Instructions to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16 # Run inference directly in the terminal: llama cli -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16 # Run inference directly in the terminal: llama cli -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16 # Run inference directly in the terminal: ./llama-cli -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16 # Run inference directly in the terminal: ./build/bin/llama-cli -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Use Docker
docker model run hf.co/random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
- LM Studio
- Jan
- Ollama
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with Ollama:
ollama run hf.co/random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
- Unsloth Studio
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP to start chatting
- Pi
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with Docker Model Runner:
docker model run hf.co/random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
- Lemonade
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Run and chat with the model
lemonade run user.Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP-BF16
List all available models
lemonade list
- Hermes Agent
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-No-MTP:BF16" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
| base_model: wangzhang/Qwen3.5-122B-A10B-abliterix | |
| base_model_relation: quantized | |
| license: apache-2.0 | |
| license_link: https://huggingface.co/Qwen/Qwen3.5-122B-A10B/blob/main/LICENSE | |
| > **Disclaimer:** Decensored model. | |
| > * Safe input → Safe output *(rare exceptions may apply)* | |
| > * Harmful input → Harmful output | |
| > * Best for testing; avoid public-facing applications. | |
| > **Note:** | |
| > * The internal MTP layer was ditched via `--no-mtp`. ~The uploader was *too lazy* to patch the conversion bugs.~ | |
| > * MTP-grafted variant available: [random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-MTP](https://huggingface.co/random257/Qwen3.5-122B-A10B-Wangzhang-Abliterix-MTP) | |
| > * **Experimental conversions**; minor tensor misalignments, loading issues, or long-context quirks may occur. Use at your own risk. | |
| **BF16**: Directly converted from [wangzhang/Qwen3.5-122B-A10B-abliterix](https://huggingface.co/wangzhang/Qwen3.5-122B-A10B-abliterix) to GGUF (without MTP), split into multiple files, with updated `general.name`. | |
| **MQ68**: Mixed precision (Q6_K - Q8_0). Uses **Q6_K for large tensors** in expert blocks; **BF16** for the output tensor; the rest remains at **Q8_0**. Inference speed is **0.8x** that of pure Q6_K. | |
| **MQ516_special**: Mixed precision (Q5_K - BF16). Uses **Q5_K** only for *large tensors* in *early SSM* blocks, **Q6_K** for large tensors in *other expert blocks*, and retains **BF16** for all *small-to-medium and critical* tensors (including the output projection). Since all medium-sized mandatory (dense) tensors are BF16, inference speed drops to **0.55x** that of pure Q6_K. *Experimental; the strategy is purely heuristic and unverified.* | |
| Both MQ68 and MQ516_special can squeeze into a gfx1151-128G device. However on Windows, RAM offloading may cause further slowdowns compared to pure Q6_K (model size >96G, \~107G at 262k context), while also increasing OOM risks that require careful management; unlike pure Q6_K (\~104G at 262k context) which leaves sufficient margin and thus remains stable. | |
| **BF16** | |
| <details> | |
| <summary>convert_hf_to_gguf params and gguf-dump</summary> | |
| --outtype bf16 | |
| --model-name "Qwen3.5 122B A10B Wangzhang Abliterix No MTP" | |
| --split-max-size 42G | |
| --no-mtp | |
| 1: UINT32 | 1 | GGUF.version = 3 | |
| 2: UINT64 | 1 | GGUF.tensor_count = 879 | |
| 3: UINT64 | 1 | GGUF.kv_count = 46 | |
| 4: STRING | 1 | general.architecture = 'qwen35moe' | |
| 5: STRING | 1 | general.type = 'model' | |
| 6: INT32 | 1 | general.sampling.top_k = 20 | |
| 7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071 | |
| 8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579 | |
| 9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Wangzhang Abliterix No MTP' | |
| 10: STRING | 1 | general.size_label = '256x6.6B' | |
| 11: UINT32 | 1 | general.base_model.count = 1 | |
| 12: STRING | 1 | general.base_model.0.name = 'Qwen3.5 122B A10B' | |
| 13: STRING | 1 | general.base_model.0.organization = 'Qwen' | |
| 14: STRING | 1 | general.base_model.0.repo_url = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B' | |
| 15: [STRING] | 5 | general.tags = ['abliterix', 'uncensored', 'decensored', 'abliterated', 'text-generation'] | |
| 16: UINT32 | 1 | qwen35moe.block_count = 48 | |
| 17: UINT32 | 1 | qwen35moe.context_length = 262144 | |
| 18: UINT32 | 1 | qwen35moe.embedding_length = 3072 | |
| 19: UINT32 | 1 | qwen35moe.attention.head_count = 32 | |
| 20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2 | |
| 21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0] | |
| 22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0 | |
| 23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07 | |
| 24: UINT32 | 1 | qwen35moe.expert_count = 256 | |
| 25: UINT32 | 1 | qwen35moe.expert_used_count = 8 | |
| 26: UINT32 | 1 | qwen35moe.attention.key_length = 256 | |
| 27: UINT32 | 1 | qwen35moe.attention.value_length = 256 | |
| 28: UINT32 | 1 | general.file_type = 32 | |
| 29: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024 | |
| 30: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024 | |
| 31: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4 | |
| 32: UINT32 | 1 | qwen35moe.ssm.state_size = 128 | |
| 33: UINT32 | 1 | qwen35moe.ssm.group_count = 16 | |
| 34: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64 | |
| 35: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192 | |
| 36: UINT32 | 1 | qwen35moe.full_attention_interval = 4 | |
| 37: UINT32 | 1 | qwen35moe.rope.dimension_count = 64 | |
| 38: UINT32 | 1 | general.quantization_version = 2 | |
| 39: STRING | 1 | tokenizer.ggml.model = 'gpt2' | |
| 40: STRING | 1 | tokenizer.ggml.pre = 'qwen35' | |
| 41: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...] | |
| 42: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...] | |
| 43: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...] | |
| 44: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046 | |
| 45: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044 | |
| 46: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...' | |
| 47: UINT16 | 1 | split.no = 0 | |
| 48: INT32 | 1 | split.tensors.count = 879 | |
| 49: UINT16 | 1 | split.count = 0 | |
| 1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight | |
| 2: 762839040 | 3072, 248320, 1, 1 | BF16 | token_embd.weight | |
| 3: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight | |
| 4: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a | |
| 5: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight | |
| 6: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias | |
| 7: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_alpha.weight | |
| 8: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_beta.weight | |
| 9: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.0.attn_qkv.weight | |
| 10: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.0.attn_gate.weight | |
| 11: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight | |
| 12: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.0.ssm_out.weight | |
| 13: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.0.ffn_down_exps.weight | |
| 14: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.0.ffn_gate_exps.weight | |
| 15: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.0.ffn_up_exps.weight | |
| 16: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight | |
| 17: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.0.ffn_down_shexp.weight | |
| 18: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_gate_shexp.weight | |
| 19: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_up_shexp.weight | |
| 20: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight | |
| 21: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight | |
| 22: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight | |
| 23: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a | |
| 24: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight | |
| 25: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias | |
| 26: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_alpha.weight | |
| 27: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_beta.weight | |
| 28: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.1.attn_qkv.weight | |
| 29: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.1.attn_gate.weight | |
| 30: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight | |
| 31: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.1.ssm_out.weight | |
| 32: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.1.ffn_down_exps.weight | |
| 33: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.1.ffn_gate_exps.weight | |
| 34: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.1.ffn_up_exps.weight | |
| 35: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight | |
| 36: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.1.ffn_down_shexp.weight | |
| 37: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_gate_shexp.weight | |
| 38: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_up_shexp.weight | |
| 39: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight | |
| 40: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight | |
| 41: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight | |
| 42: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a | |
| 43: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight | |
| 44: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias | |
| 45: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_alpha.weight | |
| 46: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_beta.weight | |
| 47: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.2.attn_qkv.weight | |
| 48: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.2.attn_gate.weight | |
| 49: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight | |
| 50: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.2.ssm_out.weight | |
| 51: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.2.ffn_down_exps.weight | |
| 52: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.2.ffn_gate_exps.weight | |
| 53: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.2.ffn_up_exps.weight | |
| 54: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight | |
| 55: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.2.ffn_down_shexp.weight | |
| 56: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_gate_shexp.weight | |
| 57: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_up_shexp.weight | |
| 58: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight | |
| 59: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight | |
| 60: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight | |
| 61: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.3.ffn_down_exps.weight | |
| 62: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.3.ffn_gate_exps.weight | |
| 63: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.3.ffn_up_exps.weight | |
| 64: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight | |
| 65: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.3.ffn_down_shexp.weight | |
| 66: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_gate_shexp.weight | |
| 67: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_up_shexp.weight | |
| 68: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight | |
| 69: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight | |
| 70: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight | |
| 71: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_k.weight | |
| 72: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.3.attn_output.weight | |
| 73: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight | |
| 74: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.3.attn_q.weight | |
| 75: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_v.weight | |
| 76: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight | |
| 77: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a | |
| 78: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight | |
| 79: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias | |
| 80: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_alpha.weight | |
| 81: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_beta.weight | |
| 82: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.4.attn_qkv.weight | |
| 83: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.4.attn_gate.weight | |
| 84: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight | |
| 85: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.4.ssm_out.weight | |
| 86: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.4.ffn_down_exps.weight | |
| 87: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.4.ffn_gate_exps.weight | |
| 88: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.4.ffn_up_exps.weight | |
| 89: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight | |
| 90: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.4.ffn_down_shexp.weight | |
| 91: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_gate_shexp.weight | |
| 92: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_up_shexp.weight | |
| 93: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight | |
| 94: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight | |
| 95: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight | |
| 96: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a | |
| 97: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight | |
| 98: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias | |
| 99: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_alpha.weight | |
| 100: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_beta.weight | |
| 101: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.5.attn_qkv.weight | |
| 102: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.5.attn_gate.weight | |
| 103: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight | |
| 104: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.5.ssm_out.weight | |
| 105: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.5.ffn_down_exps.weight | |
| 106: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.5.ffn_gate_exps.weight | |
| 107: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.5.ffn_up_exps.weight | |
| 108: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight | |
| 109: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.5.ffn_down_shexp.weight | |
| 110: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_gate_shexp.weight | |
| 111: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_up_shexp.weight | |
| 112: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight | |
| 113: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight | |
| 114: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight | |
| 115: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a | |
| 116: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight | |
| 117: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias | |
| 118: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_alpha.weight | |
| 119: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_beta.weight | |
| 120: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.6.attn_qkv.weight | |
| 121: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.6.attn_gate.weight | |
| 122: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight | |
| 123: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.6.ssm_out.weight | |
| 124: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.6.ffn_down_exps.weight | |
| 125: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.6.ffn_gate_exps.weight | |
| 126: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.6.ffn_up_exps.weight | |
| 127: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight | |
| 128: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.6.ffn_down_shexp.weight | |
| 129: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_gate_shexp.weight | |
| 130: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_up_shexp.weight | |
| 131: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight | |
| 132: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight | |
| 133: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight | |
| 134: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.7.ffn_down_exps.weight | |
| 135: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.7.ffn_gate_exps.weight | |
| 136: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.7.ffn_up_exps.weight | |
| 137: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight | |
| 138: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.7.ffn_down_shexp.weight | |
| 139: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_gate_shexp.weight | |
| 140: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_up_shexp.weight | |
| 141: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight | |
| 142: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight | |
| 143: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight | |
| 144: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_k.weight | |
| 145: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.7.attn_output.weight | |
| 146: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight | |
| 147: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.7.attn_q.weight | |
| 148: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_v.weight | |
| 149: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight | |
| 150: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a | |
| 151: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight | |
| 152: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias | |
| 153: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_alpha.weight | |
| 154: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_beta.weight | |
| 155: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.8.attn_qkv.weight | |
| 156: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.8.attn_gate.weight | |
| 157: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight | |
| 158: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.8.ssm_out.weight | |
| 159: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.8.ffn_down_exps.weight | |
| 160: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.8.ffn_gate_exps.weight | |
| 161: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.8.ffn_up_exps.weight | |
| 162: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight | |
| 163: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.8.ffn_down_shexp.weight | |
| 164: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_gate_shexp.weight | |
| 165: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_up_shexp.weight | |
| 166: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight | |
| 167: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight | |
| 168: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight | |
| 169: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a | |
| 170: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight | |
| 171: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias | |
| 172: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_alpha.weight | |
| 173: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_beta.weight | |
| 174: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.9.attn_qkv.weight | |
| 175: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.9.attn_gate.weight | |
| 176: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight | |
| 177: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.9.ssm_out.weight | |
