{ "architectures": [ "ApertusForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "torch_dtype": "bfloat16", "eos_token_id": 68, "hidden_act": "xielu", "hidden_dropout": 0.0, "hidden_size": 8192, "initializer_range": 0.02, "intermediate_size": 43008, "max_position_embeddings": 65536, "mlp_bias": false, "model_type": "apertus", "num_attention_heads": 64, "num_hidden_layers": 80, "num_key_value_heads": 8, "pad_token_id": 3, "post_norm": false, "qk_norm": true, "quantization_config": { "_load_in_4bit": true, "_load_in_8bit": false, "bnb_4bit_compute_dtype": "bfloat16", "bnb_4bit_quant_storage": "uint8", "bnb_4bit_quant_type": "nf4", "bnb_4bit_use_double_quant": true, "llm_int8_enable_fp32_cpu_offload": false, "llm_int8_has_fp16_weight": false, "llm_int8_skip_modules": [ "embed_tokens", "embedding", "lm_head", "multi_modal_projector", "merger", "modality_projection", "router", "visual", "vision_tower", "model.layers.12.self_attn", "model.layers.13.self_attn", "model.layers.78.mlp", "model.layers.62.mlp", "model.layers.3.self_attn", "model.layers.61.mlp", "model.layers.4.self_attn", "model.layers.60.mlp", "model.layers.1.self_attn", "model.layers.9.self_attn", "model.layers.8.self_attn", "model.layers.77.mlp", "model.layers.59.mlp", "model.layers.6.self_attn", "model.layers.2.self_attn", "model.layers.10.self_attn", "model.layers.7.self_attn", "model.layers.69.mlp", "model.layers.5.self_attn", "model.layers.58.mlp", "model.layers.57.mlp", "model.layers.3.mlp", "model.layers.4.mlp", "model.layers.56.mlp", "model.layers.55.mlp", "model.layers.5.mlp", "model.layers.54.mlp", "model.layers.48.mlp", "model.layers.7.mlp", "model.layers.53.mlp", "model.layers.16.mlp", "model.layers.47.mlp", "model.layers.52.mlp", "model.layers.18.mlp", "model.layers.13.mlp", "model.layers.17.mlp", "model.layers.19.mlp", "model.layers.14.mlp", "model.layers.11.mlp", "model.layers.20.mlp", "model.layers.12.mlp", "model.layers.21.mlp", "model.layers.46.mlp", "model.layers.15.mlp", "model.layers.22.mlp", "model.layers.23.mlp", "model.layers.51.mlp", "model.layers.24.mlp", "model.layers.25.mlp", "model.layers.26.mlp", "model.layers.27.mlp", "model.layers.28.mlp", "model.layers.45.mlp", "model.layers.6.mlp", "model.layers.29.mlp", "model.layers.50.mlp", "model.layers.30.mlp", "model.layers.8.mlp", "model.layers.44.mlp", "model.layers.10.mlp", "model.layers.31.mlp", "model.layers.43.mlp", "model.layers.49.mlp", "model.layers.32.mlp", "model.layers.9.mlp", "model.layers.33.mlp", "model.layers.34.mlp", "model.layers.42.mlp", "model.layers.39.mlp", "model.layers.41.mlp", "model.layers.35.mlp", "model.layers.38.mlp", "model.layers.40.mlp", "model.layers.37.mlp", "model.layers.36.mlp", "model.layers.70.mlp", "model.layers.2.mlp", "model.layers.0.mlp", "model.layers.0.self_attn", "model.layers.1.mlp", "model.layers.0.self_attn.o_proj", "model.layers.0.mlp.down_proj", "model.layers.1.self_attn.o_proj", "model.layers.2.self_attn.o_proj", "model.layers.3.self_attn.o_proj", "model.layers.4.self_attn.o_proj", "model.layers.5.self_attn.o_proj", "model.layers.6.self_attn.o_proj", "model.layers.7.self_attn.o_proj", "model.layers.8.self_attn.o_proj", "model.layers.9.self_attn.o_proj", "model.layers.10.self_attn.o_proj", "model.layers.12.self_attn.o_proj", "model.layers.18.self_attn.o_proj", "model.layers.25.self_attn.o_proj", "model.layers.27.self_attn.o_proj", "model.layers.31.self_attn.o_proj", "model.layers.32.self_attn.o_proj", "model.layers.33.self_attn.o_proj", "model.layers.35.self_attn.o_proj", "model.layers.46.self_attn.o_proj", "model.layers.51.self_attn.o_proj", "model.layers.59.self_attn.o_proj", "model.layers.68.self_attn.o_proj", "model.layers.69.self_attn.o_proj", "model.layers.70.self_attn.o_proj", "model.layers.71.self_attn.o_proj", "model.layers.72.self_attn.o_proj", "model.layers.73.self_attn.o_proj", "model.layers.76.self_attn.o_proj", "model.layers.79.self_attn.o_proj" ], "llm_int8_threshold": 6.0, "load_in_4bit": true, "load_in_8bit": false, "quant_method": "bitsandbytes" }, "rms_norm_eps": 1e-05, "rope_scaling": { "factor": 8.0, "high_freq_factor": 4.0, "low_freq_factor": 1.0, "original_max_position_embeddings": 8192, "rope_type": "llama3", "type": "llama3" }, "rope_theta": 12000000, "tie_word_embeddings": false, "transformers_version": "4.56.2", "unsloth_fixed": true, "use_cache": false, "vocab_size": 131072 }