blk.16.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 233/ 833] blk.16.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 234/ 833] blk.16.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 235/ 833] blk.16.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 236/ 833] blk.16.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 237/ 833] blk.16.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 238/ 833] blk.16.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 239/ 833] blk.16.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 240/ 833] blk.17.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 241/ 833] blk.17.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 242/ 833] blk.17.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 243/ 833] blk.17.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 244/ 833] blk.17.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 245/ 833] blk.17.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 246/ 833] blk.17.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 247/ 833] blk.17.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 248/ 833] blk.17.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 249/ 833] blk.17.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 250/ 833] blk.17.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 251/ 833] blk.17.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 252/ 833] blk.17.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 253/ 833] blk.18.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 254/ 833] blk.18.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 255/ 833] blk.18.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 256/ 833] blk.18.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 257/ 833] blk.18.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 258/ 833] blk.18.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 259/ 833] blk.18.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 260/ 833] blk.18.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 261/ 833] blk.18.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 262/ 833] blk.18.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 263/ 833] blk.18.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 264/ 833] blk.18.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 265/ 833] blk.18.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 266/ 833] blk.18.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 267/ 833] blk.19.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 268/ 833] blk.19.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 269/ 833] blk.19.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 270/ 833] blk.19.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 271/ 833] blk.19.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 272/ 833] blk.19.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 273/ 833] blk.19.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 274/ 833] blk.19.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 275/ 833] blk.19.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 276/ 833] blk.19.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 277/ 833] blk.19.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 278/ 833] blk.19.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 279/ 833] blk.19.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 280/ 833] blk.19.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 281/ 833] blk.20.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 282/ 833] blk.20.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 283/ 833] blk.20.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 284/ 833] blk.20.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 285/ 833] blk.20.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 286/ 833] blk.20.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 287/ 833] blk.20.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 288/ 833] blk.20.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 289/ 833] blk.20.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 290/ 833] blk.20.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 291/ 833] blk.20.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 292/ 833] blk.20.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 293/ 833] blk.20.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 294/ 833] blk.20.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 295/ 833] blk.21.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 296/ 833] blk.21.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 297/ 833] blk.21.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 298/ 833] blk.21.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 299/ 833] blk.21.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 300/ 833] blk.21.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 301/ 833] blk.21.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 302/ 833] blk.21.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 303/ 833] blk.21.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 304/ 833] blk.21.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 305/ 833] blk.21.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 306/ 833] blk.21.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 307/ 833] blk.21.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 308/ 833] blk.21.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 309/ 833] blk.22.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 310/ 833] blk.22.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 311/ 833] blk.22.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 312/ 833] blk.22.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 313/ 833] blk.22.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 314/ 833] blk.22.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 315/ 833] blk.22.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 316/ 833] blk.22.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 317/ 833] blk.22.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 318/ 833] blk.22.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 319/ 833] blk.22.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 320/ 833] blk.22.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 321/ 833] blk.22.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 322/ 833] blk.22.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 323/ 833] blk.23.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 324/ 833] blk.23.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 325/ 833] blk.23.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 326/ 833] blk.23.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 327/ 833] blk.23.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 328/ 833] blk.23.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 329/ 833] blk.23.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 330/ 833] blk.23.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 331/ 833] blk.23.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 332/ 833] blk.23.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 333/ 833] blk.23.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 334/ 833] blk.23.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 335/ 833] blk.23.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 336/ 833] blk.24.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 337/ 833] blk.24.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 338/ 833] blk.24.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 339/ 833] blk.24.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 340/ 833] blk.24.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 341/ 833] blk.24.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 342/ 833] blk.24.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 343/ 833] blk.24.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 344/ 833] blk.24.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 345/ 833] blk.24.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 346/ 833] blk.24.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 347/ 833] blk.24.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 348/ 833] blk.24.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 349/ 833] blk.24.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 350/ 833] blk.25.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 351/ 833] blk.25.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 352/ 833] blk.25.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 353/ 833] blk.25.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 354/ 833] blk.25.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 355/ 833] blk.25.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 356/ 833] blk.25.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 357/ 833] blk.25.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 358/ 833] blk.25.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 359/ 833] blk.25.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 360/ 833] blk.25.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 361/ 833] blk.25.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 362/ 833] blk.25.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 363/ 833] blk.25.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 364/ 833] blk.26.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 365/ 833] blk.26.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 366/ 833] blk.26.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 367/ 833] blk.26.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 368/ 833] blk.26.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 369/ 833] blk.26.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 370/ 833] blk.26.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 371/ 833] blk.26.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 372/ 833] blk.26.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 373/ 833] blk.26.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 374/ 833] blk.26.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 375/ 833] blk.26.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 376/ 833] blk.26.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 377/ 833] blk.26.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 378/ 833] blk.27.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 379/ 833] blk.27.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 380/ 833] blk.27.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 381/ 833] blk.27.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 382/ 833] blk.27.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 383/ 833] blk.27.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 384/ 833] blk.27.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 385/ 833] blk.27.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 386/ 833] blk.27.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 387/ 833] blk.27.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 388/ 833] blk.27.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 389/ 833] blk.27.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 390/ 833] blk.27.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 391/ 833] blk.27.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 392/ 833] blk.28.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 393/ 833] blk.28.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 394/ 833] blk.28.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 395/ 833] blk.28.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 396/ 833] blk.28.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 397/ 833] blk.28.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 398/ 833] blk.28.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 399/ 833] blk.28.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 400/ 833] blk.28.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 401/ 833] blk.28.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 402/ 833] blk.28.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 403/ 833] blk.28.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 404/ 833] blk.28.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 405/ 833] blk.28.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 406/ 833] blk.29.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 407/ 833] blk.29.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 408/ 833] blk.29.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 409/ 833] blk.29.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 410/ 833] blk.29.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 411/ 833] blk.29.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 412/ 833] blk.29.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 413/ 833] blk.29.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 414/ 833] blk.29.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 415/ 833] blk.29.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 416/ 833] blk.29.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 417/ 833] blk.29.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 418/ 833] blk.29.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 419/ 833] blk.30.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 420/ 833] blk.30.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 421/ 833] blk.30.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 422/ 833] blk.30.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 423/ 833] blk.30.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 424/ 833] blk.30.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 425/ 833] blk.30.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 426/ 833] blk.30.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 427/ 833] blk.30.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 428/ 833] blk.30.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 429/ 833] blk.30.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 430/ 833] blk.30.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 431/ 833] blk.30.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 432/ 833] blk.30.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 433/ 833] blk.31.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 434/ 833] blk.31.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 435/ 833] blk.31.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 436/ 833] blk.31.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 437/ 833] blk.31.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 438/ 833] blk.31.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 439/ 833] blk.31.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 440/ 833] blk.31.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 441/ 833] blk.31.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 442/ 833] blk.31.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 443/ 833] blk.31.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 444/ 833] blk.31.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 445/ 833] blk.31.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 446/ 833] blk.31.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 447/ 833] blk.32.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 448/ 833] blk.32.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 449/ 833] blk.32.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 450/ 833] blk.32.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 451/ 833] blk.32.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 452/ 833] blk.32.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 453/ 833] blk.32.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 454/ 833] blk.32.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 455/ 833] blk.32.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 456/ 833] blk.32.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 457/ 833] blk.32.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 458/ 833] blk.32.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 459/ 833] blk.32.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 460/ 833] blk.32.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 461/ 833] blk.33.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 462/ 833] blk.33.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 463/ 833] blk.33.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 464/ 833] blk.33.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 465/ 833] blk.33.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 466/ 833] blk.33.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 467/ 833] blk.33.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 468/ 833] blk.33.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 469/ 833] blk.33.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 470/ 833] blk.33.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 471/ 833] blk.33.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 472/ 833] blk.33.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 473/ 833] blk.33.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 474/ 833] blk.33.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 475/ 833] blk.34.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 476/ 833] blk.34.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 477/ 833] blk.34.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 478/ 833] blk.34.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 479/ 833] blk.34.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 480/ 833] blk.34.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 481/ 833] blk.34.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 482/ 833] blk.34.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 483/ 833] blk.34.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 484/ 833] blk.34.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 485/ 833] blk.34.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 486/ 833] blk.34.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 487/ 833] blk.34.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 488/ 833] blk.34.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 489/ 833] blk.35.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 490/ 833] blk.35.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 491/ 833] blk.35.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 492/ 833] blk.35.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 493/ 833] blk.35.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 494/ 833] blk.35.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 495/ 833] blk.35.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 496/ 833] blk.35.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 497/ 833] blk.35.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 498/ 833] blk.35.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 499/ 833] blk.35.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 500/ 833] blk.35.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 501/ 833] blk.35.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 502/ 833] blk.36.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 503/ 833] blk.36.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 504/ 833] blk.36.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 505/ 833] blk.36.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 506/ 833] blk.36.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 507/ 833] blk.36.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 508/ 833] blk.36.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 509/ 833] blk.36.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 510/ 833] blk.36.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 511/ 833] blk.36.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 512/ 833] blk.36.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 513/ 833] blk.36.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 514/ 833] blk.36.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 515/ 833] blk.36.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 516/ 833] blk.37.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 517/ 833] blk.37.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 518/ 833] blk.37.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 519/ 833] blk.37.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 520/ 833] blk.37.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 521/ 833] blk.37.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 522/ 833] blk.37.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 523/ 833] blk.37.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 524/ 833] blk.37.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 525/ 833] blk.37.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 526/ 833] blk.37.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 527/ 833] blk.37.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 528/ 833] blk.37.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 529/ 833] blk.37.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 530/ 833] blk.38.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 531/ 833] blk.38.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 532/ 833] blk.38.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 533/ 833] blk.38.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 534/ 833] blk.38.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 535/ 833] blk.38.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 536/ 833] blk.38.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 537/ 833] blk.38.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 538/ 833] blk.38.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 539/ 833] blk.38.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 540/ 833] blk.38.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 541/ 833] blk.38.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 542/ 833] blk.38.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 543/ 833] blk.38.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 544/ 833] blk.39.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 545/ 833] blk.39.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 546/ 833] blk.39.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 547/ 833] blk.39.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 548/ 833] blk.39.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 549/ 833] blk.39.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 550/ 833] blk.39.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 551/ 833] blk.39.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 552/ 833] blk.39.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 553/ 833] blk.39.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 554/ 833] blk.39.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 555/ 833] blk.39.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 556/ 833] blk.39.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 557/ 833] blk.39.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 558/ 833] blk.40.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 559/ 833] blk.40.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 560/ 833] blk.40.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 561/ 833] blk.40.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 562/ 833] blk.40.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 563/ 833] blk.40.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 564/ 833] blk.40.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 565/ 833] blk.40.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 566/ 833] blk.40.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 567/ 833] blk.40.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 568/ 833] blk.40.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 569/ 833] blk.40.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 570/ 833] blk.40.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 571/ 833] blk.40.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 572/ 833] blk.41.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 573/ 833] blk.41.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 574/ 833] blk.41.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 575/ 833] blk.41.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 576/ 833] blk.41.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 577/ 833] blk.41.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 578/ 833] blk.41.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 579/ 833] blk.41.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 580/ 833] blk.41.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 581/ 833] blk.41.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 582/ 833] blk.41.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 583/ 833] blk.41.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 584/ 833] blk.41.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 585/ 833] blk.42.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 586/ 833] blk.42.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 587/ 833] blk.42.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 588/ 833] blk.42.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 589/ 833] blk.42.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 590/ 833] blk.42.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 591/ 833] blk.42.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 592/ 833] blk.42.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 593/ 833] blk.42.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 594/ 833] blk.42.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 595/ 833] blk.42.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 596/ 833] blk.42.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 597/ 833] blk.42.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 598/ 833] blk.42.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 599/ 833] blk.43.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 600/ 833] blk.43.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 601/ 833] blk.43.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 602/ 833] blk.43.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 603/ 833] blk.43.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 604/ 833] blk.43.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 605/ 833] blk.43.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 606/ 833] blk.43.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 607/ 833] blk.43.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 608/ 833] blk.43.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 609/ 833] blk.43.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 610/ 833] blk.43.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 611/ 833] blk.43.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 612/ 833] blk.43.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 613/ 833] blk.44.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 614/ 833] blk.44.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 615/ 833] blk.44.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 616/ 833] blk.44.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 617/ 833] blk.44.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 618/ 833] blk.44.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 619/ 833] blk.44.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 620/ 833] blk.44.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 621/ 833] blk.44.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 622/ 833] blk.44.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 623/ 833] blk.44.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 624/ 833] blk.44.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 625/ 833] blk.44.