| 178: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight | |
| 179: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a | |
| 180: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight | |
| 181: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias | |
| 182: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_alpha.weight | |
| 183: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_beta.weight | |
| 184: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.10.attn_qkv.weight | |
| 185: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.10.attn_gate.weight | |
| 186: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight | |
| 187: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.10.ssm_out.weight | |
| 188: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.10.ffn_down_exps.weight | |
| 189: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.10.ffn_gate_exps.weight | |
| 190: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.10.ffn_up_exps.weight | |
| 191: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight | |
| 192: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.10.ffn_down_shexp.weight | |
| 193: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_gate_shexp.weight | |
| 194: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_up_shexp.weight | |
| 195: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight | |
| 196: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight | |
| 197: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight | |
| 198: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.11.ffn_down_exps.weight | |
| 199: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.11.ffn_gate_exps.weight | |
| 200: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.11.ffn_up_exps.weight | |
| 201: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight | |
| 202: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.11.ffn_down_shexp.weight | |
| 203: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_gate_shexp.weight | |
| 204: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_up_shexp.weight | |
| 205: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight | |
| 206: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight | |
| 207: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight | |
| 208: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_k.weight | |
| 209: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.11.attn_output.weight | |
| 210: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight | |
| 211: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.11.attn_q.weight | |
| 212: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_v.weight | |
| 213: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight | |
| 214: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a | |
| 215: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight | |
| 216: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias | |
| 217: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_alpha.weight | |
| 218: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_beta.weight | |
| 219: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.12.attn_qkv.weight | |
| 220: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.12.attn_gate.weight | |
| 221: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight | |
| 222: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.12.ssm_out.weight | |
| 223: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.12.ffn_down_exps.weight | |
| 224: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.12.ffn_gate_exps.weight | |
| 225: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.12.ffn_up_exps.weight | |
| 226: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight | |
| 227: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.12.ffn_down_shexp.weight | |
| 228: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_gate_shexp.weight | |
| 229: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_up_shexp.weight | |
| 230: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight | |
| 231: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight | |
| 232: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight | |
| 233: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a | |
| 234: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight | |
| 235: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias | |
| 236: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_alpha.weight | |
| 237: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_beta.weight | |
| 238: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.13.attn_qkv.weight | |
| 239: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.13.attn_gate.weight | |
| 240: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight | |
| 241: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.13.ssm_out.weight | |
| 242: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.13.ffn_down_exps.weight | |
| 243: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.13.ffn_gate_exps.weight | |
| 244: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.13.ffn_up_exps.weight | |
| 245: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight | |
| 246: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.13.ffn_down_shexp.weight | |
| 247: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_gate_shexp.weight | |
| 248: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_up_shexp.weight | |
| 249: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight | |
| 250: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight | |
| 251: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight | |
| 252: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a | |
| 253: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight | |
| 254: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias | |
| 255: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_alpha.weight | |
| 256: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_beta.weight | |
| 257: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.14.attn_qkv.weight | |
| 258: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.14.attn_gate.weight | |
| 259: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight | |
| 260: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.14.ssm_out.weight | |
| 261: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.14.ffn_down_exps.weight | |
| 262: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.14.ffn_gate_exps.weight | |
| 263: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.14.ffn_up_exps.weight | |
| 264: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight | |
| 265: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.14.ffn_down_shexp.weight | |
| 266: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_gate_shexp.weight | |
| 267: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_up_shexp.weight | |
| 268: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight | |
| 269: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight | |
| 270: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight | |
| 271: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.15.ffn_down_exps.weight | |
| 272: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.15.ffn_gate_exps.weight | |
| 273: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.15.ffn_up_exps.weight | |
| 274: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight | |
| 275: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.15.ffn_down_shexp.weight | |
| 276: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_gate_shexp.weight | |
| 277: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_up_shexp.weight | |
| 278: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight | |
| 279: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight | |
| 280: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight | |
| 281: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_k.weight | |
| 282: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.15.attn_output.weight | |
| 283: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight | |
| 284: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.15.attn_q.weight | |
| 285: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_v.weight | |
| 286: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight | |
| 287: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a | |
| 288: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight | |
| 289: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias | |
| 290: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_alpha.weight | |
| 291: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_beta.weight | |
| 292: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.16.attn_qkv.weight | |
| 293: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.16.attn_gate.weight | |
| 294: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight | |
| 295: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.16.ssm_out.weight | |
| 296: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.16.ffn_down_exps.weight | |
| 297: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.16.ffn_gate_exps.weight | |
| 298: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.16.ffn_up_exps.weight | |
| 299: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight | |
| 300: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.16.ffn_down_shexp.weight | |
| 301: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_gate_shexp.weight | |
| 302: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_up_shexp.weight | |
| 303: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight | |
| 304: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight | |
| 305: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight | |
| 306: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a | |
| 307: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight | |
| 308: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias | |
| 309: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_alpha.weight | |
| 310: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_beta.weight | |
| 311: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.17.attn_qkv.weight | |
| 312: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.17.attn_gate.weight | |
| 313: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight | |
| 314: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.17.ssm_out.weight | |
| 315: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.17.ffn_down_exps.weight | |
| 316: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.17.ffn_gate_exps.weight | |
| 317: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.17.ffn_up_exps.weight | |
| 318: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight | |
| 319: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.17.ffn_down_shexp.weight | |
| 320: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_gate_shexp.weight | |
| 321: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_up_shexp.weight | |
| 322: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight | |
| 323: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight | |
| 324: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight | |
| 325: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a | |
| 326: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight | |
| 327: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias | |
| 328: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_alpha.weight | |
| 329: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_beta.weight | |
| 330: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.18.attn_qkv.weight | |
| 331: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.18.attn_gate.weight | |
| 332: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight | |
| 333: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.18.ssm_out.weight | |
| 334: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.9.ffn_down_exps.weight | |
| 335: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.9.ffn_gate_exps.weight | |
| 336: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.9.ffn_up_exps.weight | |
| 337: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight | |
| 338: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.9.ffn_down_shexp.weight | |
| 339: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_gate_shexp.weight | |
| 340: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_up_shexp.weight | |
| 341: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight | |
| 342: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight | |
| 343: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.18.ffn_down_exps.weight | |
| 344: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.18.ffn_gate_exps.weight | |
| 345: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.18.ffn_up_exps.weight | |
| 346: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight | |
| 347: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.18.ffn_down_shexp.weight | |
| 348: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_gate_shexp.weight | |
| 349: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_up_shexp.weight | |
| 350: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight | |
| 351: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight | |
| 352: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight | |
| 353: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.19.ffn_down_exps.weight | |
| 354: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.19.ffn_gate_exps.weight | |
| 355: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.19.ffn_up_exps.weight | |
| 356: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight | |
| 357: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.19.ffn_down_shexp.weight | |
| 358: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_gate_shexp.weight | |
| 359: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_up_shexp.weight | |
| 360: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight | |
| 361: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight | |
| 362: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight | |
| 363: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_k.weight | |
| 364: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.19.attn_output.weight | |
| 365: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight | |
| 366: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.19.attn_q.weight | |
| 367: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_v.weight | |
| 368: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight | |
| 369: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a | |
| 370: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight | |
| 371: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias | |
| 372: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_alpha.weight | |
| 373: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_beta.weight | |
| 374: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.20.attn_qkv.weight | |
| 375: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.20.attn_gate.weight | |
| 376: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight | |
| 377: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.20.ssm_out.weight | |
| 378: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.20.ffn_down_exps.weight | |
| 379: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.20.ffn_gate_exps.weight | |
| 380: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.20.ffn_up_exps.weight | |
| 381: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight | |
| 382: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.20.ffn_down_shexp.weight | |
| 383: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_gate_shexp.weight | |
| 384: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_up_shexp.weight | |
| 385: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight | |
| 386: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight | |
| 387: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight | |
| 388: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a | |
| 389: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight | |
| 390: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias | |
| 391: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_alpha.weight | |
| 392: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_beta.weight | |
| 393: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.21.attn_qkv.weight | |
| 394: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.21.attn_gate.weight | |
| 395: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight | |
| 396: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.21.ssm_out.weight | |
| 397: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.21.ffn_down_exps.weight | |
| 398: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.21.ffn_gate_exps.weight | |
| 399: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.21.ffn_up_exps.weight | |
| 400: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight | |
| 401: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.21.ffn_down_shexp.weight | |
| 402: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_gate_shexp.weight | |
| 403: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_up_shexp.weight | |
| 404: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight | |
| 405: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight | |
| 406: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight | |
| 407: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a | |
| 408: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight | |
| 409: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias | |
| 410: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_alpha.weight | |
| 411: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_beta.weight | |
| 412: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.22.attn_qkv.weight | |
| 413: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.22.attn_gate.weight | |
| 414: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight | |
| 415: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.22.ssm_out.weight | |
| 416: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.22.ffn_down_exps.weight | |
| 417: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.22.ffn_gate_exps.weight | |
| 418: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.22.ffn_up_exps.weight | |
| 419: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight | |
| 420: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.22.ffn_down_shexp.weight | |
| 421: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_gate_shexp.weight | |
| 422: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_up_shexp.weight | |
| 423: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight | |
| 424: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight | |
| 425: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight | |
| 426: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.23.ffn_down_exps.weight | |
| 427: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.23.ffn_gate_exps.weight | |
| 428: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.23.ffn_up_exps.weight | |
| 429: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight | |
| 430: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.23.ffn_down_shexp.weight | |
| 431: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_gate_shexp.weight | |
| 432: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_up_shexp.weight | |
| 433: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight | |
| 434: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight | |
| 435: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight | |
| 436: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_k.weight | |
| 437: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.23.attn_output.weight | |
| 438: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight | |