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 626/ 833] blk.44.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 627/ 833] blk.45.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 628/ 833] blk.45.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 629/ 833] blk.45.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 630/ 833] blk.45.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 631/ 833] blk.45.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 632/ 833] blk.45.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 633/ 833] blk.45.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 634/ 833] blk.45.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 635/ 833] blk.45.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 636/ 833] blk.45.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 637/ 833] blk.45.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 638/ 833] blk.45.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 639/ 833] blk.45.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 640/ 833] blk.45.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 641/ 833] blk.46.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 642/ 833] blk.46.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 643/ 833] blk.46.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 644/ 833] blk.46.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 645/ 833] blk.46.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 646/ 833] blk.46.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 647/ 833] blk.46.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 648/ 833] blk.46.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 649/ 833] blk.46.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 650/ 833] blk.46.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 651/ 833] blk.46.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 652/ 833] blk.46.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 653/ 833] blk.46.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 654/ 833] blk.46.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 655/ 833] blk.47.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 656/ 833] blk.47.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 657/ 833] blk.47.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 658/ 833] blk.47.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 659/ 833] blk.47.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 660/ 833] blk.47.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 661/ 833] blk.47.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 662/ 833] blk.47.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 663/ 833] blk.47.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 664/ 833] blk.47.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 665/ 833] blk.47.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 666/ 833] blk.47.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 667/ 833] blk.47.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 668/ 833] blk.48.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 669/ 833] blk.48.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 670/ 833] blk.48.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 671/ 833] blk.48.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 672/ 833] blk.48.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 673/ 833] blk.48.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 674/ 833] blk.48.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 675/ 833] blk.48.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 676/ 833] blk.48.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 677/ 833] blk.48.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 678/ 833] blk.48.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 679/ 833] blk.48.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 680/ 833] blk.48.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 681/ 833] blk.48.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 682/ 833] blk.49.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 683/ 833] blk.49.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 684/ 833] blk.49.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 685/ 833] blk.49.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 686/ 833] blk.49.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 687/ 833] blk.49.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 688/ 833] blk.49.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 689/ 833] blk.49.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 690/ 833] blk.49.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 691/ 833] blk.49.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 692/ 833] blk.49.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 693/ 833] blk.49.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 694/ 833] blk.49.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 695/ 833] blk.49.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 696/ 833] blk.50.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 697/ 833] blk.50.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 698/ 833] blk.50.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 699/ 833] blk.50.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 700/ 833] blk.50.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 701/ 833] blk.50.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 702/ 833] blk.50.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 703/ 833] blk.50.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 704/ 833] blk.50.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 705/ 833] blk.50.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 706/ 833] blk.50.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 707/ 833] blk.50.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 708/ 833] blk.50.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 709/ 833] blk.50.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 710/ 833] blk.51.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 711/ 833] blk.51.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 712/ 833] blk.51.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 713/ 833] blk.51.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 714/ 833] blk.51.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 715/ 833] blk.51.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 716/ 833] blk.51.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 717/ 833] blk.51.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 718/ 833] blk.51.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 719/ 833] blk.51.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 720/ 833] blk.51.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 721/ 833] blk.51.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 722/ 833] blk.51.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 723/ 833] blk.51.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 724/ 833] blk.52.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 725/ 833] blk.52.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 726/ 833] blk.52.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 727/ 833] blk.52.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 728/ 833] blk.52.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 729/ 833] blk.52.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 730/ 833] blk.52.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 731/ 833] blk.52.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 732/ 833] blk.52.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 733/ 833] blk.52.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 734/ 833] blk.52.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 735/ 833] blk.52.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 736/ 833] blk.52.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 737/ 833] blk.52.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 738/ 833] blk.53.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 739/ 833] blk.53.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 740/ 833] blk.53.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 741/ 833] blk.53.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 742/ 833] blk.53.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 743/ 833] blk.53.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 744/ 833] blk.53.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 745/ 833] blk.53.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 746/ 833] blk.53.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 747/ 833] blk.53.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 748/ 833] blk.53.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 749/ 833] blk.53.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 750/ 833] blk.53.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 751/ 833] blk.54.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 752/ 833] blk.54.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 753/ 833] blk.54.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 754/ 833] blk.54.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 755/ 833] blk.54.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 756/ 833] blk.54.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 757/ 833] blk.54.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 758/ 833] blk.54.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 759/ 833] blk.54.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 760/ 833] blk.54.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 761/ 833] blk.54.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 762/ 833] blk.54.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 763/ 833] blk.54.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 764/ 833] blk.54.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 765/ 833] blk.55.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 766/ 833] blk.55.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 767/ 833] blk.55.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 768/ 833] blk.55.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 769/ 833] blk.55.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 770/ 833] blk.55.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 771/ 833] blk.55.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 772/ 833] blk.55.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 773/ 833] blk.55.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 774/ 833] blk.55.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 775/ 833] blk.55.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 776/ 833] blk.55.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 777/ 833] blk.55.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 778/ 833] blk.55.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 779/ 833] blk.56.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 780/ 833] blk.56.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 781/ 833] blk.56.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 782/ 833] blk.56.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 783/ 833] blk.56.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 784/ 833] blk.56.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 785/ 833] blk.56.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 786/ 833] blk.56.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 787/ 833] blk.56.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 788/ 833] blk.56.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 789/ 833] blk.56.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 790/ 833] blk.56.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 791/ 833] blk.56.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 792/ 833] blk.56.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 793/ 833] blk.57.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 794/ 833] blk.57.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 795/ 833] blk.57.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 796/ 833] blk.57.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 797/ 833] blk.57.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 798/ 833] blk.57.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 799/ 833] blk.57.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 800/ 833] blk.57.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 801/ 833] blk.57.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 802/ 833] blk.57.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 803/ 833] blk.57.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 804/ 833] blk.57.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 805/ 833] blk.57.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 806/ 833] blk.57.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 807/ 833] blk.58.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 808/ 833] blk.58.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 809/ 833] blk.58.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 810/ 833] blk.58.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 811/ 833] blk.58.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 812/ 833] blk.58.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 813/ 833] blk.58.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q8_0 .. size = 42.00 MiB -> 22.31 MiB [ 814/ 833] blk.58.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 815/ 833] blk.58.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 816/ 833] blk.58.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 817/ 833] blk.58.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 818/ 833] blk.58.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 819/ 833] blk.58.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 820/ 833] blk.58.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 821/ 833] blk.59.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 822/ 833] blk.59.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 823/ 833] blk.59.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 824/ 833] blk.59.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 825/ 833] blk.59.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 168.00 MiB -> 89.25 MiB [ 826/ 833] blk.59.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 827/ 833] blk.59.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 828/ 833] blk.59.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 829/ 833] blk.59.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 830/ 833] blk.59.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q8_0 .. size = 220.50 MiB -> 117.14 MiB [ 831/ 833] blk.59.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 832/ 833] blk.59.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 833/ 833] blk.59.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB llama_model_quantize_impl: model size = 58553.08 MiB (16.00 BPW) llama_model_quantize_impl: quant size = 31108.71 MiB (8.50 BPW) llama_quantize: quantize time = 199173.81 ms llama_quantize: total time = 199173.81 ms + python3 ./llama.cpp/convert_hf_to_gguf.py ./model-temp-gemma_4_31B_it-PRIMARY --outtype q8_0 --outfile ./upload-gemma_4_31B_it/mmproj-gemma-4-31B-it-Q8_0.gguf --mmproj --model-name gemma-4-31B-it INFO:hf-to-gguf:Loading model: model-temp-gemma_4_31B_it-PRIMARY INFO:hf-to-gguf:Model architecture: Gemma4ForConditionalGeneration INFO:hf-to-gguf:gguf: loading model weight map from 'model.safetensors.index.json' INFO:hf-to-gguf:gguf: indexing model part 'model-00001-of-00002.safetensors' INFO:hf-to-gguf:gguf: indexing model part 'model-00002-of-00002.safetensors' INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only INFO:hf-to-gguf:Exporting model... INFO:hf-to-gguf:mm.input_projection.weight, torch.bfloat16 --> Q8_0, shape = {1152, 5376} INFO:hf-to-gguf:v.blk.0.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.0.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.0.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.0.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.0.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.0.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.0.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.0.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.0.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.0.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.1.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.1.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.1.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.1.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.1.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.1.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.1.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.1.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.10.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.10.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.10.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.10.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.10.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.10.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.10.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.10.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.11.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.11.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.11.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.11.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.11.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.11.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.11.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.11.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.12.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.12.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.12.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.12.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.12.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.12.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.12.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.12.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.13.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.13.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.13.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.13.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.13.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.13.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.13.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.13.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.14.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.14.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.14.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.14.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.14.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.14.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.14.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.14.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.15.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.15.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.15.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.15.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.15.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.15.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.15.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.15.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.16.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.16.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.16.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.16.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.16.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.16.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.16.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.16.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.17.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.17.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.17.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.17.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.17.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.17.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.17.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.17.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.18.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.18.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.18.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.18.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.18.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.18.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.18.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.18.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.19.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.19.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.19.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.19.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.19.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.19.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.19.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.19.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.2.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.2.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.2.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.2.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.2.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.2.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.2.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.2.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.20.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.20.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.20.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.20.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.20.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.20.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.20.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.20.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.21.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.21.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.21.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.21.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.21.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.21.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.21.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.21.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.22.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.22.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.22.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.22.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.22.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.22.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.22.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.22.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.23.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.23.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.23.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.23.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.23.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.23.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.23.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.23.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.24.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.24.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.24.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.24.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.24.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.24.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.24.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.24.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.25.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.25.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.25.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.25.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.25.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.25.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.25.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.25.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.26.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.26.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.26.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.26.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.26.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.26.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.26.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.26.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.3.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.3.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.3.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.3.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.3.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.3.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.3.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.3.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.4.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.4.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.4.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.4.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.4.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.4.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.4.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.4.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.5.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.5.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.5.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.5.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.5.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.5.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.5.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.5.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.6.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.6.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.6.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.6.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.6.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.6.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.6.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.6.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.7.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.7.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.7.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.7.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.7.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.7.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.7.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.7.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.8.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.8.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.8.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.8.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.8.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.8.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.8.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.8.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.ln1.weight, torch.bfloat16 --> F32, shape = {1152} WARNING:hf-to-gguf:Can't quantize tensor with shape (1152, 4304) to Q8_0, falling back to F16 INFO:hf-to-gguf:v.blk.9.ffn_down.weight, torch.bfloat16 --> F16, shape = {4304, 1152} INFO:hf-to-gguf:v.blk.9.ffn_gate.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.9.ffn_up.weight, torch.bfloat16 --> Q8_0, shape = {1152, 4304} INFO:hf-to-gguf:v.blk.9.attn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.ffn_post_norm.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.ln2.weight, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.blk.9.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.9.attn_k.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.attn_out.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {72} INFO:hf-to-gguf:v.blk.9.attn_q.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.blk.9.attn_v.weight, torch.bfloat16 --> Q8_0, shape = {1152, 1152} INFO:hf-to-gguf:v.patch_embd.weight, torch.bfloat16 --> F32, shape = {16, 16, 3, 1152} INFO:hf-to-gguf:v.position_embd.weight, torch.bfloat16 --> F32, shape = {1152, 10240, 2} INFO:hf-to-gguf:v.std_bias, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:v.std_scale, torch.bfloat16 --> F32, shape = {1152} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:Set model quantization version INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/mmproj-gemma-4-31B-it-Q8_0.gguf: n_tensors = 356, total_size = 809.5M Writing: 0%| | 0.00/810M [00:00", "", "", "", ... llama_model_loader: - kv 35: tokenizer.ggml.scores arr[f32,262144] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 36: tokenizer.ggml.token_type arr[i32,262144] = [3, 3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 37: tokenizer.ggml.merges arr[str,514906] = ["\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n", ... llama_model_loader: - kv 38: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 39: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 40: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 41: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 42: tokenizer.ggml.mask_token_id u32 = 4 llama_model_loader: - kv 43: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 44: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - type f32: 26 tensors llama_model_loader: - type bf16: 23 tensors [ 1/ 49] nextn.post_projection.weight - [ 1024, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 2/ 49] nextn.pre_projection.weight - [ 10752, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 21.00 MiB -> 11.16 MiB [ 3/ 49] output_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 4/ 49] rope_freqs.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 5/ 49] token_embd.weight - [ 1024, 262144, 1, 1], type = bf16, converting to q8_0 .. size = 512.00 MiB -> 272.00 MiB [ 6/ 49] blk.0.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 7/ 49] blk.0.attn_output.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 8/ 49] blk.0.attn_q.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 9/ 49] blk.0.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 10/ 49] blk.0.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 11/ 49] blk.0.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 12/ 49] blk.0.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 13/ 49] blk.0.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 14/ 49] blk.0.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 15/ 49] blk.0.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 16/ 49] blk.0.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 17/ 49] blk.1.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 18/ 49] blk.1.attn_output.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 19/ 49] blk.1.attn_q.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 20/ 49] blk.1.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 21/ 49] blk.1.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 22/ 49] blk.1.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 23/ 49] blk.1.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 24/ 49] blk.1.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 25/ 49] blk.1.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 26/ 49] blk.1.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 27/ 49] blk.1.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 28/ 49] blk.2.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 29/ 49] blk.2.attn_output.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 30/ 49] blk.2.attn_q.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 31/ 49] blk.2.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 32/ 49] blk.2.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 33/ 49] blk.2.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 34/ 49] blk.2.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 35/ 49] blk.2.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 36/ 49] blk.2.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 37/ 49] blk.2.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 38/ 49] blk.2.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 39/ 49] blk.3.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 40/ 49] blk.3.attn_output.weight - [ 16384, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 32.00 MiB -> 17.00 MiB [ 41/ 49] blk.3.attn_q.weight - [ 1024, 16384, 1, 1], type = bf16, converting to q8_0 .. size = 32.00 MiB -> 17.00 MiB [ 42/ 49] blk.3.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 43/ 49] blk.3.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 44/ 49] blk.3.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 45/ 49] blk.3.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 46/ 49] blk.3.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 16.00 MiB -> 8.50 MiB [ 47/ 49] blk.3.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 48/ 49] blk.3.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 49/ 49] blk.3.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB llama_model_quantize_impl: model size = 895.57 MiB (16.00 BPW) llama_model_quantize_impl: quant size = 475.81 MiB (8.50 BPW) llama_quantize: quantize time = 5684.26 ms llama_quantize: total time = 5684.26 ms + python3 ./llama.cpp/convert_hf_to_gguf.py ./model-temp-gemma_4_31B_it-QAT_Q4_0 --outtype bf16 --outfile ./upload-gemma_4_31B_it/mtp-gemma-4-31B-it-QAT-BF16.gguf --model-name gemma-4-31B-it INFO:hf-to-gguf:Loading model: model-temp-gemma_4_31B_it-QAT_Q4_0 INFO:hf-to-gguf:Model architecture: Gemma4AssistantForCausalLM INFO:hf-to-gguf:gguf: indexing model part 'model.safetensors' INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only INFO:hf-to-gguf:Exporting model... INFO:hf-to-gguf:rope_freqs.weight, torch.float32 --> F32, shape = {256} INFO:hf-to-gguf:token_embd.weight, torch.bfloat16 --> BF16, shape = {1024, 262144} INFO:hf-to-gguf:blk.0.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.0.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.0.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.0.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.0.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.0.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.0.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.1.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.1.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.1.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.1.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.1.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.1.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.1.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.2.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.2.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.2.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.2.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.2.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.2.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.2.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.3.attn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.3.ffn_down.weight, torch.bfloat16 --> BF16, shape = {8192, 1024} INFO:hf-to-gguf:blk.3.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.3.ffn_up.weight, torch.bfloat16 --> BF16, shape = {1024, 8192} INFO:hf-to-gguf:blk.3.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.ffn_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:blk.3.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 1024} INFO:hf-to-gguf:blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.3.attn_q.weight, torch.bfloat16 --> BF16, shape = {1024, 16384} INFO:hf-to-gguf:output_norm.weight, torch.bfloat16 --> F32, shape = {1024} INFO:hf-to-gguf:nextn.post_projection.weight, torch.bfloat16 --> BF16, shape = {1024, 5376} INFO:hf-to-gguf:nextn.pre_projection.weight, torch.bfloat16 --> BF16, shape = {10752, 1024} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:gguf: context length = 262144 INFO:hf-to-gguf:gguf: embedding length = 1024 INFO:hf-to-gguf:gguf: feed forward length = 8192 INFO:hf-to-gguf:gguf: head count = 32 INFO:hf-to-gguf:gguf: key-value head count = 16 WARNING:hf-to-gguf:Unknown RoPE type: proportional INFO:hf-to-gguf:gguf: rope scaling type = NONE INFO:hf-to-gguf:gguf: rope theta = 1000000.0 INFO:hf-to-gguf:gguf: rope theta swa = 10000.0 INFO:hf-to-gguf:gguf: rms norm epsilon = 1e-06 INFO:hf-to-gguf:gguf: file type = 32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.context_length', overwriting it with new value 262144 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.head_count', overwriting it with new value 32 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.layer_norm_rms_epsilon', overwriting it with new value 1e-06 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.key_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.value_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.rope.freq_base', overwriting it with new value 1000000.0 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.head_count_kv', overwriting it with new value 16 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.key_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.value_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4-assistant.attention.head_count_kv', overwriting it with new value [16, 16, 16, 4] of type ARRAY INFO:hf-to-gguf:Set model quantization version INFO:hf-to-gguf:Set model tokenizer INFO:hf-to-gguf:Token '<|tool_call>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|tool_response>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|"|>' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|channel>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:gguf.vocab:Adding 514906 merge(s). INFO:gguf.vocab:Setting special token type bos to 2 INFO:gguf.vocab:Setting special token type eos to 1 INFO:gguf.vocab:Setting special token type unk to 3 INFO:gguf.vocab:Setting special token type pad to 0 INFO:gguf.vocab:Setting special token type mask to 4 INFO:gguf.vocab:Setting chat_template to {# Template: Google Gemma 4 Canonical Chat Template Author: Google Gemma Engineering Team Published: 2026-07-09 Context: Fixed tool-calling loops, turn closures, and thinking content-ordering. #} {%- macro format_parameters(properties, required, filter_keys=false) -%} {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%} {%- set ns = namespace(found_first=false) -%} {%- for key, value in properties | dictsort -%} {%- set add_comma = false -%} {%- if not filter_keys or key not in standard_keys -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {{ key }}:{ {%- if value['description'] -%} description:<|"|>{{ value['description'] }}<|"|> {%- set add_comma = true -%} {%- endif -%} {%- if value['type'] | upper == 'STRING' -%} {%- if value['enum'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} enum:{{ format_argument(value['enum']) }} {%- endif -%} {%- elif value['type'] | upper == 'ARRAY' -%} {%- if value['items'] is mapping and value['items'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} items:{ {%- set ns_items = namespace(found_first=false) -%} {%- for item_key, item_value in value['items'] | dictsort -%} {%- if item_value is not none -%} {%- if ns_items.found_first %},{% endif -%} {%- set ns_items.found_first = true -%} {%- if item_key == 'properties' -%} properties:{ {%- if item_value is mapping -%} {{- format_parameters(item_value, value['items']['required'] | default([])) -}} {%- endif -%} } {%- elif item_key == 'required' -%} required:[ {%- for req_item in item_value -%} <|"|>{{- req_item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- elif item_key == 'type' -%} {%- if item_value is string -%} type:{{ format_argument(item_value | upper) }} {%- else -%} type:{{ format_argument(item_value | map('upper') | list) }} {%- endif -%} {%- else -%} {{ item_key }}:{{ format_argument(item_value) }} {%- endif -%} {%- endif -%} {%- endfor -%} } {%- endif -%} {%- endif -%} {%- if value['nullable'] %} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} nullable:true {%- endif -%} {%- if value['type'] | upper == 'OBJECT' -%} {%- if value['properties'] is defined and value['properties'] is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value['properties'], value['required'] | default([])) -}} } {%- elif value is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value, value['required'] | default([]), filter_keys=true) -}} } {%- endif -%} {%- if value['required'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} required:[ {%- for item in value['required'] | default([]) -%} <|"|>{{- item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- endif -%} {%- endif -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} type:<|"|>{{ value['type'] | upper }}<|"|>} {%- endif -%} {%- endfor -%} {%- endmacro -%} {%- macro format_function_declaration(tool_data) -%} declaration:{{- tool_data['function']['name'] -}}{description:<|"|>{{- tool_data['function']['description'] -}}<|"|> {%- set params = tool_data['function']['parameters'] -%} {%- if params -%} ,parameters:{ {%- if params['properties'] -%} properties:{ {{- format_parameters(params['properties'], params['required']) -}} }, {%- endif -%} {%- if params['required'] -%} required:[ {%- for item in params['required'] -%} <|"|>{{- item -}}<|"|> {{- ',' if not loop.last -}} {%- endfor -%} ], {%- endif -%} {%- if params['type'] -%} type:<|"|>{{- params['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} {%- if 'response' in tool_data['function'] -%} {%- set response_declaration = tool_data['function']['response'] -%} ,response:{ {%- if response_declaration['description'] -%} description:<|"|>{{- response_declaration['description'] -}}<|"|>, {%- endif -%} {%- if response_declaration['type'] | upper == 'OBJECT' -%} type:<|"|>{{- response_declaration['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} } {%- endmacro -%} {%- macro format_argument(argument, escape_keys=True) -%} {%- if argument is none -%} {{- 'null' -}} {%- elif argument is string -%} {{- '<|"|>' + argument + '<|"|>' -}} {%- elif argument is boolean -%} {{- 'true' if argument else 'false' -}} {%- elif argument is mapping -%} {{- '{' -}} {%- set ns = namespace(found_first=false) -%} {%- for key, value in argument | dictsort -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {%- if escape_keys -%} {{- '<|"|>' + key + '<|"|>' -}} {%- else -%} {{- key -}} {%- endif -%} :{{- format_argument(value, escape_keys=escape_keys) -}} {%- endfor -%} {{- '}' -}} {%- elif argument is sequence -%} {{- '[' -}} {%- for item in argument -%} {{- format_argument(item, escape_keys=escape_keys) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- ']' -}} {%- else -%} {{- argument -}} {%- endif -%} {%- endmacro -%} {%- macro strip_thinking(text) -%} {%- set ns = namespace(result='') -%} {%- for part in text.split('') -%} {%- if '<|channel>' in part -%} {%- set ns.result = ns.result + part.split('<|channel>')[0] -%} {%- else -%} {%- set ns.result = ns.result + part -%} {%- endif -%} {%- endfor -%} {{- ns.result | trim -}} {%- endmacro -%} {%- macro format_tool_response_block(tool_name, response) -%} {{- '<|tool_response>' -}} {%- if response is mapping -%} {{- 'response:' + tool_name + '{' -}} {%- for key, value in response | dictsort -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- '}' -}} {%- else -%} {{- 'response:' + tool_name + '{value:' + format_argument(response, escape_keys=False) + '}' -}} {%- endif -%} {{- '' -}} {%- endmacro -%} {#- ===== SETUP ===== -#} {%- set ns = namespace(prev_message_type=None, prev_non_tool_role=None) -%} {%- set loop_messages = messages -%} {%- set enable_thinking = enable_thinking | default(false) -%} {%- set preserve_thinking = preserve_thinking | default(false) -%} {{- bos_token -}} {#- Handle System/Tool Definitions Block -#} {%- if enable_thinking or tools or (messages and messages[0]['role'] in ['system', 'developer']) -%} {{- '<|turn>system\n' -}} {#- Inject Thinking token at the very top of the FIRST system turn -#} {%- if enable_thinking -%} {{- '<|think|>\n' -}} {%- set ns.prev_message_type = 'think' -%} {%- endif -%} {%- if messages and messages[0]['role'] in ['system', 'developer'] -%} {%- if messages[0]['content'] is string -%} {{- messages[0]['content'] | trim -}} {%- elif messages[0]['content'] is sequence -%} {%- for item in messages[0]['content'] -%} {{- item['text'] | trim + ' '-}} {%- endfor -%} {%- endif -%} {%- set loop_messages = messages[1:] -%} {%- endif -%} {%- if tools -%} {%- for tool in tools %} {{- '<|tool>' -}} {{- format_function_declaration(tool) | trim -}} {{- '' -}} {%- endfor %} {%- set ns.prev_message_type = 'tool' -%} {%- endif -%} {{- '\n' -}} {%- endif %} {#- Pre-scan: find last user message index for reasoning guard -#} {%- set ns_turn = namespace(last_user_idx=-1) -%} {%- for i in range(loop_messages | length) -%} {%- if loop_messages[i]['role'] == 'user' -%} {%- set ns_turn.last_user_idx = i -%} {%- endif -%} {%- endfor -%} {#- Loop through messages -#} {%- for message in loop_messages -%} {%- if message['role'] != 'tool' -%} {%- set ns.prev_message_type = None -%} {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%} {#- Detect continuation using tracked state — O(1) instead of O(n) backward scan -#} {%- set continue_same_model_turn = (role == 'model' and ns.prev_non_tool_role == 'assistant') -%} {%- if not continue_same_model_turn -%} {{- '<|turn>' + role + '\n' }} {%- endif -%} {#- Render reasoning/reasoning_content as thinking channel -#} {%- set thinking_text = message.get('reasoning') or message.get('reasoning_content') -%} {%- set thinking_gate = (loop.index0 > ns_turn.last_user_idx) or (preserve_thinking and message.get('tool_calls')) -%} {%- if thinking_text and thinking_gate -%} {{- '<|channel>thought\n' + thinking_text + '\n' -}} {%- endif -%} {%- if message.get('tool_calls') -%} {%- for tool_call in message.get('tool_calls') -%} {%- set function = tool_call['function'] -%} {{- '<|tool_call>call:' + function['name'] + '{' -}} {%- if function['arguments'] is mapping -%} {%- set ns_args = namespace(found_first=false) -%} {%- for key, value in function['arguments'] | dictsort -%} {%- if ns_args.found_first %},{% endif -%} {%- set ns_args.found_first = true -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- endfor -%} {%- elif function['arguments'] is none -%} {%- else -%} {{- raise_exception( "chat_template: tool_calls[].function.arguments must be a " "JSON object (mapping), not a string. Deserialize arguments " "before passing to the template." ) -}} {%- endif -%} {{- '}' -}} {%- endfor -%} {%- set ns.prev_message_type = 'tool_call' -%} {%- endif -%} {%- set ns_tr_out = namespace(flag=false) -%} {%- if message.get('tool_responses') -%} {#- Legacy: tool_responses embedded on the assistant message (Google/Gemma native) -#} {%- for tool_response in message.get('tool_responses') -%} {{- format_tool_response_block(tool_response['name'] | default('unknown', true), tool_response['response']) -}} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endfor -%} {%- elif message.get('tool_calls') -%} {#- OpenAI Chat Completions: forward-scan consecutive role:tool messages -#} {%- set ns_tool_scan = namespace(stopped=false) -%} {%- for k in range(loop.index0 + 1, loop_messages | length) -%} {%- if ns_tool_scan.stopped -%} {%- elif loop_messages[k]['role'] != 'tool' -%} {%- set ns_tool_scan.stopped = true -%} {%- else -%} {%- set follow = loop_messages[k] -%} {#- Resolve tool_call_id to function name -#} {%- set ns_tname = namespace(name=follow.get('name') or 'unknown') -%} {%- for tc in message.get('tool_calls') -%} {%- if tc.get('id') == follow.get('tool_call_id') -%} {%- set ns_tname.name = tc['function']['name'] -%} {%- endif -%} {%- endfor -%} {#- Handle content as string or content-parts array -#} {%- set tool_body = follow.get('content') -%} {%- if tool_body is string -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- elif tool_body is sequence and tool_body is not string -%} {%- set ns_txt = namespace(s='') -%} {%- for part in tool_body -%} {%- if part.get('type') == 'text' -%} {%- set ns_txt.s = ns_txt.s + (part.get('text') | default('')) -%} {%- endif -%} {%- endfor -%} {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}} {%- for part in tool_body -%} {%- if part.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif part.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif part.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- else -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- endif -%} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endif -%} {%- endfor -%} {%- endif -%} {%- set captured_content -%} {%- if message.get('content') is string -%} {%- if role == 'model' -%} {{- strip_thinking(message['content']) -}} {%- else -%} {{- message['content'] | trim -}} {%- endif -%} {%- elif message.get('content') is sequence -%} {%- for item in message['content'] -%} {%- if item.get('type') == 'text' -%} {%- if role == 'model' -%} {{- strip_thinking(item['text']) -}} {%- else -%} {{- item['text'] | trim -}} {%- endif -%} {%- elif item.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif item.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif item.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- endif -%} {%- endset -%} {{- captured_content -}} {%- set has_content = captured_content | trim | length > 0 -%} {#- Forward-scan: find next non-tool message role for continuation detection -#} {%- set next_nt = namespace(role=None, found=false) -%} {%- for j in range(loop.index0 + 1, loop_messages | length) -%} {%- if not next_nt.found -%} {%- if loop_messages[j]['role'] != 'tool' -%} {%- set next_nt.role = loop_messages[j]['role'] -%} {%- set next_nt.found = true -%} {%- endif -%} {%- endif -%} {%- endfor -%} {%- set continues_into_next = ( role == 'model' and next_nt.role == 'assistant' and (not message.get('tool_calls') or ns_tr_out.flag) ) -%} {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%} {{- '<|tool_response>' -}} {%- elif continues_into_next -%} {%- elif not (ns_tr_out.flag and not has_content and not next_nt.found) -%} {{- '\n' -}} {%- endif -%} {#- Track previous non-tool role for next iteration (avoids O(n) backward scan) -#} {%- set ns.prev_non_tool_role = message['role'] -%} {%- endif -%} {%- endfor -%} {%- if add_generation_prompt -%} {%- if ns.prev_message_type != 'tool_response' and ns.prev_message_type != 'tool_call' -%} {{- '<|turn>model\n' -}} {%- if not enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- elif ns.prev_message_type == 'tool_response' and enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- endif -%} INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/mtp-gemma-4-31B-it-QAT-BF16.gguf: n_tensors = 49, total_size = 939.1M Writing: 0%| | 0.00/939M [00:00", "", "", "", ... llama_model_loader: - kv 39: tokenizer.ggml.scores arr[f32,262144] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 40: tokenizer.ggml.token_type arr[i32,262144] = [3, 3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 41: tokenizer.ggml.merges arr[str,514906] = ["\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n", ... llama_model_loader: - kv 42: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 43: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 44: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 45: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 46: tokenizer.ggml.mask_token_id u32 = 4 llama_model_loader: - kv 47: tokenizer.chat_template str = {#\n Template: Google Gemma 4 Canonic... llama_model_loader: - kv 48: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 49: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - type f32: 26 tensors llama_model_loader: - type bf16: 23 tensors [ 1/ 49] nextn.post_projection.weight - [ 1024, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 10.50 MiB -> 2.95 MiB [ 2/ 49] nextn.pre_projection.weight - [ 10752, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 3/ 49] output_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 4/ 49] rope_freqs.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 5/ 49] token_embd.weight - [ 1024, 262144, 1, 1], type = bf16, converting to q4_0 .. size = 512.00 MiB -> 144.00 MiB [ 6/ 49] blk.0.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 7/ 49] blk.0.attn_output.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 8/ 49] blk.0.attn_q.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 9/ 49] blk.0.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 10/ 49] blk.0.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 11/ 49] blk.0.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 12/ 49] blk.0.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 13/ 49] blk.0.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 14/ 49] blk.0.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 15/ 49] blk.0.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 16/ 49] blk.0.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 17/ 49] blk.1.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 18/ 49] blk.1.attn_output.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 19/ 49] blk.1.attn_q.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 20/ 49] blk.1.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 21/ 49] blk.1.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 22/ 49] blk.1.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 23/ 49] blk.1.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 24/ 49] blk.1.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 25/ 49] blk.1.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 26/ 49] blk.1.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 27/ 49] blk.1.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 28/ 49] blk.2.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 29/ 49] blk.2.attn_output.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 30/ 49] blk.2.attn_q.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 31/ 49] blk.2.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 32/ 49] blk.2.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 33/ 49] blk.2.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 34/ 49] blk.2.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 35/ 49] blk.2.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 36/ 49] blk.2.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 37/ 49] blk.2.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 38/ 49] blk.2.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 39/ 49] blk.3.attn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 40/ 49] blk.3.attn_output.weight - [ 16384, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 32.00 MiB -> 9.00 MiB [ 41/ 49] blk.3.attn_q.weight - [ 1024, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 32.00 MiB -> 9.00 MiB [ 42/ 49] blk.3.