| 439: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.23.attn_q.weight | |
| 440: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_v.weight | |
| 441: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight | |
| 442: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a | |
| 443: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight | |
| 444: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias | |
| 445: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_alpha.weight | |
| 446: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_beta.weight | |
| 447: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.24.attn_qkv.weight | |
| 448: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.24.attn_gate.weight | |
| 449: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight | |
| 450: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.24.ssm_out.weight | |
| 451: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.24.ffn_down_exps.weight | |
| 452: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.24.ffn_gate_exps.weight | |
| 453: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.24.ffn_up_exps.weight | |
| 454: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight | |
| 455: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.24.ffn_down_shexp.weight | |
| 456: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_gate_shexp.weight | |
| 457: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_up_shexp.weight | |
| 458: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight | |
| 459: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight | |
| 460: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight | |
| 461: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a | |
| 462: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight | |
| 463: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias | |
| 464: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_alpha.weight | |
| 465: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_beta.weight | |
| 466: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.25.attn_qkv.weight | |
| 467: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.25.attn_gate.weight | |
| 468: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight | |
| 469: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.25.ssm_out.weight | |
| 470: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.25.ffn_down_exps.weight | |
| 471: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.25.ffn_gate_exps.weight | |
| 472: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.25.ffn_up_exps.weight | |
| 473: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight | |
| 474: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.25.ffn_down_shexp.weight | |
| 475: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_gate_shexp.weight | |
| 476: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_up_shexp.weight | |
| 477: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight | |
| 478: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight | |
| 479: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight | |
| 480: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a | |
| 481: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight | |
| 482: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias | |
| 483: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_alpha.weight | |
| 484: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_beta.weight | |
| 485: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.26.attn_qkv.weight | |
| 486: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.26.attn_gate.weight | |
| 487: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight | |
| 488: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.26.ssm_out.weight | |
| 489: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.26.ffn_down_exps.weight | |
| 490: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.26.ffn_gate_exps.weight | |
| 491: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.26.ffn_up_exps.weight | |
| 492: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight | |
| 493: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.26.ffn_down_shexp.weight | |
| 494: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_gate_shexp.weight | |
| 495: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_up_shexp.weight | |
| 496: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight | |
| 497: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight | |
| 498: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight | |
| 499: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.27.ffn_down_exps.weight | |
| 500: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.27.ffn_gate_exps.weight | |
| 501: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.27.ffn_up_exps.weight | |
| 502: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight | |
| 503: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.27.ffn_down_shexp.weight | |
| 504: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_gate_shexp.weight | |
| 505: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_up_shexp.weight | |
| 506: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight | |
| 507: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight | |
| 508: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight | |
| 509: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_k.weight | |
| 510: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.27.attn_output.weight | |
| 511: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight | |
| 512: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.27.attn_q.weight | |
| 513: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_v.weight | |
| 514: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight | |
| 515: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a | |
| 516: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight | |
| 517: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias | |
| 518: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_alpha.weight | |
| 519: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_beta.weight | |
| 520: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.28.attn_qkv.weight | |
| 521: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.28.attn_gate.weight | |
| 522: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight | |
| 523: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.28.ssm_out.weight | |
| 524: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.28.ffn_down_exps.weight | |
| 525: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.28.ffn_gate_exps.weight | |
| 526: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.28.ffn_up_exps.weight | |
| 527: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight | |
| 528: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.28.ffn_down_shexp.weight | |
| 529: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_gate_shexp.weight | |
| 530: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_up_shexp.weight | |
| 531: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight | |
| 532: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight | |
| 533: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight | |
| 534: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a | |
| 535: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight | |
| 536: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias | |
| 537: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_alpha.weight | |
| 538: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_beta.weight | |
| 539: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.29.attn_qkv.weight | |
| 540: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.29.attn_gate.weight | |
| 541: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight | |
| 542: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.29.ssm_out.weight | |
| 543: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.29.ffn_down_exps.weight | |
| 544: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.29.ffn_gate_exps.weight | |
| 545: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.29.ffn_up_exps.weight | |
| 546: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight | |
| 547: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.29.ffn_down_shexp.weight | |
| 548: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_gate_shexp.weight | |
| 549: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_up_shexp.weight | |
| 550: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight | |
| 551: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight | |
| 552: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight | |
| 553: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a | |
| 554: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight | |
| 555: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias | |
| 556: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_alpha.weight | |
| 557: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_beta.weight | |
| 558: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.30.attn_qkv.weight | |
| 559: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.30.attn_gate.weight | |
| 560: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight | |
| 561: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.30.ssm_out.weight | |
| 562: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.30.ffn_down_exps.weight | |
| 563: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.30.ffn_gate_exps.weight | |
| 564: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.30.ffn_up_exps.weight | |
| 565: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight | |
| 566: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.30.ffn_down_shexp.weight | |
| 567: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_gate_shexp.weight | |
| 568: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_up_shexp.weight | |
| 569: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight | |
| 570: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight | |
| 571: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight | |
| 572: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.31.ffn_down_exps.weight | |
| 573: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.31.ffn_gate_exps.weight | |
| 574: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.31.ffn_up_exps.weight | |
| 575: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight | |
| 576: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.31.ffn_down_shexp.weight | |
| 577: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_gate_shexp.weight | |
| 578: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_up_shexp.weight | |
| 579: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight | |
| 580: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight | |
| 581: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight | |
| 582: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_k.weight | |
| 583: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.31.attn_output.weight | |
| 584: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight | |
| 585: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.31.attn_q.weight | |
| 586: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_v.weight | |
| 587: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight | |
| 588: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a | |
| 589: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight | |
| 590: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias | |
| 591: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_alpha.weight | |
| 592: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_beta.weight | |
| 593: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.32.attn_qkv.weight | |
| 594: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.32.attn_gate.weight | |
| 595: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight | |
| 596: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.32.ssm_out.weight | |
| 597: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.32.ffn_down_exps.weight | |
| 598: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.32.ffn_gate_exps.weight | |
| 599: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.32.ffn_up_exps.weight | |
| 600: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight | |
| 601: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.32.ffn_down_shexp.weight | |
| 602: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_gate_shexp.weight | |
| 603: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_up_shexp.weight | |
| 604: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight | |
| 605: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight | |
| 606: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight | |
| 607: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a | |
| 608: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight | |
| 609: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias | |
| 610: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_alpha.weight | |
| 611: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_beta.weight | |
| 612: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.33.attn_qkv.weight | |
| 613: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.33.attn_gate.weight | |
| 614: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight | |
| 615: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.33.ssm_out.weight | |
| 616: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.33.ffn_down_exps.weight | |
| 617: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.33.ffn_gate_exps.weight | |
| 618: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.33.ffn_up_exps.weight | |
| 619: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight | |
| 620: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.33.ffn_down_shexp.weight | |
| 621: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_gate_shexp.weight | |
| 622: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_up_shexp.weight | |
| 623: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight | |
| 624: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight | |
| 625: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight | |
| 626: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a | |
| 627: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight | |
| 628: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias | |
| 629: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_alpha.weight | |
| 630: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_beta.weight | |
| 631: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.34.attn_qkv.weight | |
| 632: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.34.attn_gate.weight | |
| 633: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight | |
| 634: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.34.ssm_out.weight | |
| 635: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.34.ffn_down_exps.weight | |
| 636: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.34.ffn_gate_exps.weight | |
| 637: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.34.ffn_up_exps.weight | |
| 638: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight | |
| 639: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.34.ffn_down_shexp.weight | |
| 640: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_gate_shexp.weight | |
| 641: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_up_shexp.weight | |
| 642: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight | |
| 643: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight | |
| 644: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight | |
| 645: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.35.ffn_down_exps.weight | |
| 646: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.35.ffn_gate_exps.weight | |
| 647: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.35.ffn_up_exps.weight | |
| 648: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight | |
| 649: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.35.ffn_down_shexp.weight | |
| 650: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_gate_shexp.weight | |
| 651: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_up_shexp.weight | |
| 652: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight | |
| 653: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight | |
| 654: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight | |
| 655: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_k.weight | |
| 656: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.35.attn_output.weight | |
| 657: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight | |
| 658: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.35.attn_q.weight | |
| 659: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_v.weight | |
| 660: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight | |
| 661: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a | |
| 662: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight | |
| 663: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias | |
| 664: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_alpha.weight | |
| 665: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_beta.weight | |
| 666: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.36.attn_qkv.weight | |
| 667: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.36.attn_gate.weight | |
| 668: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight | |
| 669: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.36.ssm_out.weight | |
| 670: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.36.ffn_down_exps.weight | |
| 671: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.36.ffn_gate_exps.weight | |
| 672: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.36.ffn_up_exps.weight | |
| 673: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight | |
| 674: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.36.ffn_down_shexp.weight | |
| 675: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_gate_shexp.weight | |
| 676: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_up_shexp.weight | |
| 677: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight | |
| 678: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight | |
| 679: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight | |
| 680: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a | |
| 681: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight | |
| 682: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias | |
| 683: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_alpha.weight | |
| 684: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_beta.weight | |
| 685: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.37.attn_qkv.weight | |
| 686: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.37.attn_gate.weight | |
| 687: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight | |
| 688: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.37.ssm_out.weight | |
| 689: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.37.ffn_down_exps.weight | |
| 690: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.37.ffn_gate_exps.weight | |
| 691: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.37.ffn_up_exps.weight | |
| 692: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight | |
| 693: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.37.ffn_down_shexp.weight | |
| 694: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_gate_shexp.weight | |
| 695: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_up_shexp.weight | |
| 696: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight | |
| 697: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight | |
| 698: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight | |
| 699: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a | |
| 700: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight | |
| 701: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias | |
| 702: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_alpha.weight | |
| 703: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_beta.weight | |
| 704: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.38.attn_qkv.weight | |
| 705: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.38.attn_gate.weight | |
| 706: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight | |
| 707: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.38.ssm_out.weight | |
| 708: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.38.ffn_down_exps.weight | |
| 709: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.38.ffn_gate_exps.weight | |
| 710: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.38.ffn_up_exps.weight | |
| 711: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight | |
| 712: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.38.ffn_down_shexp.weight | |
| 713: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_gate_shexp.weight | |
| 714: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_up_shexp.weight | |
| 715: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight | |
| 716: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight | |
| 717: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight | |
| 718: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.39.ffn_down_exps.weight | |
| 719: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.39.ffn_gate_exps.weight | |
| 720: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.39.ffn_up_exps.weight | |
| 721: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight | |
| 722: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.39.ffn_down_shexp.weight | |
| 723: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_gate_shexp.weight | |
| 724: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_up_shexp.weight | |
| 725: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight | |
| 726: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight | |
| 727: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight | |
| 728: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_k.weight | |
| 729: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.39.attn_output.weight | |
| 730: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight | |
| 731: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.39.attn_q.weight | |
| 732: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_v.weight | |
| 733: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight | |
| 734: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a | |
| 735: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight | |
| 736: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias | |
| 737: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_alpha.weight | |
| 738: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_beta.weight | |
| 739: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.40.attn_qkv.weight | |
| 740: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.40.attn_gate.weight | |
| 741: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight | |
| 742: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.40.ssm_out.weight | |
| 743: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.40.ffn_down_exps.weight | |
| 744: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.40.ffn_gate_exps.weight | |
| 745: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.40.ffn_up_exps.weight | |
| 746: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight | |
| 747: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.40.ffn_down_shexp.weight | |
| 748: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_gate_shexp.weight | |
| 749: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_up_shexp.weight | |
| 750: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight | |
| 751: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight | |
| 752: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight | |
| 753: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a | |
| 754: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight | |
| 755: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias | |
| 756: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_alpha.weight | |
| 757: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_beta.weight | |
| 758: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.41.attn_qkv.weight | |
| 759: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.41.attn_gate.weight | |
| 760: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight | |
| 761: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.41.ssm_out.weight | |
| 762: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.41.ffn_down_exps.weight | |
| 763: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.41.ffn_gate_exps.weight | |
| 764: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.41.ffn_up_exps.weight | |
| 765: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight | |
| 766: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.41.ffn_down_shexp.weight | |
| 767: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_gate_shexp.weight | |
| 768: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_up_shexp.weight | |
| 769: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight | |
| 770: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight | |
| 771: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight | |
| 772: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a | |
| 773: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight | |
| 774: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias | |
| 775: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_alpha.weight | |
| 776: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_beta.weight | |
| 777: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.42.attn_qkv.weight | |
| 778: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.42.attn_gate.weight | |
| 779: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight | |
| 780: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.42.ssm_out.weight | |
| 781: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.42.ffn_down_exps.weight | |
| 782: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.42.ffn_gate_exps.weight | |
| 783: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.42.ffn_up_exps.weight | |
| 784: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight | |
| 785: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.42.ffn_down_shexp.weight | |
| 786: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_gate_shexp.weight | |
| 787: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_up_shexp.weight | |
| 788: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight | |
| 789: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight | |
| 790: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight | |
| 791: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.43.ffn_down_exps.weight | |
| 792: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.43.ffn_gate_exps.weight | |
| 793: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.43.ffn_up_exps.weight | |
| 794: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight | |
| 795: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.43.ffn_down_shexp.weight | |
| 796: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_gate_shexp.weight | |
| 797: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_up_shexp.weight | |
| 798: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight | |
| 799: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight | |
| 800: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight | |
| 801: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_k.weight | |
| 802: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.43.attn_output.weight | |
| 803: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight | |
| 804: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.43.attn_q.weight | |
| 805: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_v.weight | |
| 806: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight | |
| 807: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a | |
| 808: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight | |
| 809: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias | |
| 810: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_alpha.weight | |
| 811: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_beta.weight | |
| 812: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.44.attn_qkv.weight | |
| 813: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.44.attn_gate.weight | |
| 814: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight | |
| 815: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.44.ssm_out.weight | |
| 816: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.44.ffn_down_exps.weight | |
| 817: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.44.ffn_gate_exps.weight | |
| 818: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.44.ffn_up_exps.weight | |
| 819: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight | |
| 820: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.44.ffn_down_shexp.weight | |
| 821: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_gate_shexp.weight | |
| 822: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_up_shexp.weight | |
| 823: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight | |
| 824: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight | |
| 825: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight | |
| 826: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a | |
| 827: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight | |
| 828: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias | |
| 829: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_alpha.weight | |
| 830: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_beta.weight | |
| 831: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.45.attn_qkv.weight | |
| 832: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.45.attn_gate.weight | |
| 833: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight | |
| 834: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.45.ssm_out.weight | |
| 835: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.45.ffn_down_exps.weight | |
| 836: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.45.ffn_gate_exps.weight | |
| 837: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.45.ffn_up_exps.weight | |
| 838: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight | |
| 839: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.45.ffn_down_shexp.weight | |
| 840: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_gate_shexp.weight | |
| 841: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_up_shexp.weight | |
| 842: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight | |
| 843: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight | |
| 844: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight | |
| 845: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a | |
| 846: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight | |
| 847: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias | |
| 848: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_alpha.weight | |
| 849: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_beta.weight | |
| 850: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.46.attn_qkv.weight | |
| 851: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.46.attn_gate.weight | |
| 852: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight | |
| 853: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.46.ssm_out.weight | |
| 854: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.46.ffn_down_exps.weight | |
| 855: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.46.ffn_gate_exps.weight | |
| 856: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.46.ffn_up_exps.weight | |
| 857: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight | |
| 858: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.46.ffn_down_shexp.weight | |
| 859: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_gate_shexp.weight | |
| 860: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_up_shexp.weight | |
| 861: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight | |
| 862: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight | |
| 863: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight | |
| 864: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.47.ffn_down_exps.weight | |
| 865: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.47.ffn_gate_exps.weight | |
| 866: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.47.ffn_up_exps.weight | |
| 867: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight | |
| 868: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.47.ffn_down_shexp.weight | |
| 869: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_gate_shexp.weight | |
| 870: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_up_shexp.weight | |
| 871: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight | |
| 872: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight | |
| 873: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight | |
| 874: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_k.weight | |
| 875: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.47.attn_output.weight | |
| 876: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight | |
| 877: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.47.attn_q.weight | |
| 878: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_v.weight | |
| 879: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight | |
| </details> | |
| **MQ68 (Mixed Quant Q6_K - Q8_0)** | |
| <details> | |
| <summary>llama-quantize params and gguf-dump</summary> | |
| --leave-output-tensor | |
| --tensor-type ".*ffn_down_exps\.weight=q6_k" | |
| --tensor-type ".*ffn_gate_exps\.weight=q6_k" | |
| --tensor-type ".*ffn_up_exps\.weight=q6_k" | |
| Q8_0 | |
| 1: UINT32 | 1 | GGUF.version = 3 | |
| 2: UINT64 | 1 | GGUF.tensor_count = 879 | |
| 3: UINT64 | 1 | GGUF.kv_count = 43 | |
| 4: STRING | 1 | general.architecture = 'qwen35moe' | |
| 5: STRING | 1 | general.type = 'model' | |
| 6: INT32 | 1 | general.sampling.top_k = 20 | |
| 7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071 | |
| 8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579 | |
| 9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Wangzhang Abliterix No MTP' | |
| 10: STRING | 1 | general.size_label = '256x6.6B' | |
| 11: UINT32 | 1 | general.base_model.count = 1 | |
| 12: STRING | 1 | general.base_model.0.name = 'Qwen3.5 122B A10B' | |
| 13: STRING | 1 | general.base_model.0.organization = 'Qwen' | |
| 14: STRING | 1 | general.base_model.0.repo_url = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B' | |
| 15: [STRING] | 5 | general.tags = ['abliterix', 'uncensored', 'decensored', 'abliterated', 'text-generation'] | |
| 16: UINT32 | 1 | qwen35moe.block_count = 48 | |
| 17: UINT32 | 1 | qwen35moe.context_length = 262144 | |
| 18: UINT32 | 1 | qwen35moe.embedding_length = 3072 | |
| 19: UINT32 | 1 | qwen35moe.attention.head_count = 32 | |
| 20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2 | |
| 21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0] | |
| 22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0 | |
| 23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07 | |
| 24: UINT32 | 1 | qwen35moe.expert_count = 256 | |
| 25: UINT32 | 1 | qwen35moe.expert_used_count = 8 | |
| 26: UINT32 | 1 | qwen35moe.attention.key_length = 256 | |
| 27: UINT32 | 1 | qwen35moe.attention.value_length = 256 | |
| 28: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024 | |
| 29: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024 | |
| 30: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4 | |
| 31: UINT32 | 1 | qwen35moe.ssm.state_size = 128 | |
| 32: UINT32 | 1 | qwen35moe.ssm.group_count = 16 | |
| 33: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64 | |
| 34: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192 | |
| 35: UINT32 | 1 | qwen35moe.full_attention_interval = 4 | |
| 36: UINT32 | 1 | qwen35moe.rope.dimension_count = 64 | |
| 37: STRING | 1 | tokenizer.ggml.model = 'gpt2' | |
| 38: STRING | 1 | tokenizer.ggml.pre = 'qwen35' | |
| 39: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...] | |
| 40: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...] | |
| 41: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...] | |
| 42: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046 | |
| 43: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044 | |
| 44: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...' | |
| 45: UINT32 | 1 | general.quantization_version = 2 | |
| 46: UINT32 | 1 | general.file_type = 7 | |
| 1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight | |
| 2: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight | |
| 3: 762839040 | 3072, 248320, 1, 1 | Q8_0 | token_embd.weight | |
| 4: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.0.attn_gate.weight | |
| 5: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight | |
| 6: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.0.attn_qkv.weight | |
| 7: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.0.ffn_down_exps.weight | |
| 8: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.0.ffn_down_shexp.weight | |
| 9: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.0.ffn_gate_exps.weight | |
| 10: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight | |
| 11: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight | |
| 12: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.0.ffn_gate_shexp.weight | |
| 13: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.0.ffn_up_exps.weight | |
| 14: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.0.ffn_up_shexp.weight | |
| 15: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight | |
| 16: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a | |
| 17: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.0.ssm_alpha.weight | |
| 18: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.0.ssm_beta.weight | |
| 19: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight | |
| 20: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias | |
| 21: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight | |
| 22: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.0.ssm_out.weight | |
| 23: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.1.attn_gate.weight | |
| 24: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight | |
| 25: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.1.attn_qkv.weight | |
| 26: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.1.ffn_down_exps.weight | |
| 27: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.1.ffn_down_shexp.weight | |
| 28: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.1.ffn_gate_exps.weight | |
| 29: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight | |
| 30: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight | |
| 31: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.1.ffn_gate_shexp.weight | |
| 32: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.1.ffn_up_exps.weight | |
| 33: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.1.ffn_up_shexp.weight | |
| 34: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight | |
| 35: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a | |
| 36: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.1.ssm_alpha.weight | |
| 37: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.1.ssm_beta.weight | |
| 38: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight | |
| 39: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias | |
| 40: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight | |
| 41: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.1.ssm_out.weight | |