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 43/ 49] blk.3.ffn_down.weight - [ 8192, 1024, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 44/ 49] blk.3.ffn_gate.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 45/ 49] blk.3.ffn_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 46/ 49] blk.3.ffn_up.weight - [ 1024, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 16.00 MiB -> 4.50 MiB [ 47/ 49] blk.3.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 48/ 49] blk.3.post_attention_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB [ 49/ 49] blk.3.post_ffw_norm.weight - [ 1024, 1, 1, 1], type = f32, size = 0.004 MiB llama_model_quantize_impl: model size = 895.57 MiB (16.00 BPW) llama_model_quantize_impl: quant size = 251.93 MiB (4.50 BPW) llama_quantize: quantize time = 2619.23 ms llama_quantize: total time = 2619.23 ms + python3 ./llama.cpp/convert_hf_to_gguf.py ./model-temp-gemma_4_31B_it-QAT_Q4_0_UNQUANTIZED --outtype bf16 --outfile ./upload-gemma_4_31B_it/gemma-4-31B-it-QAT-BF16.gguf --model-name gemma-4-31B-it INFO:hf-to-gguf:Loading model: model-temp-gemma_4_31B_it-QAT_Q4_0_UNQUANTIZED INFO:hf-to-gguf:Model architecture: Gemma4ForConditionalGeneration INFO:hf-to-gguf:gguf: loading model weight map from 'model.safetensors.index.json' INFO:hf-to-gguf:gguf: indexing model part 'model-00001-of-00002.safetensors' INFO:hf-to-gguf:gguf: indexing model part 'model-00002-of-00002.safetensors' INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only INFO:hf-to-gguf:Exporting model... INFO:hf-to-gguf:rope_freqs.weight, torch.float32 --> F32, shape = {256} INFO:hf-to-gguf:token_embd.weight, torch.bfloat16 --> BF16, shape = {5376, 262144} INFO:hf-to-gguf:blk.0.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.0.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.0.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.0.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.0.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.0.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.0.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.0.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.0.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.1.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.1.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.1.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.1.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.1.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.1.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.1.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.1.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.1.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.10.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.10.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.10.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.10.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.10.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.10.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.10.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.10.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.10.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.12.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.12.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.12.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.12.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.12.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.12.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.12.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.12.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.12.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.12.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.12.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.13.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.13.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.13.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.13.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.13.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.13.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.13.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.13.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.13.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.13.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.13.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.14.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.14.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.14.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.14.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.14.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.14.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.14.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.14.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.14.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.14.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.14.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.15.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.15.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.15.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.15.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.15.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.15.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.15.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.15.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.15.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.15.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.15.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.16.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.16.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.16.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.16.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.16.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.16.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.16.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.16.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.16.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.18.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.18.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.18.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.18.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.18.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.18.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.18.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.18.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.18.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.18.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.18.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.19.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.19.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.19.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.19.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.19.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.19.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.19.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.19.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.19.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.19.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.19.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.2.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.2.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.2.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.2.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.2.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.2.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.2.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.2.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.2.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.20.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.20.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.20.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.20.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.20.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.20.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.20.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.20.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.20.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.20.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.20.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.21.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.21.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.21.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.21.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.21.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.21.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.21.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.21.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.21.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.21.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.21.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.22.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.22.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.22.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.22.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.22.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.22.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.22.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.22.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.22.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.24.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.24.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.24.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.24.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.24.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.24.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.24.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.24.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.24.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.24.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.24.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.25.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.25.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.25.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.25.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.25.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.25.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.25.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.25.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.25.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.25.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.25.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.26.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.26.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.26.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.26.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.26.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.26.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.26.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.26.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.26.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.26.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.26.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.27.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.27.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.27.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.27.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.27.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.27.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.27.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.27.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.27.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.27.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.27.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.28.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.28.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.28.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.28.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.28.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.28.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.28.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.28.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.28.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.3.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.3.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.3.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.3.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.3.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.3.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.3.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.3.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.3.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.30.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.30.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.30.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.30.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.30.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.30.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.30.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.30.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.30.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.30.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.30.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.31.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.31.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.31.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.31.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.31.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.31.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.31.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.31.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.31.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.31.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.31.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.32.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.32.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.32.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.32.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.32.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.32.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.32.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.32.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.32.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.32.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.32.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.33.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.33.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.33.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.33.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.33.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.33.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.33.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.33.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.33.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.33.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.33.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.34.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.34.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.34.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.34.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.34.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.34.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.34.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.34.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.34.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.36.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.36.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.36.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.36.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.36.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.36.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.36.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.36.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.36.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.36.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.36.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.37.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.37.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.37.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.37.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.37.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.37.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.37.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.37.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.37.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.37.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.37.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.38.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.38.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.38.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.38.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.38.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.38.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.38.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.38.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.38.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.38.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.38.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.39.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.39.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.39.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.39.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.39.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.39.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.39.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.39.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.39.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.39.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.39.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.4.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.4.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.4.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.4.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.4.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.4.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.4.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.4.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.4.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.4.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.40.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.40.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.40.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.40.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.40.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.40.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.40.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.40.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.40.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.42.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.42.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.42.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.42.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.42.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.42.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.42.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.42.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.42.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.42.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.42.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.43.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.43.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.43.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.43.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.43.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.43.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.43.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.43.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.43.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.43.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.43.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.44.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.44.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.44.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.44.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.44.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.44.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.44.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.44.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.44.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.44.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.44.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.45.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.45.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.45.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.45.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.45.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.45.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.45.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.45.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.45.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.45.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.45.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.46.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.46.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.46.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.46.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.46.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.46.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.46.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.46.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.46.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.48.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.48.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.48.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.48.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.48.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.48.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.48.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.48.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.48.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.48.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.48.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.49.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.49.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.49.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.49.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.49.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.49.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.49.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.49.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.49.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.49.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.49.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.50.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.50.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.50.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.50.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.50.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.50.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.50.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.50.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.50.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.50.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.50.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.51.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.51.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.51.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.51.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.51.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.51.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.51.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.51.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.51.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.51.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.51.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.52.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.52.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.52.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.52.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.52.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.52.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.52.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.52.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.52.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.54.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.54.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.54.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.54.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.54.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.54.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.54.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.54.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.54.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.54.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.54.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.55.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.55.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.55.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.55.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.55.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.55.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.55.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.55.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.55.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.55.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.55.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.56.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.56.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.56.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.56.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.56.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.56.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.56.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.56.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.56.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.56.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.56.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.57.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.57.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.57.