| 42: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.2.attn_gate.weight | |
| 43: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight | |
| 44: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.2.attn_qkv.weight | |
| 45: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.2.ffn_down_exps.weight | |
| 46: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.2.ffn_down_shexp.weight | |
| 47: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.2.ffn_gate_exps.weight | |
| 48: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight | |
| 49: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight | |
| 50: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.2.ffn_gate_shexp.weight | |
| 51: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.2.ffn_up_exps.weight | |
| 52: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.2.ffn_up_shexp.weight | |
| 53: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight | |
| 54: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a | |
| 55: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.2.ssm_alpha.weight | |
| 56: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.2.ssm_beta.weight | |
| 57: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight | |
| 58: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias | |
| 59: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight | |
| 60: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.2.ssm_out.weight | |
| 61: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.3.attn_k.weight | |
| 62: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight | |
| 63: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight | |
| 64: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.3.attn_output.weight | |
| 65: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.3.attn_q.weight | |
| 66: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight | |
| 67: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.3.attn_v.weight | |
| 68: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.3.ffn_down_exps.weight | |
| 69: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.3.ffn_down_shexp.weight | |
| 70: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_gate_exps.weight | |
| 71: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight | |
| 72: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight | |
| 73: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.3.ffn_gate_shexp.weight | |
| 74: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_up_exps.weight | |
| 75: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.3.ffn_up_shexp.weight | |
| 76: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight | |
| 77: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.4.attn_gate.weight | |
| 78: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight | |
| 79: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.4.attn_qkv.weight | |
| 80: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.4.ffn_down_exps.weight | |
| 81: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.4.ffn_down_shexp.weight | |
| 82: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.4.ffn_gate_exps.weight | |
| 83: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight | |
| 84: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight | |
| 85: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.4.ffn_gate_shexp.weight | |
| 86: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.4.ffn_up_exps.weight | |
| 87: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.4.ffn_up_shexp.weight | |
| 88: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight | |
| 89: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a | |
| 90: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.4.ssm_alpha.weight | |
| 91: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.4.ssm_beta.weight | |
| 92: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight | |
| 93: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias | |
| 94: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight | |
| 95: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.4.ssm_out.weight | |
| 96: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.5.attn_gate.weight | |
| 97: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight | |
| 98: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.5.attn_qkv.weight | |
| 99: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.5.ffn_down_exps.weight | |
| 100: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.5.ffn_down_shexp.weight | |
| 101: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.5.ffn_gate_exps.weight | |
| 102: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight | |
| 103: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight | |
| 104: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.5.ffn_gate_shexp.weight | |
| 105: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.5.ffn_up_exps.weight | |
| 106: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.5.ffn_up_shexp.weight | |
| 107: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight | |
| 108: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a | |
| 109: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.5.ssm_alpha.weight | |
| 110: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.5.ssm_beta.weight | |
| 111: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight | |
| 112: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias | |
| 113: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight | |
| 114: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.5.ssm_out.weight | |
| 115: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.6.attn_gate.weight | |
| 116: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight | |
| 117: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.6.attn_qkv.weight | |
| 118: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.6.ffn_down_exps.weight | |
| 119: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.6.ffn_down_shexp.weight | |
| 120: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.6.ffn_gate_exps.weight | |
| 121: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight | |
| 122: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight | |
| 123: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.6.ffn_gate_shexp.weight | |
| 124: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.6.ffn_up_exps.weight | |
| 125: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.6.ffn_up_shexp.weight | |
| 126: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight | |
| 127: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a | |
| 128: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.6.ssm_alpha.weight | |
| 129: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.6.ssm_beta.weight | |
| 130: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight | |
| 131: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias | |
| 132: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight | |
| 133: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.6.ssm_out.weight | |
| 134: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.7.attn_k.weight | |
| 135: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight | |
| 136: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight | |
| 137: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.7.attn_output.weight | |
| 138: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.7.attn_q.weight | |
| 139: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight | |
| 140: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.7.attn_v.weight | |
| 141: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.7.ffn_down_exps.weight | |
| 142: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.7.ffn_down_shexp.weight | |
| 143: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_gate_exps.weight | |
| 144: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight | |
| 145: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight | |
| 146: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.7.ffn_gate_shexp.weight | |
| 147: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_up_exps.weight | |
| 148: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.7.ffn_up_shexp.weight | |
| 149: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight | |
| 150: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.8.attn_gate.weight | |
| 151: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight | |
| 152: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.8.attn_qkv.weight | |
| 153: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.8.ffn_down_exps.weight | |
| 154: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.8.ffn_down_shexp.weight | |
| 155: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.8.ffn_gate_exps.weight | |
| 156: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight | |
| 157: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight | |
| 158: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.8.ffn_gate_shexp.weight | |
| 159: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.8.ffn_up_exps.weight | |
| 160: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.8.ffn_up_shexp.weight | |
| 161: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight | |
| 162: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a | |
| 163: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.8.ssm_alpha.weight | |
| 164: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.8.ssm_beta.weight | |
| 165: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight | |
| 166: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias | |
| 167: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight | |
| 168: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.8.ssm_out.weight | |
| 169: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.9.attn_gate.weight | |
| 170: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight | |
| 171: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.9.attn_qkv.weight | |
| 172: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.9.ffn_down_exps.weight | |
| 173: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.9.ffn_down_shexp.weight | |
| 174: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.9.ffn_gate_exps.weight | |
| 175: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight | |
| 176: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight | |
| 177: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.9.ffn_gate_shexp.weight | |
| 178: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.9.ffn_up_exps.weight | |
| 179: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.9.ffn_up_shexp.weight | |
| 180: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight | |
| 181: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a | |
| 182: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.9.ssm_alpha.weight | |
| 183: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.9.ssm_beta.weight | |
| 184: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight | |
| 185: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias | |
| 186: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight | |
| 187: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.9.ssm_out.weight | |
| 188: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.10.attn_gate.weight | |
| 189: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight | |
| 190: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.10.attn_qkv.weight | |
| 191: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.10.ffn_down_exps.weight | |
| 192: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.10.ffn_down_shexp.weight | |
| 193: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.10.ffn_gate_exps.weight | |
| 194: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight | |
| 195: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight | |
| 196: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.10.ffn_gate_shexp.weight | |
| 197: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.10.ffn_up_exps.weight | |
| 198: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.10.ffn_up_shexp.weight | |
| 199: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight | |
| 200: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a | |
| 201: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.10.ssm_alpha.weight | |
| 202: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.10.ssm_beta.weight | |
| 203: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight | |
| 204: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias | |
| 205: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight | |
| 206: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.10.ssm_out.weight | |
| 207: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.11.attn_k.weight | |
| 208: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight | |
| 209: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight | |
| 210: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.11.attn_output.weight | |
| 211: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.11.attn_q.weight | |
| 212: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight | |
| 213: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.11.attn_v.weight | |
| 214: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.11.ffn_down_exps.weight | |
| 215: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.11.ffn_down_shexp.weight | |
| 216: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_gate_exps.weight | |
| 217: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight | |
| 218: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight | |
| 219: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.11.ffn_gate_shexp.weight | |
| 220: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_up_exps.weight | |
| 221: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.11.ffn_up_shexp.weight | |
| 222: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight | |
| 223: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.12.attn_gate.weight | |
| 224: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight | |
| 225: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.12.attn_qkv.weight | |
| 226: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.12.ffn_down_exps.weight | |
| 227: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.12.ffn_down_shexp.weight | |
| 228: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_gate_exps.weight | |
| 229: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight | |
| 230: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight | |
| 231: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.12.ffn_gate_shexp.weight | |
| 232: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_up_exps.weight | |
| 233: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.12.ffn_up_shexp.weight | |
| 234: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight | |
| 235: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a | |
| 236: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.12.ssm_alpha.weight | |
| 237: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.12.ssm_beta.weight | |
| 238: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight | |
| 239: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias | |
| 240: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight | |
| 241: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.12.ssm_out.weight | |
| 242: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.13.attn_gate.weight | |
| 243: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight | |
| 244: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.13.attn_qkv.weight | |
| 245: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.13.ffn_down_exps.weight | |
| 246: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.13.ffn_down_shexp.weight | |
| 247: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_gate_exps.weight | |
| 248: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight | |
| 249: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight | |
| 250: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.13.ffn_gate_shexp.weight | |
| 251: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_up_exps.weight | |
| 252: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.13.ffn_up_shexp.weight | |
| 253: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight | |
| 254: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a | |
| 255: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.13.ssm_alpha.weight | |
| 256: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.13.ssm_beta.weight | |
| 257: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight | |
| 258: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias | |
| 259: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight | |
| 260: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.13.ssm_out.weight | |
| 261: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.14.attn_gate.weight | |
| 262: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight | |
| 263: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.14.attn_qkv.weight | |
| 264: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.14.ffn_down_exps.weight | |
| 265: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.14.ffn_down_shexp.weight | |
| 266: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_gate_exps.weight | |
| 267: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight | |
| 268: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight | |
| 269: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.14.ffn_gate_shexp.weight | |
| 270: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_up_exps.weight | |
| 271: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.14.ffn_up_shexp.weight | |
| 272: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight | |
| 273: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a | |
| 274: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.14.ssm_alpha.weight | |
| 275: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.14.ssm_beta.weight | |
| 276: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight | |
| 277: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias | |
| 278: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight | |
| 279: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.14.ssm_out.weight | |
| 280: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.15.attn_k.weight | |
| 281: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight | |
| 282: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight | |
| 283: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.15.attn_output.weight | |
| 284: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.15.attn_q.weight | |
| 285: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight | |
| 286: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.15.attn_v.weight | |
| 287: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.15.ffn_down_exps.weight | |
| 288: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.15.ffn_down_shexp.weight | |
| 289: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_gate_exps.weight | |
| 290: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight | |
| 291: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight | |
| 292: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.15.ffn_gate_shexp.weight | |
| 293: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_up_exps.weight | |
| 294: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.15.ffn_up_shexp.weight | |
| 295: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight | |
| 296: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.16.attn_gate.weight | |
| 297: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight | |
| 298: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.16.attn_qkv.weight | |
| 299: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.16.ffn_down_exps.weight | |
| 300: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.16.ffn_down_shexp.weight | |