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.57.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.57.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.57.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.57.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.57.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.57.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.57.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.57.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.58.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.58.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.58.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.58.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.58.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.58.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.58.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.58.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.58.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.6.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.6.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.6.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.6.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.6.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.6.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.6.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.6.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.6.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.6.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.6.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.7.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.7.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.7.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.7.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.7.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.7.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.7.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.7.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.7.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.7.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.7.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.8.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.8.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.8.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.8.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.8.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.8.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.8.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.8.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.8.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.8.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.8.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.9.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.9.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.9.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.9.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.9.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.9.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.9.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:blk.9.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.9.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {256} INFO:hf-to-gguf:blk.9.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.9.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 4096} INFO:hf-to-gguf:output_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.10.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.10.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.11.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.11.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.11.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.11.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.11.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.11.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.11.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.11.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.11.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.16.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.16.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.16.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.16.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.16.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.17.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.17.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.17.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.17.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.17.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.17.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.17.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.17.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.17.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.22.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.22.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.22.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.22.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.22.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.23.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.23.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.23.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.23.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.23.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.23.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.23.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.23.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.23.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.28.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.28.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.28.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.28.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.28.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.29.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.29.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.29.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.29.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.29.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.29.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.29.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.29.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.29.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.34.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.34.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.34.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.34.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.34.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.35.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.35.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.35.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.35.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.35.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.35.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.35.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.35.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.35.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.4.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.40.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.40.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.41.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.41.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.41.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.41.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.41.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.41.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.41.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.41.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.41.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.46.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.46.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.46.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.46.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.46.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.47.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.47.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.47.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.47.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.47.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.47.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.47.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.47.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.47.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.5.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.5.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.5.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.5.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.5.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.5.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.5.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.5.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.5.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.5.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.52.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.52.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.52.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.52.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.52.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.53.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.53.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.53.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.53.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.53.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.53.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.53.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.53.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.53.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:blk.58.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.58.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.58.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.58.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.58.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.layer_output_scale.weight, torch.bfloat16 --> F32, shape = {1} INFO:hf-to-gguf:blk.59.ffn_down.weight, torch.bfloat16 --> BF16, shape = {21504, 5376} INFO:hf-to-gguf:blk.59.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.59.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 21504} INFO:hf-to-gguf:blk.59.post_attention_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.post_ffw_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.59.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.59.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 2048} INFO:hf-to-gguf:blk.59.attn_output.weight, torch.bfloat16 --> BF16, shape = {16384, 5376} INFO:hf-to-gguf:blk.59.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {512} INFO:hf-to-gguf:blk.59.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 16384} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:gguf: context length = 262144 INFO:hf-to-gguf:gguf: embedding length = 5376 INFO:hf-to-gguf:gguf: feed forward length = 21504 INFO:hf-to-gguf:gguf: head count = 32 INFO:hf-to-gguf:gguf: key-value head count = 16 WARNING:hf-to-gguf:Unknown RoPE type: proportional INFO:hf-to-gguf:gguf: rope scaling type = NONE INFO:hf-to-gguf:gguf: rope theta = 1000000.0 INFO:hf-to-gguf:gguf: rope theta swa = 10000.0 INFO:hf-to-gguf:gguf: rms norm epsilon = 1e-06 INFO:hf-to-gguf:gguf: file type = 32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.context_length', overwriting it with new value 262144 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.head_count', overwriting it with new value 32 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.layer_norm_rms_epsilon', overwriting it with new value 1e-06 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.key_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.value_length', overwriting it with new value 256 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.rope.freq_base', overwriting it with new value 1000000.0 of type FLOAT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.head_count_kv', overwriting it with new value 16 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.key_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.value_length', overwriting it with new value 512 of type UINT32 WARNING:gguf.gguf_writer:Duplicated key name 'gemma4.attention.head_count_kv', overwriting it with new value [16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4, 16, 16, 16, 16, 16, 4] of type ARRAY INFO:hf-to-gguf:Set model quantization version INFO:hf-to-gguf:Set model tokenizer INFO:hf-to-gguf:Token '<|tool_call>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|tool_response>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|"|>' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|channel>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:gguf.vocab:Adding 514906 merge(s). INFO:gguf.vocab:Setting special token type bos to 2 INFO:gguf.vocab:Setting special token type eos to 1 INFO:gguf.vocab:Setting special token type unk to 3 INFO:gguf.vocab:Setting special token type pad to 0 INFO:gguf.vocab:Setting special token type mask to 4 INFO:gguf.vocab:Setting chat_template to {# Template: Google Gemma 4 Canonical Chat Template Author: Google Gemma Engineering Team Published: 2026-07-09 Context: Fixed tool-calling loops, turn closures, and thinking content-ordering. #} {%- macro format_parameters(properties, required, filter_keys=false) -%} {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%} {%- set ns = namespace(found_first=false) -%} {%- for key, value in properties | dictsort -%} {%- set add_comma = false -%} {%- if not filter_keys or key not in standard_keys -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {{ key }}:{ {%- if value['description'] -%} description:<|"|>{{ value['description'] }}<|"|> {%- set add_comma = true -%} {%- endif -%} {%- if value['type'] | upper == 'STRING' -%} {%- if value['enum'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} enum:{{ format_argument(value['enum']) }} {%- endif -%} {%- elif value['type'] | upper == 'ARRAY' -%} {%- if value['items'] is mapping and value['items'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} items:{ {%- set ns_items = namespace(found_first=false) -%} {%- for item_key, item_value in value['items'] | dictsort -%} {%- if item_value is not none -%} {%- if ns_items.found_first %},{% endif -%} {%- set ns_items.found_first = true -%} {%- if item_key == 'properties' -%} properties:{ {%- if item_value is mapping -%} {{- format_parameters(item_value, value['items']['required'] | default([])) -}} {%- endif -%} } {%- elif item_key == 'required' -%} required:[ {%- for req_item in item_value -%} <|"|>{{- req_item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- elif item_key == 'type' -%} {%- if item_value is string -%} type:{{ format_argument(item_value | upper) }} {%- else -%} type:{{ format_argument(item_value | map('upper') | list) }} {%- endif -%} {%- else -%} {{ item_key }}:{{ format_argument(item_value) }} {%- endif -%} {%- endif -%} {%- endfor -%} } {%- endif -%} {%- endif -%} {%- if value['nullable'] %} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} nullable:true {%- endif -%} {%- if value['type'] | upper == 'OBJECT' -%} {%- if value['properties'] is defined and value['properties'] is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value['properties'], value['required'] | default([])) -}} } {%- elif value is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value, value['required'] | default([]), filter_keys=true) -}} } {%- endif -%} {%- if value['required'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} required:[ {%- for item in value['required'] | default([]) -%} <|"|>{{- item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- endif -%} {%- endif -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} type:<|"|>{{ value['type'] | upper }}<|"|>} {%- endif -%} {%- endfor -%} {%- endmacro -%} {%- macro format_function_declaration(tool_data) -%} declaration:{{- tool_data['function']['name'] -}}{description:<|"|>{{- tool_data['function']['description'] -}}<|"|> {%- set params = tool_data['function']['parameters'] -%} {%- if params -%} ,parameters:{ {%- if params['properties'] -%} properties:{ {{- format_parameters(params['properties'], params['required']) -}} }, {%- endif -%} {%- if params['required'] -%} required:[ {%- for item in params['required'] -%} <|"|>{{- item -}}<|"|> {{- ',' if not loop.last -}} {%- endfor -%} ], {%- endif -%} {%- if params['type'] -%} type:<|"|>{{- params['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} {%- if 'response' in tool_data['function'] -%} {%- set response_declaration = tool_data['function']['response'] -%} ,response:{ {%- if response_declaration['description'] -%} description:<|"|>{{- response_declaration['description'] -}}<|"|>, {%- endif -%} {%- if response_declaration['type'] | upper == 'OBJECT' -%} type:<|"|>{{- response_declaration['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} } {%- endmacro -%} {%- macro format_argument(argument, escape_keys=True) -%} {%- if argument is none -%} {{- 'null' -}} {%- elif argument is string -%} {{- '<|"|>' + argument + '<|"|>' -}} {%- elif argument is boolean -%} {{- 'true' if argument else 'false' -}} {%- elif argument is mapping -%} {{- '{' -}} {%- set ns = namespace(found_first=false) -%} {%- for key, value in argument | dictsort -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {%- if escape_keys -%} {{- '<|"|>' + key + '<|"|>' -}} {%- else -%} {{- key -}} {%- endif -%} :{{- format_argument(value, escape_keys=escape_keys) -}} {%- endfor -%} {{- '}' -}} {%- elif argument is sequence -%} {{- '[' -}} {%- for item in argument -%} {{- format_argument(item, escape_keys=escape_keys) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- ']' -}} {%- else -%} {{- argument -}} {%- endif -%} {%- endmacro -%} {%- macro strip_thinking(text) -%} {%- set ns = namespace(result='') -%} {%- for part in text.split('') -%} {%- if '<|channel>' in part -%} {%- set ns.result = ns.result + part.split('<|channel>')[0] -%} {%- else -%} {%- set ns.result = ns.result + part -%} {%- endif -%} {%- endfor -%} {{- ns.result | trim -}} {%- endmacro -%} {%- macro format_tool_response_block(tool_name, response) -%} {{- '<|tool_response>' -}} {%- if response is mapping -%} {{- 'response:' + tool_name + '{' -}} {%- for key, value in response | dictsort -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- '}' -}} {%- else -%} {{- 'response:' + tool_name + '{value:' + format_argument(response, escape_keys=False) + '}' -}} {%- endif -%} {{- '' -}} {%- endmacro -%} {#- ===== SETUP ===== -#} {%- set ns = namespace(prev_message_type=None, prev_non_tool_role=None) -%} {%- set loop_messages = messages -%} {%- set enable_thinking = enable_thinking | default(false) -%} {%- set preserve_thinking = preserve_thinking | default(false) -%} {{- bos_token -}} {#- Handle System/Tool Definitions Block -#} {%- if enable_thinking or tools or (messages and messages[0]['role'] in ['system', 'developer']) -%} {{- '<|turn>system\n' -}} {#- Inject Thinking token at the very top of the FIRST system turn -#} {%- if enable_thinking -%} {{- '<|think|>\n' -}} {%- set ns.prev_message_type = 'think' -%} {%- endif -%} {%- if messages and messages[0]['role'] in ['system', 'developer'] -%} {%- if messages[0]['content'] is string -%} {{- messages[0]['content'] | trim -}} {%- elif messages[0]['content'] is sequence -%} {%- for item in messages[0]['content'] -%} {{- item['text'] | trim + ' '-}} {%- endfor -%} {%- endif -%} {%- set loop_messages = messages[1:] -%} {%- endif -%} {%- if tools -%} {%- for tool in tools %} {{- '<|tool>' -}} {{- format_function_declaration(tool) | trim -}} {{- '' -}} {%- endfor %} {%- set ns.prev_message_type = 'tool' -%} {%- endif -%} {{- '\n' -}} {%- endif %} {#- Pre-scan: find last user message index for reasoning guard -#} {%- set ns_turn = namespace(last_user_idx=-1) -%} {%- for i in range(loop_messages | length) -%} {%- if loop_messages[i]['role'] == 'user' -%} {%- set ns_turn.last_user_idx = i -%} {%- endif -%} {%- endfor -%} {#- Loop through messages -#} {%- for message in loop_messages -%} {%- if message['role'] != 'tool' -%} {%- set ns.prev_message_type = None -%} {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%} {#- Detect continuation using tracked state — O(1) instead of O(n) backward scan -#} {%- set continue_same_model_turn = (role == 'model' and ns.prev_non_tool_role == 'assistant') -%} {%- if not continue_same_model_turn -%} {{- '<|turn>' + role + '\n' }} {%- endif -%} {#- Render reasoning/reasoning_content as thinking channel -#} {%- set thinking_text = message.get('reasoning') or message.get('reasoning_content') -%} {%- set thinking_gate = (loop.index0 > ns_turn.last_user_idx) or (preserve_thinking and message.get('tool_calls')) -%} {%- if thinking_text and thinking_gate -%} {{- '<|channel>thought\n' + thinking_text + '\n' -}} {%- endif -%} {%- if message.get('tool_calls') -%} {%- for tool_call in message.get('tool_calls') -%} {%- set function = tool_call['function'] -%} {{- '<|tool_call>call:' + function['name'] + '{' -}} {%- if function['arguments'] is mapping -%} {%- set ns_args = namespace(found_first=false) -%} {%- for key, value in function['arguments'] | dictsort -%} {%- if ns_args.found_first %},{% endif -%} {%- set ns_args.found_first = true -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- endfor -%} {%- elif function['arguments'] is none -%} {%- else -%} {{- raise_exception( "chat_template: tool_calls[].function.arguments must be a " "JSON object (mapping), not a string. Deserialize arguments " "before passing to the template." ) -}} {%- endif -%} {{- '}' -}} {%- endfor -%} {%- set ns.prev_message_type = 'tool_call' -%} {%- endif -%} {%- set ns_tr_out = namespace(flag=false) -%} {%- if message.get('tool_responses') -%} {#- Legacy: tool_responses embedded on the assistant message (Google/Gemma native) -#} {%- for tool_response in message.get('tool_responses') -%} {{- format_tool_response_block(tool_response['name'] | default('unknown', true), tool_response['response']) -}} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endfor -%} {%- elif message.get('tool_calls') -%} {#- OpenAI Chat Completions: forward-scan consecutive role:tool messages -#} {%- set ns_tool_scan = namespace(stopped=false) -%} {%- for k in range(loop.index0 + 1, loop_messages | length) -%} {%- if ns_tool_scan.stopped -%} {%- elif loop_messages[k]['role'] != 'tool' -%} {%- set ns_tool_scan.stopped = true -%} {%- else -%} {%- set follow = loop_messages[k] -%} {#- Resolve tool_call_id to function name -#} {%- set ns_tname = namespace(name=follow.get('name') or 'unknown') -%} {%- for tc in message.get('tool_calls') -%} {%- if tc.get('id') == follow.get('tool_call_id') -%} {%- set ns_tname.name = tc['function']['name'] -%} {%- endif -%} {%- endfor -%} {#- Handle content as string or content-parts array -#} {%- set tool_body = follow.get('content') -%} {%- if tool_body is string -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- elif tool_body is sequence and tool_body is not string -%} {%- set ns_txt = namespace(s='') -%} {%- for part in tool_body -%} {%- if part.get('type') == 'text' -%} {%- set ns_txt.s = ns_txt.s + (part.get('text') | default('')) -%} {%- endif -%} {%- endfor -%} {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}} {%- for part in tool_body -%} {%- if part.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif part.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif part.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- else -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- endif -%} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endif -%} {%- endfor -%} {%- endif -%} {%- set captured_content -%} {%- if message.get('content') is string -%} {%- if role == 'model' -%} {{- strip_thinking(message['content']) -}} {%- else -%} {{- message['content'] | trim -}} {%- endif -%} {%- elif message.get('content') is sequence -%} {%- for item in message['content'] -%} {%- if item.get('type') == 'text' -%} {%- if role == 'model' -%} {{- strip_thinking(item['text']) -}} {%- else -%} {{- item['text'] | trim -}} {%- endif -%} {%- elif item.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif item.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif item.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- endif -%} {%- endset -%} {{- captured_content -}} {%- set has_content = captured_content | trim | length > 0 -%} {#- Forward-scan: find next non-tool message role for continuation detection -#} {%- set next_nt = namespace(role=None, found=false) -%} {%- for j in range(loop.index0 + 1, loop_messages | length) -%} {%- if not next_nt.found -%} {%- if loop_messages[j]['role'] != 'tool' -%} {%- set next_nt.role = loop_messages[j]['role'] -%} {%- set next_nt.found = true -%} {%- endif -%} {%- endif -%} {%- endfor -%} {%- set continues_into_next = ( role == 'model' and next_nt.role == 'assistant' and (not message.get('tool_calls') or ns_tr_out.flag) ) -%} {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%} {{- '<|tool_response>' -}} {%- elif continues_into_next -%} {%- elif not (ns_tr_out.flag and not has_content and not next_nt.found) -%} {{- '\n' -}} {%- endif -%} {#- Track previous non-tool role for next iteration (avoids O(n) backward scan) -#} {%- set ns.prev_non_tool_role = message['role'] -%} {%- endif -%} {%- endfor -%} {%- if add_generation_prompt -%} {%- if ns.prev_message_type != 'tool_response' and ns.prev_message_type != 'tool_call' -%} {{- '<|turn>model\n' -}} {%- if not enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- elif ns.prev_message_type == 'tool_response' and enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- endif -%} INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/gemma-4-31B-it-QAT-BF16.gguf: n_tensors = 833, total_size = 61.4G Writing: 0%| | 0.00/61.4G [00:00", "", "", "", ... llama_model_loader: - kv 38: tokenizer.ggml.scores arr[f32,262144] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 39: tokenizer.ggml.token_type arr[i32,262144] = [3, 3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 40: tokenizer.ggml.merges arr[str,514906] = ["\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n", ... llama_model_loader: - kv 41: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 42: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 43: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 44: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 45: tokenizer.ggml.mask_token_id u32 = 4 llama_model_loader: - kv 46: tokenizer.chat_template str = {#\n Template: Google Gemma 4 Canonic... llama_model_loader: - kv 47: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 48: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - type f32: 422 tensors llama_model_loader: - type bf16: 411 tensors llama_tensor_get_type: token_embd.weight - applying manual override: q4_0 -> q8_0 [ 1/ 833] output_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 2/ 833] rope_freqs.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 3/ 833] token_embd.weight - [ 5376, 262144, 1, 1], type = bf16, converting to q8_0 .. size = 2688.00 MiB -> 1428.00 MiB [ 4/ 833] blk.0.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 5/ 833] blk.0.