| 301: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_gate_exps.weight | |
| 302: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight | |
| 303: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight | |
| 304: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.16.ffn_gate_shexp.weight | |
| 305: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_up_exps.weight | |
| 306: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.16.ffn_up_shexp.weight | |
| 307: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight | |
| 308: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a | |
| 309: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.16.ssm_alpha.weight | |
| 310: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.16.ssm_beta.weight | |
| 311: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight | |
| 312: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias | |
| 313: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight | |
| 314: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.16.ssm_out.weight | |
| 315: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.17.attn_gate.weight | |
| 316: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight | |
| 317: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.17.attn_qkv.weight | |
| 318: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.17.ffn_down_exps.weight | |
| 319: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.17.ffn_down_shexp.weight | |
| 320: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_gate_exps.weight | |
| 321: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight | |
| 322: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight | |
| 323: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.17.ffn_gate_shexp.weight | |
| 324: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_up_exps.weight | |
| 325: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.17.ffn_up_shexp.weight | |
| 326: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight | |
| 327: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a | |
| 328: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.17.ssm_alpha.weight | |
| 329: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.17.ssm_beta.weight | |
| 330: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight | |
| 331: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias | |
| 332: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight | |
| 333: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.17.ssm_out.weight | |
| 334: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.18.attn_gate.weight | |
| 335: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight | |
| 336: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.18.attn_qkv.weight | |
| 337: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.18.ffn_down_exps.weight | |
| 338: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.18.ffn_down_shexp.weight | |
| 339: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_gate_exps.weight | |
| 340: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight | |
| 341: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight | |
| 342: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.18.ffn_gate_shexp.weight | |
| 343: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_up_exps.weight | |
| 344: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.18.ffn_up_shexp.weight | |
| 345: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight | |
| 346: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a | |
| 347: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.18.ssm_alpha.weight | |
| 348: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.18.ssm_beta.weight | |
| 349: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight | |
| 350: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias | |
| 351: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight | |
| 352: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.18.ssm_out.weight | |
| 353: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.19.attn_k.weight | |
| 354: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight | |
| 355: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight | |
| 356: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.19.attn_output.weight | |
| 357: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.19.attn_q.weight | |
| 358: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight | |
| 359: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.19.attn_v.weight | |
| 360: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.19.ffn_down_exps.weight | |
| 361: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.19.ffn_down_shexp.weight | |
| 362: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_gate_exps.weight | |
| 363: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight | |
| 364: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight | |
| 365: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.19.ffn_gate_shexp.weight | |
| 366: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_up_exps.weight | |
| 367: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.19.ffn_up_shexp.weight | |
| 368: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight | |
| 369: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.20.attn_gate.weight | |
| 370: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight | |
| 371: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.20.attn_qkv.weight | |
| 372: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.20.ffn_down_exps.weight | |
| 373: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.20.ffn_down_shexp.weight | |
| 374: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_gate_exps.weight | |
| 375: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight | |
| 376: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight | |
| 377: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.20.ffn_gate_shexp.weight | |
| 378: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_up_exps.weight | |
| 379: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.20.ffn_up_shexp.weight | |
| 380: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight | |
| 381: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a | |
| 382: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.20.ssm_alpha.weight | |
| 383: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.20.ssm_beta.weight | |
| 384: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight | |
| 385: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias | |
| 386: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight | |
| 387: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.20.ssm_out.weight | |
| 388: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.21.attn_gate.weight | |
| 389: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight | |
| 390: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.21.attn_qkv.weight | |
| 391: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.21.ffn_down_exps.weight | |
| 392: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.21.ffn_down_shexp.weight | |
| 393: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_gate_exps.weight | |
| 394: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight | |
| 395: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight | |
| 396: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.21.ffn_gate_shexp.weight | |
| 397: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_up_exps.weight | |
| 398: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.21.ffn_up_shexp.weight | |
| 399: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight | |
| 400: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a | |
| 401: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.21.ssm_alpha.weight | |
| 402: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.21.ssm_beta.weight | |
| 403: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight | |
| 404: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias | |
| 405: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight | |
| 406: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.21.ssm_out.weight | |
| 407: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.22.attn_gate.weight | |
| 408: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight | |
| 409: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.22.attn_qkv.weight | |
| 410: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.22.ffn_down_exps.weight | |
| 411: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.22.ffn_down_shexp.weight | |
| 412: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_gate_exps.weight | |
| 413: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight | |
| 414: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight | |
| 415: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.22.ffn_gate_shexp.weight | |
| 416: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_up_exps.weight | |
| 417: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.22.ffn_up_shexp.weight | |
| 418: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight | |
| 419: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a | |
| 420: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.22.ssm_alpha.weight | |
| 421: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.22.ssm_beta.weight | |
| 422: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight | |
| 423: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias | |
| 424: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight | |
| 425: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.22.ssm_out.weight | |
| 426: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.23.attn_k.weight | |
| 427: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight | |
| 428: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight | |
| 429: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.23.attn_output.weight | |
| 430: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.23.attn_q.weight | |
| 431: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight | |
| 432: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.23.attn_v.weight | |
| 433: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.23.ffn_down_exps.weight | |
| 434: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.23.ffn_down_shexp.weight | |
| 435: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_gate_exps.weight | |
| 436: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight | |
| 437: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight | |
| 438: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.23.ffn_gate_shexp.weight | |
| 439: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_up_exps.weight | |
| 440: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.23.ffn_up_shexp.weight | |
| 441: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight | |
| 442: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.24.attn_gate.weight | |
| 443: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight | |
| 444: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.24.attn_qkv.weight | |
| 445: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.24.ffn_down_exps.weight | |
| 446: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.24.ffn_down_shexp.weight | |
| 447: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_gate_exps.weight | |
| 448: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight | |
| 449: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight | |
| 450: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.24.ffn_gate_shexp.weight | |
| 451: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_up_exps.weight | |
| 452: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.24.ffn_up_shexp.weight | |
| 453: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight | |
| 454: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a | |
| 455: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.24.ssm_alpha.weight | |
| 456: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.24.ssm_beta.weight | |
| 457: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight | |
| 458: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias | |
| 459: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight | |
| 460: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.24.ssm_out.weight | |
| 461: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.25.attn_gate.weight | |
| 462: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight | |
| 463: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.25.attn_qkv.weight | |
| 464: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.25.ffn_down_exps.weight | |
| 465: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.25.ffn_down_shexp.weight | |
| 466: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_gate_exps.weight | |
| 467: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight | |
| 468: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight | |
| 469: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.25.ffn_gate_shexp.weight | |
| 470: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_up_exps.weight | |
| 471: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.25.ffn_up_shexp.weight | |
| 472: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight | |
| 473: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a | |
| 474: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.25.ssm_alpha.weight | |
| 475: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.25.ssm_beta.weight | |
| 476: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight | |
| 477: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias | |
| 478: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight | |
| 479: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.25.ssm_out.weight | |
| 480: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.26.attn_gate.weight | |
| 481: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight | |
| 482: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.26.attn_qkv.weight | |
| 483: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.26.ffn_down_exps.weight | |
| 484: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.26.ffn_down_shexp.weight | |
| 485: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_gate_exps.weight | |
| 486: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight | |
| 487: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight | |
| 488: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.26.ffn_gate_shexp.weight | |
| 489: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_up_exps.weight | |
| 490: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.26.ffn_up_shexp.weight | |
| 491: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight | |
| 492: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a | |
| 493: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.26.ssm_alpha.weight | |
| 494: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.26.ssm_beta.weight | |
| 495: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight | |
| 496: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias | |
| 497: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight | |
| 498: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.26.ssm_out.weight | |
| 499: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.27.attn_k.weight | |
| 500: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight | |
| 501: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight | |
| 502: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.27.attn_output.weight | |
| 503: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.27.attn_q.weight | |
| 504: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight | |
| 505: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.27.attn_v.weight | |
| 506: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.27.ffn_down_exps.weight | |
| 507: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.27.ffn_down_shexp.weight | |
| 508: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_gate_exps.weight | |
| 509: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight | |
| 510: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight | |
| 511: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.27.ffn_gate_shexp.weight | |
| 512: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_up_exps.weight | |
| 513: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.27.ffn_up_shexp.weight | |
| 514: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight | |
| 515: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.28.attn_gate.weight | |
| 516: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight | |
| 517: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.28.attn_qkv.weight | |
| 518: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.28.ffn_down_exps.weight | |
| 519: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.28.ffn_down_shexp.weight | |
| 520: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_gate_exps.weight | |
| 521: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight | |
| 522: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight | |
| 523: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.28.ffn_gate_shexp.weight | |
| 524: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_up_exps.weight | |
| 525: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.28.ffn_up_shexp.weight | |
| 526: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight | |
| 527: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a | |
| 528: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.28.ssm_alpha.weight | |
| 529: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.28.ssm_beta.weight | |
| 530: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight | |
| 531: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias | |
| 532: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight | |
| 533: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.28.ssm_out.weight | |
| 534: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.29.attn_gate.weight | |
| 535: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight | |
| 536: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.29.attn_qkv.weight | |
| 537: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.29.ffn_down_exps.weight | |
| 538: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.29.ffn_down_shexp.weight | |
| 539: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_gate_exps.weight | |
| 540: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight | |
| 541: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight | |
| 542: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.29.ffn_gate_shexp.weight | |
| 543: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_up_exps.weight | |
| 544: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.29.ffn_up_shexp.weight | |
| 545: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight | |
| 546: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a | |
| 547: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.29.ssm_alpha.weight | |
| 548: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.29.ssm_beta.weight | |
| 549: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight | |
| 550: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias | |
| 551: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight | |
| 552: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.29.ssm_out.weight | |
| 553: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.30.attn_gate.weight | |
| 554: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight | |
| 555: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.30.attn_qkv.weight | |
| 556: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.30.ffn_down_exps.weight | |
| 557: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.30.ffn_down_shexp.weight | |
| 558: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_gate_exps.weight | |
| 559: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight | |