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 6/ 833] blk.0.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 7/ 833] blk.0.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 8/ 833] blk.0.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 9/ 833] blk.0.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 10/ 833] blk.0.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 11/ 833] blk.0.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 12/ 833] blk.0.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 13/ 833] blk.0.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 14/ 833] blk.0.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 15/ 833] blk.0.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 16/ 833] blk.0.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 17/ 833] blk.0.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 18/ 833] blk.1.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 19/ 833] blk.1.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 20/ 833] blk.1.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 21/ 833] blk.1.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 22/ 833] blk.1.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 23/ 833] blk.1.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 24/ 833] blk.1.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 25/ 833] blk.1.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 26/ 833] blk.1.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 27/ 833] blk.1.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 28/ 833] blk.1.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 29/ 833] blk.1.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 30/ 833] blk.1.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 31/ 833] blk.1.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 32/ 833] blk.2.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 33/ 833] blk.2.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 34/ 833] blk.2.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 35/ 833] blk.2.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 36/ 833] blk.2.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 37/ 833] blk.2.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 38/ 833] blk.2.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 39/ 833] blk.2.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 40/ 833] blk.2.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 41/ 833] blk.2.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 42/ 833] blk.2.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 43/ 833] blk.2.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 44/ 833] blk.2.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 45/ 833] blk.2.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 46/ 833] blk.3.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 47/ 833] blk.3.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 48/ 833] blk.3.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 49/ 833] blk.3.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 50/ 833] blk.3.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 51/ 833] blk.3.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 52/ 833] blk.3.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 53/ 833] blk.3.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 54/ 833] blk.3.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 55/ 833] blk.3.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 56/ 833] blk.3.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 57/ 833] blk.3.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 58/ 833] blk.3.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 59/ 833] blk.3.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 60/ 833] blk.4.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 61/ 833] blk.4.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 62/ 833] blk.4.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 63/ 833] blk.4.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 64/ 833] blk.4.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 65/ 833] blk.4.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 66/ 833] blk.4.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 67/ 833] blk.4.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 68/ 833] blk.4.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 69/ 833] blk.4.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 70/ 833] blk.4.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 71/ 833] blk.4.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 72/ 833] blk.4.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 73/ 833] blk.4.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 74/ 833] blk.5.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 75/ 833] blk.5.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 76/ 833] blk.5.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 77/ 833] blk.5.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 78/ 833] blk.5.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 79/ 833] blk.5.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 80/ 833] blk.5.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 81/ 833] blk.5.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 82/ 833] blk.5.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 83/ 833] blk.5.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 84/ 833] blk.5.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 85/ 833] blk.5.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 86/ 833] blk.5.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 87/ 833] blk.6.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 88/ 833] blk.6.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 89/ 833] blk.6.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 90/ 833] blk.6.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 91/ 833] blk.6.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 92/ 833] blk.6.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 93/ 833] blk.6.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 94/ 833] blk.6.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 95/ 833] blk.6.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 96/ 833] blk.6.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 97/ 833] blk.6.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 98/ 833] blk.6.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 99/ 833] blk.6.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 100/ 833] blk.6.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 101/ 833] blk.7.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 102/ 833] blk.7.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 103/ 833] blk.7.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 104/ 833] blk.7.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 105/ 833] blk.7.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 106/ 833] blk.7.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 107/ 833] blk.7.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 108/ 833] blk.7.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 109/ 833] blk.7.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 110/ 833] blk.7.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 111/ 833] blk.7.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 112/ 833] blk.7.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 113/ 833] blk.7.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 114/ 833] blk.7.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 115/ 833] blk.8.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 116/ 833] blk.8.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 117/ 833] blk.8.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 118/ 833] blk.8.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 119/ 833] blk.8.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 120/ 833] blk.8.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 121/ 833] blk.8.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 122/ 833] blk.8.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 123/ 833] blk.8.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 124/ 833] blk.8.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 125/ 833] blk.8.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 126/ 833] blk.8.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 127/ 833] blk.8.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 128/ 833] blk.8.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 129/ 833] blk.9.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 130/ 833] blk.9.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 131/ 833] blk.9.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 132/ 833] blk.9.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 133/ 833] blk.9.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 134/ 833] blk.9.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 135/ 833] blk.9.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 136/ 833] blk.9.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 137/ 833] blk.9.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 138/ 833] blk.9.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 139/ 833] blk.9.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 140/ 833] blk.9.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 141/ 833] blk.9.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 142/ 833] blk.9.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 143/ 833] blk.10.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 144/ 833] blk.10.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 145/ 833] blk.10.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 146/ 833] blk.10.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 147/ 833] blk.10.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 148/ 833] blk.10.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 149/ 833] blk.10.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 150/ 833] blk.10.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 151/ 833] blk.10.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 152/ 833] blk.10.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 153/ 833] blk.10.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 154/ 833] blk.10.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 155/ 833] blk.10.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 156/ 833] blk.10.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 157/ 833] blk.11.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 158/ 833] blk.11.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 159/ 833] blk.11.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 160/ 833] blk.11.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 161/ 833] blk.11.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 162/ 833] blk.11.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 163/ 833] blk.11.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 164/ 833] blk.11.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 165/ 833] blk.11.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 166/ 833] blk.11.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 167/ 833] blk.11.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 168/ 833] blk.11.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 169/ 833] blk.11.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 170/ 833] blk.12.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 171/ 833] blk.12.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 172/ 833] blk.12.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 173/ 833] blk.12.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 174/ 833] blk.12.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 175/ 833] blk.12.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 176/ 833] blk.12.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 177/ 833] blk.12.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 178/ 833] blk.12.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 179/ 833] blk.12.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 180/ 833] blk.12.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 181/ 833] blk.12.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 182/ 833] blk.12.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 183/ 833] blk.12.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 184/ 833] blk.13.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 185/ 833] blk.13.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 186/ 833] blk.13.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 187/ 833] blk.13.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 188/ 833] blk.13.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 189/ 833] blk.13.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 190/ 833] blk.13.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 191/ 833] blk.13.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 192/ 833] blk.13.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 193/ 833] blk.13.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 194/ 833] blk.13.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 195/ 833] blk.13.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 196/ 833] blk.13.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 197/ 833] blk.13.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 198/ 833] blk.14.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 199/ 833] blk.14.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 200/ 833] blk.14.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 201/ 833] blk.14.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 202/ 833] blk.14.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 203/ 833] blk.14.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 204/ 833] blk.14.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 205/ 833] blk.14.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 206/ 833] blk.14.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 207/ 833] blk.14.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 208/ 833] blk.14.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 209/ 833] blk.14.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 210/ 833] blk.14.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 211/ 833] blk.14.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 212/ 833] blk.15.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 213/ 833] blk.15.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 214/ 833] blk.15.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 215/ 833] blk.15.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 216/ 833] blk.15.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 217/ 833] blk.15.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 218/ 833] blk.15.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 219/ 833] blk.15.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 220/ 833] blk.15.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 221/ 833] blk.15.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 222/ 833] blk.15.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 223/ 833] blk.15.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 224/ 833] blk.15.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 225/ 833] blk.15.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 226/ 833] blk.16.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 227/ 833] blk.16.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 228/ 833] blk.16.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 229/ 833] blk.16.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 230/ 833] blk.16.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 231/ 833] blk.16.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 232/ 833] blk.16.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 233/ 833] blk.16.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 234/ 833] blk.16.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 235/ 833] blk.16.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 236/ 833] blk.16.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 237/ 833] blk.16.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 238/ 833] blk.16.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 239/ 833] blk.16.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 240/ 833] blk.17.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 241/ 833] blk.17.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 242/ 833] blk.17.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 243/ 833] blk.17.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 244/ 833] blk.17.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 245/ 833] blk.17.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 246/ 833] blk.17.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 247/ 833] blk.17.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 248/ 833] blk.17.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 249/ 833] blk.17.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 250/ 833] blk.17.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 251/ 833] blk.17.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 252/ 833] blk.17.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 253/ 833] blk.18.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 254/ 833] blk.18.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 255/ 833] blk.18.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 256/ 833] blk.18.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 257/ 833] blk.18.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 258/ 833] blk.18.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 259/ 833] blk.18.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 260/ 833] blk.18.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 261/ 833] blk.18.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 262/ 833] blk.18.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 263/ 833] blk.18.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 264/ 833] blk.18.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 265/ 833] blk.18.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 266/ 833] blk.18.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 267/ 833] blk.19.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 268/ 833] blk.19.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 269/ 833] blk.19.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 270/ 833] blk.19.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 271/ 833] blk.19.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 272/ 833] blk.19.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 273/ 833] blk.19.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 274/ 833] blk.19.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 275/ 833] blk.19.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 276/ 833] blk.19.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 277/ 833] blk.19.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 278/ 833] blk.19.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 279/ 833] blk.19.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 280/ 833] blk.19.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 281/ 833] blk.20.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 282/ 833] blk.20.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 283/ 833] blk.20.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 284/ 833] blk.20.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 285/ 833] blk.20.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 286/ 833] blk.20.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 287/ 833] blk.20.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 288/ 833] blk.20.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 289/ 833] blk.20.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 290/ 833] blk.20.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 291/ 833] blk.20.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 292/ 833] blk.20.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 293/ 833] blk.20.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 294/ 833] blk.20.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 295/ 833] blk.21.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 296/ 833] blk.21.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 297/ 833] blk.21.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 298/ 833] blk.21.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 299/ 833] blk.21.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 300/ 833] blk.21.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 301/ 833] blk.21.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 302/ 833] blk.21.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 303/ 833] blk.21.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 304/ 833] blk.21.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 305/ 833] blk.21.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 306/ 833] blk.21.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 307/ 833] blk.21.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 308/ 833] blk.21.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 309/ 833] blk.22.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 310/ 833] blk.22.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 311/ 833] blk.22.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 312/ 833] blk.22.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 313/ 833] blk.22.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 314/ 833] blk.22.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 315/ 833] blk.22.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 316/ 833] blk.22.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 317/ 833] blk.22.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 318/ 833] blk.22.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 319/ 833] blk.22.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 320/ 833] blk.22.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 321/ 833] blk.22.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 322/ 833] blk.22.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 323/ 833] blk.23.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 324/ 833] blk.23.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 325/ 833] blk.23.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 326/ 833] blk.23.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 327/ 833] blk.23.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 328/ 833] blk.23.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 329/ 833] blk.23.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 330/ 833] blk.23.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 331/ 833] blk.23.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 332/ 833] blk.23.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 333/ 833] blk.23.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 334/ 833] blk.23.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 335/ 833] blk.23.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 336/ 833] blk.24.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 337/ 833] blk.24.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 338/ 833] blk.24.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 339/ 833] blk.24.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 340/ 833] blk.24.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 341/ 833] blk.24.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 342/ 833] blk.24.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 343/ 833] blk.24.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 344/ 833] blk.24.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 345/ 833] blk.24.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 346/ 833] blk.24.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 347/ 833] blk.24.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 348/ 833] blk.24.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 349/ 833] blk.24.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 350/ 833] blk.25.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 351/ 833] blk.25.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 352/ 833] blk.25.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 353/ 833] blk.25.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 354/ 833] blk.25.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 355/ 833] blk.25.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 356/ 833] blk.25.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 357/ 833] blk.25.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 358/ 833] blk.25.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 359/ 833] blk.25.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 360/ 833] blk.25.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 361/ 833] blk.25.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 362/ 833] blk.25.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 363/ 833] blk.25.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 364/ 833] blk.26.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 365/ 833] blk.26.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 366/ 833] blk.26.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 367/ 833] blk.26.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 368/ 833] blk.26.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 369/ 833] blk.26.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 370/ 833] blk.26.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 371/ 833] blk.26.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 372/ 833] blk.26.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 373/ 833] blk.26.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 374/ 833] blk.26.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 375/ 833] blk.26.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 376/ 833] blk.26.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 377/ 833] blk.26.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 378/ 833] blk.27.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 379/ 833] blk.27.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 380/ 833] blk.27.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 381/ 833] blk.27.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 382/ 833] blk.27.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 383/ 833] blk.27.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 384/ 833] blk.27.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 385/ 833] blk.27.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 386/ 833] blk.27.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 387/ 833] blk.27.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 388/ 833] blk.27.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 389/ 833] blk.27.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 390/ 833] blk.27.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 391/ 833] blk.27.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 392/ 833] blk.28.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 393/ 833] blk.28.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 394/ 833] blk.28.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 395/ 833] blk.28.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 396/ 833] blk.28.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 397/ 833] blk.28.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 398/ 833] blk.28.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 399/ 833] blk.28.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 400/ 833] blk.28.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 401/ 833] blk.28.