| 560: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight | |
| 561: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.30.ffn_gate_shexp.weight | |
| 562: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_up_exps.weight | |
| 563: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.30.ffn_up_shexp.weight | |
| 564: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight | |
| 565: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a | |
| 566: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.30.ssm_alpha.weight | |
| 567: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.30.ssm_beta.weight | |
| 568: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight | |
| 569: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias | |
| 570: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight | |
| 571: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.30.ssm_out.weight | |
| 572: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.31.attn_k.weight | |
| 573: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight | |
| 574: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight | |
| 575: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.31.attn_output.weight | |
| 576: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.31.attn_q.weight | |
| 577: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight | |
| 578: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.31.attn_v.weight | |
| 579: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.31.ffn_down_exps.weight | |
| 580: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.31.ffn_down_shexp.weight | |
| 581: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_gate_exps.weight | |
| 582: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight | |
| 583: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight | |
| 584: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.31.ffn_gate_shexp.weight | |
| 585: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_up_exps.weight | |
| 586: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.31.ffn_up_shexp.weight | |
| 587: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight | |
| 588: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.32.attn_gate.weight | |
| 589: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight | |
| 590: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.32.attn_qkv.weight | |
| 591: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.32.ffn_down_exps.weight | |
| 592: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.32.ffn_down_shexp.weight | |
| 593: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_gate_exps.weight | |
| 594: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight | |
| 595: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight | |
| 596: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.32.ffn_gate_shexp.weight | |
| 597: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_up_exps.weight | |
| 598: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.32.ffn_up_shexp.weight | |
| 599: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight | |
| 600: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a | |
| 601: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.32.ssm_alpha.weight | |
| 602: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.32.ssm_beta.weight | |
| 603: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight | |
| 604: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias | |
| 605: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight | |
| 606: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.32.ssm_out.weight | |
| 607: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.33.attn_gate.weight | |
| 608: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight | |
| 609: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.33.attn_qkv.weight | |
| 610: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.33.ffn_down_exps.weight | |
| 611: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.33.ffn_down_shexp.weight | |
| 612: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_gate_exps.weight | |
| 613: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight | |
| 614: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight | |
| 615: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.33.ffn_gate_shexp.weight | |
| 616: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_up_exps.weight | |
| 617: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.33.ffn_up_shexp.weight | |
| 618: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight | |
| 619: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a | |
| 620: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.33.ssm_alpha.weight | |
| 621: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.33.ssm_beta.weight | |
| 622: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight | |
| 623: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias | |
| 624: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight | |
| 625: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.33.ssm_out.weight | |
| 626: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.34.attn_gate.weight | |
| 627: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight | |
| 628: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.34.attn_qkv.weight | |
| 629: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.34.ffn_down_exps.weight | |
| 630: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.34.ffn_down_shexp.weight | |
| 631: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_gate_exps.weight | |
| 632: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight | |
| 633: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight | |
| 634: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.34.ffn_gate_shexp.weight | |
| 635: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_up_exps.weight | |
| 636: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.34.ffn_up_shexp.weight | |
| 637: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight | |
| 638: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a | |
| 639: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.34.ssm_alpha.weight | |
| 640: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.34.ssm_beta.weight | |
| 641: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight | |
| 642: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias | |
| 643: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight | |
| 644: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.34.ssm_out.weight | |
| 645: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.35.attn_k.weight | |
| 646: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight | |
| 647: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight | |
| 648: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.35.attn_output.weight | |
| 649: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.35.attn_q.weight | |
| 650: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight | |
| 651: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.35.attn_v.weight | |
| 652: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.35.ffn_down_exps.weight | |
| 653: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.35.ffn_down_shexp.weight | |
| 654: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_gate_exps.weight | |
| 655: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight | |
| 656: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight | |
| 657: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.35.ffn_gate_shexp.weight | |
| 658: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_up_exps.weight | |
| 659: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.35.ffn_up_shexp.weight | |
| 660: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight | |
| 661: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.36.attn_gate.weight | |
| 662: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight | |
| 663: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.36.attn_qkv.weight | |
| 664: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.36.ffn_down_exps.weight | |
| 665: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.36.ffn_down_shexp.weight | |
| 666: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_gate_exps.weight | |
| 667: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight | |
| 668: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight | |
| 669: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.36.ffn_gate_shexp.weight | |
| 670: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_up_exps.weight | |
| 671: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.36.ffn_up_shexp.weight | |
| 672: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight | |
| 673: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a | |
| 674: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.36.ssm_alpha.weight | |
| 675: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.36.ssm_beta.weight | |
| 676: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight | |
| 677: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias | |
| 678: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight | |
| 679: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.36.ssm_out.weight | |
| 680: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.37.attn_gate.weight | |
| 681: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight | |
| 682: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.37.attn_qkv.weight | |
| 683: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.37.ffn_down_exps.weight | |
| 684: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.37.ffn_down_shexp.weight | |
| 685: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_gate_exps.weight | |
| 686: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight | |
| 687: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight | |
| 688: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.37.ffn_gate_shexp.weight | |
| 689: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_up_exps.weight | |
| 690: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.37.ffn_up_shexp.weight | |
| 691: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight | |
| 692: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a | |
| 693: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.37.ssm_alpha.weight | |
| 694: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.37.ssm_beta.weight | |
| 695: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight | |
| 696: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias | |
| 697: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight | |
| 698: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.37.ssm_out.weight | |
| 699: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.38.attn_gate.weight | |
| 700: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight | |
| 701: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.38.attn_qkv.weight | |
| 702: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.38.ffn_down_exps.weight | |
| 703: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.38.ffn_down_shexp.weight | |
| 704: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_gate_exps.weight | |
| 705: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight | |
| 706: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight | |
| 707: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.38.ffn_gate_shexp.weight | |
| 708: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_up_exps.weight | |
| 709: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.38.ffn_up_shexp.weight | |
| 710: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight | |
| 711: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a | |
| 712: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.38.ssm_alpha.weight | |
| 713: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.38.ssm_beta.weight | |
| 714: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight | |
| 715: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias | |
| 716: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight | |
| 717: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.38.ssm_out.weight | |
| 718: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.39.attn_k.weight | |
| 719: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight | |
| 720: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight | |
| 721: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.39.attn_output.weight | |
| 722: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.39.attn_q.weight | |
| 723: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight | |
| 724: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.39.attn_v.weight | |
| 725: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.39.ffn_down_exps.weight | |
| 726: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.39.ffn_down_shexp.weight | |
| 727: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_gate_exps.weight | |
| 728: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight | |
| 729: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight | |
| 730: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.39.ffn_gate_shexp.weight | |
| 731: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_up_exps.weight | |
| 732: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.39.ffn_up_shexp.weight | |
| 733: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight | |
| 734: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.40.attn_gate.weight | |
| 735: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight | |
| 736: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.40.attn_qkv.weight | |
| 737: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.40.ffn_down_exps.weight | |
| 738: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.40.ffn_down_shexp.weight | |
| 739: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_gate_exps.weight | |
| 740: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight | |
| 741: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight | |
| 742: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.40.ffn_gate_shexp.weight | |
| 743: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_up_exps.weight | |
| 744: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.40.ffn_up_shexp.weight | |
| 745: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight | |
| 746: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a | |
| 747: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.40.ssm_alpha.weight | |
| 748: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.40.ssm_beta.weight | |
| 749: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight | |
| 750: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias | |
| 751: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight | |
| 752: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.40.ssm_out.weight | |
| 753: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.41.attn_gate.weight | |
| 754: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight | |
| 755: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.41.attn_qkv.weight | |
| 756: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.41.ffn_down_exps.weight | |
| 757: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.41.ffn_down_shexp.weight | |
| 758: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_gate_exps.weight | |
| 759: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight | |
| 760: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight | |
| 761: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.41.ffn_gate_shexp.weight | |
| 762: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_up_exps.weight | |
| 763: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.41.ffn_up_shexp.weight | |
| 764: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight | |
| 765: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a | |
| 766: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.41.ssm_alpha.weight | |
| 767: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.41.ssm_beta.weight | |
| 768: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight | |
| 769: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias | |
| 770: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight | |
| 771: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.41.ssm_out.weight | |
| 772: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.42.attn_gate.weight | |
| 773: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight | |
| 774: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.42.attn_qkv.weight | |
| 775: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.42.ffn_down_exps.weight | |
| 776: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.42.ffn_down_shexp.weight | |
| 777: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_gate_exps.weight | |
| 778: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight | |
| 779: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight | |
| 780: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.42.ffn_gate_shexp.weight | |
| 781: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_up_exps.weight | |
| 782: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.42.ffn_up_shexp.weight | |
| 783: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight | |
| 784: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a | |
| 785: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.42.ssm_alpha.weight | |
| 786: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.42.ssm_beta.weight | |
| 787: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight | |
| 788: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias | |
| 789: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight | |
| 790: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.42.ssm_out.weight | |
| 791: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.43.attn_k.weight | |
| 792: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight | |
| 793: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight | |
| 794: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.43.attn_output.weight | |
| 795: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.43.attn_q.weight | |
| 796: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight | |
| 797: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.43.attn_v.weight | |
| 798: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.43.ffn_down_exps.weight | |
| 799: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.43.ffn_down_shexp.weight | |
| 800: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_gate_exps.weight | |
| 801: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight | |
| 802: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight | |
| 803: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.43.ffn_gate_shexp.weight | |
| 804: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_up_exps.weight | |
| 805: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.43.ffn_up_shexp.weight | |
| 806: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight | |
| 807: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.44.attn_gate.weight | |
| 808: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight | |
| 809: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.44.attn_qkv.weight | |
| 810: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.44.ffn_down_exps.weight | |
| 811: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.44.ffn_down_shexp.weight | |
| 812: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_gate_exps.weight | |
| 813: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight | |
| 814: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight | |
| 815: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.44.ffn_gate_shexp.weight | |
| 816: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_up_exps.weight | |
| 817: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.44.ffn_up_shexp.weight | |