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 402/ 833] blk.28.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 403/ 833] blk.28.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 404/ 833] blk.28.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 405/ 833] blk.28.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 406/ 833] blk.29.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 407/ 833] blk.29.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 408/ 833] blk.29.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 409/ 833] blk.29.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 410/ 833] blk.29.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 411/ 833] blk.29.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 412/ 833] blk.29.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 413/ 833] blk.29.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 414/ 833] blk.29.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 415/ 833] blk.29.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 416/ 833] blk.29.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 417/ 833] blk.29.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 418/ 833] blk.29.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 419/ 833] blk.30.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 420/ 833] blk.30.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 421/ 833] blk.30.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 422/ 833] blk.30.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 423/ 833] blk.30.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 424/ 833] blk.30.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 425/ 833] blk.30.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 426/ 833] blk.30.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 427/ 833] blk.30.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 428/ 833] blk.30.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 429/ 833] blk.30.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 430/ 833] blk.30.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 431/ 833] blk.30.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 432/ 833] blk.30.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 433/ 833] blk.31.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 434/ 833] blk.31.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 435/ 833] blk.31.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 436/ 833] blk.31.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 437/ 833] blk.31.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 438/ 833] blk.31.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 439/ 833] blk.31.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 440/ 833] blk.31.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 441/ 833] blk.31.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 442/ 833] blk.31.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 443/ 833] blk.31.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 444/ 833] blk.31.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 445/ 833] blk.31.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 446/ 833] blk.31.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 447/ 833] blk.32.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 448/ 833] blk.32.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 449/ 833] blk.32.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 450/ 833] blk.32.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 451/ 833] blk.32.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 452/ 833] blk.32.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 453/ 833] blk.32.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 454/ 833] blk.32.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 455/ 833] blk.32.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 456/ 833] blk.32.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 457/ 833] blk.32.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 458/ 833] blk.32.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 459/ 833] blk.32.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 460/ 833] blk.32.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 461/ 833] blk.33.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 462/ 833] blk.33.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 463/ 833] blk.33.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 464/ 833] blk.33.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 465/ 833] blk.33.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 466/ 833] blk.33.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 467/ 833] blk.33.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 468/ 833] blk.33.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 469/ 833] blk.33.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 470/ 833] blk.33.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 471/ 833] blk.33.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 472/ 833] blk.33.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 473/ 833] blk.33.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 474/ 833] blk.33.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 475/ 833] blk.34.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 476/ 833] blk.34.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 477/ 833] blk.34.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 478/ 833] blk.34.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 479/ 833] blk.34.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 480/ 833] blk.34.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 481/ 833] blk.34.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 482/ 833] blk.34.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 483/ 833] blk.34.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 484/ 833] blk.34.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 485/ 833] blk.34.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 486/ 833] blk.34.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 487/ 833] blk.34.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 488/ 833] blk.34.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 489/ 833] blk.35.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 490/ 833] blk.35.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 491/ 833] blk.35.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 492/ 833] blk.35.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 493/ 833] blk.35.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 494/ 833] blk.35.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 495/ 833] blk.35.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 496/ 833] blk.35.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 497/ 833] blk.35.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 498/ 833] blk.35.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 499/ 833] blk.35.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 500/ 833] blk.35.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 501/ 833] blk.35.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 502/ 833] blk.36.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 503/ 833] blk.36.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 504/ 833] blk.36.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 505/ 833] blk.36.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 506/ 833] blk.36.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 507/ 833] blk.36.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 508/ 833] blk.36.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 509/ 833] blk.36.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 510/ 833] blk.36.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 511/ 833] blk.36.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 512/ 833] blk.36.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 513/ 833] blk.36.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 514/ 833] blk.36.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 515/ 833] blk.36.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 516/ 833] blk.37.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 517/ 833] blk.37.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 518/ 833] blk.37.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 519/ 833] blk.37.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 520/ 833] blk.37.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 521/ 833] blk.37.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 522/ 833] blk.37.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 523/ 833] blk.37.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 524/ 833] blk.37.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 525/ 833] blk.37.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 526/ 833] blk.37.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 527/ 833] blk.37.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 528/ 833] blk.37.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 529/ 833] blk.37.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 530/ 833] blk.38.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 531/ 833] blk.38.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 532/ 833] blk.38.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 533/ 833] blk.38.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 534/ 833] blk.38.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 535/ 833] blk.38.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 536/ 833] blk.38.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 537/ 833] blk.38.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 538/ 833] blk.38.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 539/ 833] blk.38.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 540/ 833] blk.38.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 541/ 833] blk.38.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 542/ 833] blk.38.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 543/ 833] blk.38.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 544/ 833] blk.39.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 545/ 833] blk.39.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 546/ 833] blk.39.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 547/ 833] blk.39.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 548/ 833] blk.39.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 549/ 833] blk.39.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 550/ 833] blk.39.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 551/ 833] blk.39.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 552/ 833] blk.39.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 553/ 833] blk.39.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 554/ 833] blk.39.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 555/ 833] blk.39.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 556/ 833] blk.39.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 557/ 833] blk.39.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 558/ 833] blk.40.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 559/ 833] blk.40.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 560/ 833] blk.40.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 561/ 833] blk.40.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 562/ 833] blk.40.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 563/ 833] blk.40.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 564/ 833] blk.40.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 565/ 833] blk.40.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 566/ 833] blk.40.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 567/ 833] blk.40.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 568/ 833] blk.40.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 569/ 833] blk.40.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 570/ 833] blk.40.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 571/ 833] blk.40.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 572/ 833] blk.41.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 573/ 833] blk.41.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 574/ 833] blk.41.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 575/ 833] blk.41.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 576/ 833] blk.41.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 577/ 833] blk.41.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 578/ 833] blk.41.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 579/ 833] blk.41.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 580/ 833] blk.41.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 581/ 833] blk.41.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 582/ 833] blk.41.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 583/ 833] blk.41.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 584/ 833] blk.41.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 585/ 833] blk.42.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 586/ 833] blk.42.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 587/ 833] blk.42.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 588/ 833] blk.42.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 589/ 833] blk.42.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 590/ 833] blk.42.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 591/ 833] blk.42.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 592/ 833] blk.42.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 593/ 833] blk.42.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 594/ 833] blk.42.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 595/ 833] blk.42.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 596/ 833] blk.42.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 597/ 833] blk.42.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 598/ 833] blk.42.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 599/ 833] blk.43.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 600/ 833] blk.43.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 601/ 833] blk.43.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 602/ 833] blk.43.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 603/ 833] blk.43.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 604/ 833] blk.43.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 605/ 833] blk.43.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 606/ 833] blk.43.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 607/ 833] blk.43.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 608/ 833] blk.43.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 609/ 833] blk.43.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 610/ 833] blk.43.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 611/ 833] blk.43.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 612/ 833] blk.43.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 613/ 833] blk.44.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 614/ 833] blk.44.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 615/ 833] blk.44.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 616/ 833] blk.44.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 617/ 833] blk.44.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 618/ 833] blk.44.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 619/ 833] blk.44.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 620/ 833] blk.44.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 621/ 833] blk.44.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 622/ 833] blk.44.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 623/ 833] blk.44.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 624/ 833] blk.44.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 625/ 833] blk.44.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 626/ 833] blk.44.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 627/ 833] blk.45.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 628/ 833] blk.45.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 629/ 833] blk.45.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 630/ 833] blk.45.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 631/ 833] blk.45.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 632/ 833] blk.45.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 633/ 833] blk.45.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 634/ 833] blk.45.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 635/ 833] blk.45.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 636/ 833] blk.45.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 637/ 833] blk.45.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 638/ 833] blk.45.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 639/ 833] blk.45.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 640/ 833] blk.45.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 641/ 833] blk.46.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 642/ 833] blk.46.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 643/ 833] blk.46.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 644/ 833] blk.46.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 645/ 833] blk.46.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 646/ 833] blk.46.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 647/ 833] blk.46.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 648/ 833] blk.46.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 649/ 833] blk.46.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 650/ 833] blk.46.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 651/ 833] blk.46.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 652/ 833] blk.46.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 653/ 833] blk.46.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 654/ 833] blk.46.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 655/ 833] blk.47.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 656/ 833] blk.47.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 657/ 833] blk.47.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 658/ 833] blk.47.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 659/ 833] blk.47.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 660/ 833] blk.47.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 661/ 833] blk.47.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 662/ 833] blk.47.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 663/ 833] blk.47.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 664/ 833] blk.47.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 665/ 833] blk.47.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 666/ 833] blk.47.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 667/ 833] blk.47.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 668/ 833] blk.48.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 669/ 833] blk.48.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 670/ 833] blk.48.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 671/ 833] blk.48.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 672/ 833] blk.48.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 673/ 833] blk.48.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 674/ 833] blk.48.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 675/ 833] blk.48.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 676/ 833] blk.48.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 677/ 833] blk.48.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 678/ 833] blk.48.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 679/ 833] blk.48.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 680/ 833] blk.48.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 681/ 833] blk.48.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 682/ 833] blk.49.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 683/ 833] blk.49.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 684/ 833] blk.49.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 685/ 833] blk.49.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 686/ 833] blk.49.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 687/ 833] blk.49.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 688/ 833] blk.49.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 689/ 833] blk.49.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 690/ 833] blk.49.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 691/ 833] blk.49.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 692/ 833] blk.49.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 693/ 833] blk.49.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 694/ 833] blk.49.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 695/ 833] blk.49.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 696/ 833] blk.50.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 697/ 833] blk.50.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 698/ 833] blk.50.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 699/ 833] blk.50.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 700/ 833] blk.50.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 701/ 833] blk.50.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 702/ 833] blk.50.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 703/ 833] blk.50.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 704/ 833] blk.50.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 705/ 833] blk.50.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 706/ 833] blk.50.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 707/ 833] blk.50.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 708/ 833] blk.50.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 709/ 833] blk.50.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 710/ 833] blk.51.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 711/ 833] blk.51.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 712/ 833] blk.51.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 713/ 833] blk.51.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 714/ 833] blk.51.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 715/ 833] blk.51.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 716/ 833] blk.51.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 717/ 833] blk.51.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 718/ 833] blk.51.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 719/ 833] blk.51.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 720/ 833] blk.51.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 721/ 833] blk.51.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 722/ 833] blk.51.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 723/ 833] blk.51.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 724/ 833] blk.52.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 725/ 833] blk.52.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 726/ 833] blk.52.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 727/ 833] blk.52.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 728/ 833] blk.52.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 729/ 833] blk.52.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 730/ 833] blk.52.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 731/ 833] blk.52.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 732/ 833] blk.52.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 733/ 833] blk.52.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 734/ 833] blk.52.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 735/ 833] blk.52.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 736/ 833] blk.52.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 737/ 833] blk.52.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 738/ 833] blk.53.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 739/ 833] blk.53.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 740/ 833] blk.53.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 741/ 833] blk.53.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 742/ 833] blk.53.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 743/ 833] blk.53.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 744/ 833] blk.53.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 745/ 833] blk.53.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 746/ 833] blk.53.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 747/ 833] blk.53.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 748/ 833] blk.53.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 749/ 833] blk.53.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 750/ 833] blk.53.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 751/ 833] blk.54.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 752/ 833] blk.54.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 753/ 833] blk.54.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 754/ 833] blk.54.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 755/ 833] blk.54.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 756/ 833] blk.54.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 757/ 833] blk.54.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 758/ 833] blk.54.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 759/ 833] blk.54.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 760/ 833] blk.54.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 761/ 833] blk.54.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 762/ 833] blk.54.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 763/ 833] blk.54.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 764/ 833] blk.54.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 765/ 833] blk.55.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 766/ 833] blk.55.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 767/ 833] blk.55.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 768/ 833] blk.55.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 769/ 833] blk.55.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 770/ 833] blk.55.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 771/ 833] blk.55.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 772/ 833] blk.55.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 773/ 833] blk.55.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 774/ 833] blk.55.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 775/ 833] blk.55.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 776/ 833] blk.55.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 777/ 833] blk.55.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 778/ 833] blk.55.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 779/ 833] blk.56.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 780/ 833] blk.56.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 781/ 833] blk.56.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 782/ 833] blk.56.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 783/ 833] blk.56.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 784/ 833] blk.56.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 785/ 833] blk.56.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 786/ 833] blk.56.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 787/ 833] blk.56.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 788/ 833] blk.56.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 789/ 833] blk.56.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 790/ 833] blk.56.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 791/ 833] blk.56.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 792/ 833] blk.56.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 793/ 833] blk.57.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 794/ 833] blk.57.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 795/ 833] blk.57.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 796/ 833] blk.57.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 797/ 833] blk.57.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 798/ 833] blk.57.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 799/ 833] blk.57.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 800/ 833] blk.57.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 801/ 833] blk.57.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 802/ 833] blk.57.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 803/ 833] blk.57.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 804/ 833] blk.57.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 805/ 833] blk.57.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 806/ 833] blk.57.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 807/ 833] blk.58.attn_k.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 808/ 833] blk.58.attn_k_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 809/ 833] blk.58.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 810/ 833] blk.58.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 811/ 833] blk.58.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q4_0 .. size = 84.00 MiB -> 23.62 MiB [ 812/ 833] blk.58.attn_q_norm.weight - [ 256, 1, 1, 1], type = f32, size = 0.001 MiB [ 813/ 833] blk.58.attn_v.weight - [ 5376, 4096, 1, 1], type = bf16, converting to q4_0 .. size = 42.00 MiB -> 11.81 MiB [ 814/ 833] blk.58.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 815/ 833] blk.58.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 816/ 833] blk.58.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 817/ 833] blk.58.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 818/ 833] blk.58.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 819/ 833] blk.58.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 820/ 833] blk.58.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 821/ 833] blk.59.attn_k.weight - [ 5376, 2048, 1, 1], type = bf16, converting to q4_0 .. size = 21.00 MiB -> 5.91 MiB [ 822/ 833] blk.59.attn_k_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 823/ 833] blk.59.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 824/ 833] blk.59.attn_output.weight - [ 16384, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 825/ 833] blk.59.attn_q.weight - [ 5376, 16384, 1, 1], type = bf16, converting to q4_0 .. size = 168.00 MiB -> 47.25 MiB [ 826/ 833] blk.59.attn_q_norm.weight - [ 512, 1, 1, 1], type = f32, size = 0.002 MiB [ 827/ 833] blk.59.ffn_down.weight - [ 21504, 5376, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 828/ 833] blk.59.ffn_gate.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 829/ 833] blk.59.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 830/ 833] blk.59.ffn_up.weight - [ 5376, 21504, 1, 1], type = bf16, converting to q4_0 .. size = 220.50 MiB -> 62.02 MiB [ 831/ 833] blk.59.layer_output_scale.weight - [ 1, 1, 1, 1], type = f32, size = 0.000 MiB [ 832/ 833] blk.59.post_attention_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 833/ 833] blk.59.post_ffw_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB llama_model_quantize_impl: model size = 58553.08 MiB (16.00 BPW) llama_model_quantize_impl: quant size = 17143.71 MiB (4.68 BPW) llama_quantize: quantize time = 162759.45 ms llama_quantize: total time = 162759.45 ms + python3 ./llama.cpp/convert_hf_to_gguf.py ./model-temp-gemma_4_31B_it-DFLASH --outtype bf16 --target-model ./model-temp-gemma_4_31B_it-PRIMARY --outfile ./upload-gemma_4_31B_it/dflash-gemma-4-31B-it-BF16.gguf --model-name gemma-4-31B-it INFO:hf-to-gguf:Loading model: model-temp-gemma_4_31B_it-DFLASH INFO:hf-to-gguf:Model architecture: DFlashDraftModel INFO:hf-to-gguf:gguf: indexing model part 'model.safetensors' INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only INFO:hf-to-gguf:Exporting model... INFO:hf-to-gguf:fc.weight, torch.bfloat16 --> BF16, shape = {32256, 5376} INFO:hf-to-gguf:enc.output_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.ffn_down.weight, torch.bfloat16 --> BF16, shape = {10752, 5376} INFO:hf-to-gguf:blk.0.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.0.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.0.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.0.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.0.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.0.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.0.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.0.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.1.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.ffn_down.weight, torch.bfloat16 --> BF16, shape = {10752, 5376} INFO:hf-to-gguf:blk.1.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.1.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.1.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.1.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.1.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.1.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.1.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.1.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.2.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.ffn_down.weight, torch.bfloat16 --> BF16, shape = {10752, 5376} INFO:hf-to-gguf:blk.2.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.2.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.2.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.2.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.2.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.2.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.2.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.2.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.3.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.ffn_down.weight, torch.bfloat16 --> BF16, shape = {10752, 5376} INFO:hf-to-gguf:blk.3.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.3.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.3.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.3.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.3.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.3.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.3.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.3.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.4.attn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.ffn_down.weight, torch.bfloat16 --> BF16, shape = {10752, 5376} INFO:hf-to-gguf:blk.4.ffn_gate.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.4.ffn_up.weight, torch.bfloat16 --> BF16, shape = {5376, 10752} INFO:hf-to-gguf:blk.4.ffn_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:blk.4.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.4.attn_k.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:blk.4.attn_output.weight, torch.bfloat16 --> BF16, shape = {8192, 5376} INFO:hf-to-gguf:blk.4.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {128} INFO:hf-to-gguf:blk.4.attn_q.weight, torch.bfloat16 --> BF16, shape = {5376, 8192} INFO:hf-to-gguf:blk.4.attn_v.weight, torch.bfloat16 --> BF16, shape = {5376, 1024} INFO:hf-to-gguf:output_norm.weight, torch.bfloat16 --> F32, shape = {5376} INFO:hf-to-gguf:Set meta model INFO:hf-to-gguf:Set model parameters INFO:hf-to-gguf:gguf: context length = 262144 INFO:hf-to-gguf:gguf: embedding length = 5376 INFO:hf-to-gguf:gguf: feed forward length = 10752 INFO:hf-to-gguf:gguf: head count = 64 INFO:hf-to-gguf:gguf: key-value head count = 8 WARNING:hf-to-gguf:Unknown RoPE type: default INFO:hf-to-gguf:gguf: rope scaling type = NONE INFO:hf-to-gguf:gguf: rope theta = 1000000 INFO:hf-to-gguf:gguf: rms norm epsilon = 1e-06 INFO:hf-to-gguf:gguf: file type = 32 INFO:hf-to-gguf:Set model quantization version INFO:hf-to-gguf:Set model tokenizer INFO:hf-to-gguf:DFlash: Using tokenizer from target model: model-temp-gemma_4_31B_it-PRIMARY INFO:hf-to-gguf:Token '<|tool_call>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|tool_response>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|"|>' is set to USER_DEFINED INFO:hf-to-gguf:Token '<|channel>' is set to USER_DEFINED INFO:hf-to-gguf:Token '' is set to USER_DEFINED INFO:gguf.vocab:Adding 514906 merge(s). INFO:gguf.vocab:Setting special token type bos to 2 INFO:gguf.vocab:Setting special token type eos to 1 INFO:gguf.vocab:Setting special token type unk to 3 INFO:gguf.vocab:Setting special token type pad to 0 INFO:gguf.vocab:Setting special token type mask to 4 INFO:gguf.vocab:Setting chat_template to {# Template: Google Gemma 4 Canonical Chat Template Author: Google Gemma Engineering Team Published: 2026-07-09 Context: Fixed tool-calling loops, turn closures, and thinking content-ordering. #} {%- macro format_parameters(properties, required, filter_keys=false) -%} {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%} {%- set ns = namespace(found_first=false) -%} {%- for key, value in properties | dictsort -%} {%- set add_comma = false -%} {%- if not filter_keys or key not in standard_keys -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {{ key }}:{ {%- if value['description'] -%} description:<|"|>{{ value['description'] }}<|"|> {%- set add_comma = true -%} {%- endif -%} {%- if value['type'] | upper == 'STRING' -%} {%- if value['enum'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} enum:{{ format_argument(value['enum']) }} {%- endif -%} {%- elif value['type'] | upper == 'ARRAY' -%} {%- if value['items'] is mapping and value['items'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} items:{ {%- set ns_items = namespace(found_first=false) -%} {%- for item_key, item_value in value['items'] | dictsort -%} {%- if item_value is not none -%} {%- if ns_items.found_first %},{% endif -%} {%- set ns_items.found_first = true -%} {%- if item_key == 'properties' -%} properties:{ {%- if item_value is mapping -%} {{- format_parameters(item_value, value['items']['required'] | default([])) -}} {%- endif -%} } {%- elif item_key == 'required' -%} required:[ {%- for req_item in item_value -%} <|"|>{{- req_item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- elif item_key == 'type' -%} {%- if item_value is string -%} type:{{ format_argument(item_value | upper) }} {%- else -%} type:{{ format_argument(item_value | map('upper') | list) }} {%- endif -%} {%- else -%} {{ item_key }}:{{ format_argument(item_value) }} {%- endif -%} {%- endif -%} {%- endfor -%} } {%- endif -%} {%- endif -%} {%- if value['nullable'] %} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} nullable:true {%- endif -%} {%- if value['type'] | upper == 'OBJECT' -%} {%- if value['properties'] is defined and value['properties'] is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value['properties'], value['required'] | default([])) -}} } {%- elif value is mapping -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} properties:{ {{- format_parameters(value, value['required'] | default([]), filter_keys=true) -}} } {%- endif -%} {%- if value['required'] -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} required:[ {%- for item in value['required'] | default([]) -%} <|"|>{{- item -}}<|"|> {%- if not loop.last %},{% endif -%} {%- endfor -%} ] {%- endif -%} {%- endif -%} {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%} type:<|"|>{{ value['type'] | upper }}<|"|>} {%- endif -%} {%- endfor -%} {%- endmacro -%} {%- macro format_function_declaration(tool_data) -%} declaration:{{- tool_data['function']['name'] -}}{description:<|"|>{{- tool_data['function']['description'] -}}<|"|> {%- set params = tool_data['function']['parameters'] -%} {%- if params -%} ,parameters:{ {%- if params['properties'] -%} properties:{ {{- format_parameters(params['properties'], params['required']) -}} }, {%- endif -%} {%- if params['required'] -%} required:[ {%- for item in params['required'] -%} <|"|>{{- item -}}<|"|> {{- ',' if not loop.last -}} {%- endfor -%} ], {%- endif -%} {%- if params['type'] -%} type:<|"|>{{- params['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} {%- if 'response' in tool_data['function'] -%} {%- set response_declaration = tool_data['function']['response'] -%} ,response:{ {%- if response_declaration['description'] -%} description:<|"|>{{- response_declaration['description'] -}}<|"|>, {%- endif -%} {%- if response_declaration['type'] | upper == 'OBJECT' -%} type:<|"|>{{- response_declaration['type'] | upper -}}<|"|>} {%- endif -%} {%- endif -%} } {%- endmacro -%} {%- macro format_argument(argument, escape_keys=True) -%} {%- if argument is none -%} {{- 'null' -}} {%- elif argument is string -%} {{- '<|"|>' + argument + '<|"|>' -}} {%- elif argument is boolean -%} {{- 'true' if argument else 'false' -}} {%- elif argument is mapping -%} {{- '{' -}} {%- set ns = namespace(found_first=false) -%} {%- for key, value in argument | dictsort -%} {%- if ns.found_first %},{% endif -%} {%- set ns.found_first = true -%} {%- if escape_keys -%} {{- '<|"|>' + key + '<|"|>' -}} {%- else -%} {{- key -}} {%- endif -%} :{{- format_argument(value, escape_keys=escape_keys) -}} {%- endfor -%} {{- '}' -}} {%- elif argument is sequence -%} {{- '[' -}} {%- for item in argument -%} {{- format_argument(item, escape_keys=escape_keys) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- ']' -}} {%- else -%} {{- argument -}} {%- endif -%} {%- endmacro -%} {%- macro strip_thinking(text) -%} {%- set ns = namespace(result='') -%} {%- for part in text.split('') -%} {%- if '<|channel>' in part -%} {%- set ns.result = ns.result + part.split('<|channel>')[0] -%} {%- else -%} {%- set ns.result = ns.result + part -%} {%- endif -%} {%- endfor -%} {{- ns.result | trim -}} {%- endmacro -%} {%- macro format_tool_response_block(tool_name, response) -%} {{- '<|tool_response>' -}} {%- if response is mapping -%} {{- 'response:' + tool_name + '{' -}} {%- for key, value in response | dictsort -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- if not loop.last %},{% endif -%} {%- endfor -%} {{- '}' -}} {%- else -%} {{- 'response:' + tool_name + '{value:' + format_argument(response, escape_keys=False) + '}' -}} {%- endif -%} {{- '' -}} {%- endmacro -%} {#- ===== SETUP ===== -#} {%- set ns = namespace(prev_message_type=None, prev_non_tool_role=None) -%} {%- set loop_messages = messages -%} {%- set enable_thinking = enable_thinking | default(false) -%} {%- set preserve_thinking = preserve_thinking | default(false) -%} {{- bos_token -}} {#- Handle System/Tool Definitions Block -#} {%- if enable_thinking or tools or (messages and messages[0]['role'] in ['system', 'developer']) -%} {{- '<|turn>system\n' -}} {#- Inject Thinking token at the very top of the FIRST system turn -#} {%- if enable_thinking -%} {{- '<|think|>\n' -}} {%- set ns.prev_message_type = 'think' -%} {%- endif -%} {%- if messages and messages[0]['role'] in ['system', 'developer'] -%} {%- if messages[0]['content'] is string -%} {{- messages[0]['content'] | trim -}} {%- elif messages[0]['content'] is sequence -%} {%- for item in messages[0]['content'] -%} {{- item['text'] | trim + ' '-}} {%- endfor -%} {%- endif -%} {%- set loop_messages = messages[1:] -%} {%- endif -%} {%- if tools -%} {%- for tool in tools %} {{- '<|tool>' -}} {{- format_function_declaration(tool) | trim -}} {{- '' -}} {%- endfor %} {%- set ns.prev_message_type = 'tool' -%} {%- endif -%} {{- '\n' -}} {%- endif %} {#- Pre-scan: find last user message index for reasoning guard -#} {%- set ns_turn = namespace(last_user_idx=-1) -%} {%- for i in range(loop_messages | length) -%} {%- if loop_messages[i]['role'] == 'user' -%} {%- set ns_turn.last_user_idx = i -%} {%- endif -%} {%- endfor -%} {#- Loop through messages -#} {%- for message in loop_messages -%} {%- if message['role'] != 'tool' -%} {%- set ns.prev_message_type = None -%} {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%} {#- Detect continuation using tracked state — O(1) instead of O(n) backward scan -#} {%- set continue_same_model_turn = (role == 'model' and ns.prev_non_tool_role == 'assistant') -%} {%- if not continue_same_model_turn -%} {{- '<|turn>' + role + '\n' }} {%- endif -%} {#- Render reasoning/reasoning_content as thinking channel -#} {%- set thinking_text = message.get('reasoning') or message.get('reasoning_content') -%} {%- set thinking_gate = (loop.index0 > ns_turn.last_user_idx) or (preserve_thinking and message.get('tool_calls')) -%} {%- if thinking_text and thinking_gate -%} {{- '<|channel>thought\n' + thinking_text + '\n' -}} {%- endif -%} {%- if message.get('tool_calls') -%} {%- for tool_call in message.get('tool_calls') -%} {%- set function = tool_call['function'] -%} {{- '<|tool_call>call:' + function['name'] + '{' -}} {%- if function['arguments'] is mapping -%} {%- set ns_args = namespace(found_first=false) -%} {%- for key, value in function['arguments'] | dictsort -%} {%- if ns_args.found_first %},{% endif -%} {%- set ns_args.found_first = true -%} {{- key -}}:{{- format_argument(value, escape_keys=False) -}} {%- endfor -%} {%- elif function['arguments'] is none -%} {%- else -%} {{- raise_exception( "chat_template: tool_calls[].function.arguments must be a " "JSON object (mapping), not a string. Deserialize arguments " "before passing to the template." ) -}} {%- endif -%} {{- '}' -}} {%- endfor -%} {%- set ns.prev_message_type = 'tool_call' -%} {%- endif -%} {%- set ns_tr_out = namespace(flag=false) -%} {%- if message.get('tool_responses') -%} {#- Legacy: tool_responses embedded on the assistant message (Google/Gemma native) -#} {%- for tool_response in message.get('tool_responses') -%} {{- format_tool_response_block(tool_response['name'] | default('unknown', true), tool_response['response']) -}} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endfor -%} {%- elif message.get('tool_calls') -%} {#- OpenAI Chat Completions: forward-scan consecutive role:tool messages -#} {%- set ns_tool_scan = namespace(stopped=false) -%} {%- for k in range(loop.index0 + 1, loop_messages | length) -%} {%- if ns_tool_scan.stopped -%} {%- elif loop_messages[k]['role'] != 'tool' -%} {%- set ns_tool_scan.stopped = true -%} {%- else -%} {%- set follow = loop_messages[k] -%} {#- Resolve tool_call_id to function name -#} {%- set ns_tname = namespace(name=follow.get('name') or 'unknown') -%} {%- for tc in message.get('tool_calls') -%} {%- if tc.get('id') == follow.get('tool_call_id') -%} {%- set ns_tname.name = tc['function']['name'] -%} {%- endif -%} {%- endfor -%} {#- Handle content as string or content-parts array -#} {%- set tool_body = follow.get('content') -%} {%- if tool_body is string -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- elif tool_body is sequence and tool_body is not string -%} {%- set ns_txt = namespace(s='') -%} {%- for part in tool_body -%} {%- if part.get('type') == 'text' -%} {%- set ns_txt.s = ns_txt.s + (part.get('text') | default('')) -%} {%- endif -%} {%- endfor -%} {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}} {%- for part in tool_body -%} {%- if part.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif part.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif part.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- else -%} {{- format_tool_response_block(ns_tname.name, tool_body) -}} {%- endif -%} {%- set ns_tr_out.flag = true -%} {%- set ns.prev_message_type = 'tool_response' -%} {%- endif -%} {%- endfor -%} {%- endif -%} {%- set captured_content -%} {%- if message.get('content') is string -%} {%- if role == 'model' -%} {{- strip_thinking(message['content']) -}} {%- else -%} {{- message['content'] | trim -}} {%- endif -%} {%- elif message.get('content') is sequence -%} {%- for item in message['content'] -%} {%- if item.get('type') == 'text' -%} {%- if role == 'model' -%} {{- strip_thinking(item['text']) -}} {%- else -%} {{- item['text'] | trim -}} {%- endif -%} {%- elif item.get('type') in ['image', 'image_url'] -%} {{- '<|image|>' -}} {%- elif item.get('type') in ['audio', 'input_audio'] -%} {{- '<|audio|>' -}} {%- elif item.get('type') == 'video' -%} {{- '<|video|>' -}} {%- endif -%} {%- endfor -%} {%- endif -%} {%- endset -%} {{- captured_content -}} {%- set has_content = captured_content | trim | length > 0 -%} {#- Forward-scan: find next non-tool message role for continuation detection -#} {%- set next_nt = namespace(role=None, found=false) -%} {%- for j in range(loop.index0 + 1, loop_messages | length) -%} {%- if not next_nt.found -%} {%- if loop_messages[j]['role'] != 'tool' -%} {%- set next_nt.role = loop_messages[j]['role'] -%} {%- set next_nt.found = true -%} {%- endif -%} {%- endif -%} {%- endfor -%} {%- set continues_into_next = ( role == 'model' and next_nt.role == 'assistant' and (not message.get('tool_calls') or ns_tr_out.flag) ) -%} {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%} {{- '<|tool_response>' -}} {%- elif continues_into_next -%} {%- elif not (ns_tr_out.flag and not has_content and not next_nt.found) -%} {{- '\n' -}} {%- endif -%} {#- Track previous non-tool role for next iteration (avoids O(n) backward scan) -#} {%- set ns.prev_non_tool_role = message['role'] -%} {%- endif -%} {%- endfor -%} {%- if add_generation_prompt -%} {%- if ns.prev_message_type != 'tool_response' and ns.prev_message_type != 'tool_call' -%} {{- '<|turn>model\n' -}} {%- if not enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- elif ns.prev_message_type == 'tool_response' and enable_thinking -%} {{- '<|channel>thought\n' -}} {%- endif -%} {%- endif -%} WARNING:gguf.gguf_writer:Duplicated key name 'tokenizer.ggml.mask_token_id', overwriting it with new value 4 of type UINT32 INFO:gguf.gguf_writer:Writing the following files: INFO:gguf.gguf_writer:upload-gemma_4_31B_it/dflash-gemma-4-31B-it-BF16.gguf: n_tensors = 58, total_size = 3.1G Writing: 0%| | 0.00/3.07G [00:00", "", "", "", ... llama_model_loader: - kv 24: tokenizer.ggml.scores arr[f32,262144] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 25: tokenizer.ggml.token_type arr[i32,262144] = [3, 3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 26: tokenizer.ggml.merges arr[str,514906] = ["\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n", ... llama_model_loader: - kv 27: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 29: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 30: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 31: tokenizer.ggml.mask_token_id u32 = 4 llama_model_loader: - kv 32: tokenizer.chat_template str = {#\n Template: Google Gemma 4 Canonic... llama_model_loader: - kv 33: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 34: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - type f32: 22 tensors llama_model_loader: - type bf16: 36 tensors load_arch_hparams: DFlash extract_layers = [2, 13, 24, 36, 47, 58] [ 1/ 58] enc.output_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 2/ 58] fc.weight - [ 32256, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 330.75 MiB -> 175.71 MiB [ 3/ 58] output_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 4/ 58] blk.0.attn_k.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 5/ 58] blk.0.attn_k_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 6/ 58] blk.0.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 7/ 58] blk.0.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 8/ 58] blk.0.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 9/ 58] blk.0.attn_q_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 10/ 58] blk.0.attn_v.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 11/ 58] blk.0.ffn_down.weight - [ 10752, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 12/ 58] blk.0.ffn_gate.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 13/ 58] blk.0.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 14/ 58] blk.0.ffn_up.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 15/ 58] blk.1.attn_k.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 16/ 58] blk.1.attn_k_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 17/ 58] blk.1.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 18/ 58] blk.1.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 19/ 58] blk.1.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 20/ 58] blk.1.attn_q_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 21/ 58] blk.1.attn_v.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 22/ 58] blk.1.ffn_down.weight - [ 10752, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 23/ 58] blk.1.ffn_gate.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 24/ 58] blk.1.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 25/ 58] blk.1.ffn_up.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 26/ 58] blk.2.attn_k.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 27/ 58] blk.2.attn_k_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 28/ 58] blk.2.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 29/ 58] blk.2.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 30/ 58] blk.2.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 31/ 58] blk.2.attn_q_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 32/ 58] blk.2.attn_v.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 33/ 58] blk.2.ffn_down.weight - [ 10752, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 34/ 58] blk.2.ffn_gate.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 35/ 58] blk.2.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 36/ 58] blk.2.ffn_up.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 37/ 58] blk.3.attn_k.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 38/ 58] blk.3.attn_k_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 39/ 58] blk.3.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 40/ 58] blk.3.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 41/ 58] blk.3.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 42/ 58] blk.3.attn_q_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 43/ 58] blk.3.attn_v.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 44/ 58] blk.3.ffn_down.weight - [ 10752, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 45/ 58] blk.3.ffn_gate.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 46/ 58] blk.3.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 47/ 58] blk.3.ffn_up.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 48/ 58] blk.4.attn_k.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 49/ 58] blk.4.attn_k_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 50/ 58] blk.4.attn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 51/ 58] blk.4.attn_output.weight - [ 8192, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 52/ 58] blk.4.attn_q.weight - [ 5376, 8192, 1, 1], type = bf16, converting to q8_0 .. size = 84.00 MiB -> 44.62 MiB [ 53/ 58] blk.4.attn_q_norm.weight - [ 128, 1, 1, 1], type = f32, size = 0.000 MiB [ 54/ 58] blk.4.attn_v.weight - [ 5376, 1024, 1, 1], type = bf16, converting to q8_0 .. size = 10.50 MiB -> 5.58 MiB [ 55/ 58] blk.4.ffn_down.weight - [ 10752, 5376, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 56/ 58] blk.4.ffn_gate.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB [ 57/ 58] blk.4.ffn_norm.weight - [ 5376, 1, 1, 1], type = f32, size = 0.021 MiB [ 58/ 58] blk.4.ffn_up.weight - [ 5376, 10752, 1, 1], type = bf16, converting to q8_0 .. size = 110.25 MiB -> 58.57 MiB llama_model_quantize_impl: model size = 2929.75 MiB (16.00 BPW) llama_model_quantize_impl: quant size = 1556.55 MiB (8.50 BPW) llama_quantize: quantize time = 2828.89 ms llama_quantize: total time = 2828.89 ms + echo gemma-4-31B-it-BF16.gguf + echo gemma-4-31B-it-Q8_0.gguf + echo mmproj-gemma-4-31B-it-BF16.gguf + echo mmproj-gemma-4-31B-it-Q8_0.gguf + echo mtp-gemma-4-31B-it-BF16.gguf + echo mtp-gemma-4-31B-it-Q8_0.gguf + echo mtp-gemma-4-31B-it-Q4_0.gguf + echo gemma-4-31B-it-Q4_0.gguf + echo dflash-gemma-4-31B-it-BF16.gguf + echo dflash-gemma-4-31B-it-Q8_0.gguf