| 818: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight | |
| 819: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a | |
| 820: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.44.ssm_alpha.weight | |
| 821: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.44.ssm_beta.weight | |
| 822: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight | |
| 823: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias | |
| 824: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight | |
| 825: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.44.ssm_out.weight | |
| 826: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.45.attn_gate.weight | |
| 827: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight | |
| 828: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.45.attn_qkv.weight | |
| 829: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.45.ffn_down_exps.weight | |
| 830: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.45.ffn_down_shexp.weight | |
| 831: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_gate_exps.weight | |
| 832: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight | |
| 833: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight | |
| 834: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.45.ffn_gate_shexp.weight | |
| 835: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_up_exps.weight | |
| 836: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.45.ffn_up_shexp.weight | |
| 837: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight | |
| 838: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a | |
| 839: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.45.ssm_alpha.weight | |
| 840: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.45.ssm_beta.weight | |
| 841: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight | |
| 842: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias | |
| 843: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight | |
| 844: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.45.ssm_out.weight | |
| 845: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.46.attn_gate.weight | |
| 846: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight | |
| 847: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.46.attn_qkv.weight | |
| 848: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.46.ffn_down_exps.weight | |
| 849: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.46.ffn_down_shexp.weight | |
| 850: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_gate_exps.weight | |
| 851: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight | |
| 852: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight | |
| 853: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.46.ffn_gate_shexp.weight | |
| 854: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_up_exps.weight | |
| 855: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.46.ffn_up_shexp.weight | |
| 856: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight | |
| 857: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a | |
| 858: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.46.ssm_alpha.weight | |
| 859: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.46.ssm_beta.weight | |
| 860: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight | |
| 861: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias | |
| 862: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight | |
| 863: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.46.ssm_out.weight | |
| 864: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.47.attn_k.weight | |
| 865: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight | |
| 866: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight | |
| 867: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.47.attn_output.weight | |
| 868: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.47.attn_q.weight | |
| 869: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight | |
| 870: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.47.attn_v.weight | |
| 871: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.47.ffn_down_exps.weight | |
| 872: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.47.ffn_down_shexp.weight | |
| 873: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_gate_exps.weight | |
| 874: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight | |
| 875: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight | |
| 876: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.47.ffn_gate_shexp.weight | |
| 877: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_up_exps.weight | |
| 878: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.47.ffn_up_shexp.weight | |
| 879: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight | |
| </details> | |
| **MQ516_special (Mixed Quant Q5_K - BF16, Special Purpose)** | |
| <details> | |
| <summary>llama-quantize params and gguf-dump</summary> | |
| --leave-output-tensor | |
| --tensor-type "blk\.(0|1|2|4|5|6|8|9|10)\..*ffn_(down|gate|up)_exps\.weight=q5_k" | |
| --tensor-type ".*ffn_(down|gate|up)_shexp\.weight=bf16" | |
| --tensor-type "blk\.(3|7|11|15|19|23|27|31|35|39|43|47)\..*attn_(q|k|v|output)\.weight=bf16" | |
| --tensor-type ".*ssm_(alpha|beta|out)\.weight=bf16" | |
| --tensor-type "token_embd\.weight=bf16" | |
| --tensor-type ".*attn_(qkv|gate)\.weight=bf16" | |
| Q6_K | |
| 1: UINT32 | 1 | GGUF.version = 3 | |
| 2: UINT64 | 1 | GGUF.tensor_count = 879 | |
| 3: UINT64 | 1 | GGUF.kv_count = 43 | |
| 4: STRING | 1 | general.architecture = 'qwen35moe' | |
| 5: STRING | 1 | general.type = 'model' | |
| 6: INT32 | 1 | general.sampling.top_k = 20 | |
| 7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071 | |
| 8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579 | |
| 9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Wangzhang Abliterix No MTP' | |
| 10: STRING | 1 | general.size_label = '256x6.6B' | |
| 11: UINT32 | 1 | general.base_model.count = 1 | |
| 12: STRING | 1 | general.base_model.0.name = 'Qwen3.5 122B A10B' | |
| 13: STRING | 1 | general.base_model.0.organization = 'Qwen' | |
| 14: STRING | 1 | general.base_model.0.repo_url = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B' | |
| 15: [STRING] | 5 | general.tags = ['abliterix', 'uncensored', 'decensored', 'abliterated', 'text-generation'] | |
| 16: UINT32 | 1 | qwen35moe.block_count = 48 | |
| 17: UINT32 | 1 | qwen35moe.context_length = 262144 | |
| 18: UINT32 | 1 | qwen35moe.embedding_length = 3072 | |
| 19: UINT32 | 1 | qwen35moe.attention.head_count = 32 | |
| 20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2 | |
| 21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0] | |
| 22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0 | |
| 23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07 | |
| 24: UINT32 | 1 | qwen35moe.expert_count = 256 | |
| 25: UINT32 | 1 | qwen35moe.expert_used_count = 8 | |
| 26: UINT32 | 1 | qwen35moe.attention.key_length = 256 | |
| 27: UINT32 | 1 | qwen35moe.attention.value_length = 256 | |
| 28: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024 | |
| 29: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024 | |
| 30: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4 | |
| 31: UINT32 | 1 | qwen35moe.ssm.state_size = 128 | |
| 32: UINT32 | 1 | qwen35moe.ssm.group_count = 16 | |
| 33: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64 | |
| 34: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192 | |
| 35: UINT32 | 1 | qwen35moe.full_attention_interval = 4 | |
| 36: UINT32 | 1 | qwen35moe.rope.dimension_count = 64 | |
| 37: STRING | 1 | tokenizer.ggml.model = 'gpt2' | |
| 38: STRING | 1 | tokenizer.ggml.pre = 'qwen35' | |
| 39: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...] | |
| 40: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...] | |
| 41: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...] | |
| 42: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046 | |
| 43: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044 | |
| 44: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...' | |
| 45: UINT32 | 1 | general.quantization_version = 2 | |
| 46: UINT32 | 1 | general.file_type = 18 | |
| 1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight | |
| 2: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight | |
| 3: 762839040 | 3072, 248320, 1, 1 | BF16 | token_embd.weight | |
| 4: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.0.attn_gate.weight | |
| 5: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight | |
| 6: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.0.attn_qkv.weight | |
| 7: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.0.ffn_down_exps.weight | |
| 8: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.0.ffn_down_shexp.weight | |
| 9: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.0.ffn_gate_exps.weight | |
| 10: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight | |
| 11: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight | |
| 12: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_gate_shexp.weight | |
| 13: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.0.ffn_up_exps.weight | |
| 14: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_up_shexp.weight | |
| 15: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight | |
| 16: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a | |
| 17: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_alpha.weight | |
| 18: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_beta.weight | |
| 19: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight | |
| 20: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias | |
| 21: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight | |
| 22: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.0.ssm_out.weight | |
| 23: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.1.attn_gate.weight | |
| 24: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight | |
| 25: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.1.attn_qkv.weight | |
| 26: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.1.ffn_down_exps.weight | |
| 27: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.1.ffn_down_shexp.weight | |
| 28: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.1.ffn_gate_exps.weight | |
| 29: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight | |
| 30: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight | |
| 31: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_gate_shexp.weight | |
| 32: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.1.ffn_up_exps.weight | |
| 33: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_up_shexp.weight | |
| 34: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight | |
| 35: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a | |
| 36: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_alpha.weight | |
| 37: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_beta.weight | |
| 38: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight | |
| 39: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias | |
| 40: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight | |
| 41: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.1.ssm_out.weight | |
| 42: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.2.attn_gate.weight | |
| 43: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight | |
| 44: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.2.attn_qkv.weight | |
| 45: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.2.ffn_down_exps.weight | |
| 46: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.2.ffn_down_shexp.weight | |
| 47: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.2.ffn_gate_exps.weight | |
| 48: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight | |
| 49: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight | |
| 50: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_gate_shexp.weight | |
| 51: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.2.ffn_up_exps.weight | |
| 52: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_up_shexp.weight | |
| 53: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight | |
| 54: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a | |
| 55: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_alpha.weight | |
| 56: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_beta.weight | |
| 57: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight | |
| 58: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias | |
| 59: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight | |
| 60: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.2.ssm_out.weight | |
| 61: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_k.weight | |
| 62: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight | |
| 63: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight | |
| 64: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.3.attn_output.weight | |
| 65: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.3.attn_q.weight | |
| 66: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight | |
| 67: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_v.weight | |
| 68: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.3.ffn_down_exps.weight | |
| 69: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.3.ffn_down_shexp.weight | |
| 70: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_gate_exps.weight | |
| 71: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight | |
| 72: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight | |
| 73: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_gate_shexp.weight | |
| 74: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_up_exps.weight | |
| 75: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_up_shexp.weight | |
| 76: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight | |
| 77: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.4.attn_gate.weight | |
| 78: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight | |
| 79: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.4.attn_qkv.weight | |
| 80: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.4.ffn_down_exps.weight | |
| 81: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.4.ffn_down_shexp.weight | |
| 82: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.4.ffn_gate_exps.weight | |
| 83: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight | |
| 84: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight | |
| 85: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_gate_shexp.weight | |
| 86: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.4.ffn_up_exps.weight | |
| 87: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_up_shexp.weight | |
| 88: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight | |
| 89: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a | |
| 90: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_alpha.weight | |
| 91: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_beta.weight | |
| 92: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight | |
| 93: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias | |
| 94: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight | |
| 95: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.4.ssm_out.weight | |
| 96: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.5.attn_gate.weight | |
| 97: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight | |
| 98: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.5.attn_qkv.weight | |
| 99: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.5.ffn_down_exps.weight | |
| 100: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.5.ffn_down_shexp.weight | |
| 101: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.5.ffn_gate_exps.weight | |
| 102: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight | |
| 103: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight | |
| 104: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_gate_shexp.weight | |
| 105: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.5.ffn_up_exps.weight | |
| 106: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_up_shexp.weight | |
| 107: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight | |
| 108: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a | |
| 109: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_alpha.weight | |
| 110: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_beta.weight | |
| 111: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight | |
| 112: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias | |
| 113: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight | |
| 114: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.5.ssm_out.weight | |
| 115: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.6.attn_gate.weight | |
| 116: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight | |
| 117: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.6.attn_qkv.weight | |
| 118: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.6.ffn_down_exps.weight | |
| 119: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.6.ffn_down_shexp.weight | |
| 120: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.6.ffn_gate_exps.weight | |
| 121: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight | |
| 122: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight | |
| 123: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_gate_shexp.weight | |
| 124: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.6.ffn_up_exps.weight | |
| 125: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_up_shexp.weight | |
| 126: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight | |
| 127: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a | |
| 128: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_alpha.weight | |
| 129: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_beta.weight | |
| 130: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight | |
| 131: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias | |
| 132: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight | |
| 133: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.6.ssm_out.weight | |
| 134: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_k.weight | |
| 135: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight | |
| 136: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight | |
| 137: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.7.attn_output.weight | |
| 138: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.7.attn_q.weight | |
| 139: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight | |
| 140: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_v.weight | |
| 141: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.7.ffn_down_exps.weight | |
| 142: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.7.ffn_down_shexp.weight | |
| 143: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_gate_exps.weight | |
| 144: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight | |
| 145: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight | |
| 146: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_gate_shexp.weight | |
| 147: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_up_exps.weight | |
| 148: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_up_shexp.weight | |
| 149: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight | |
| 150: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.8.attn_gate.weight | |
| 151: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight | |
| 152: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.8.attn_qkv.weight | |
| 153: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.8.ffn_down_exps.weight | |
| 154: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.8.ffn_down_shexp.weight | |
| 155: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.8.ffn_gate_exps.weight | |
| 156: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight | |
| 157: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight | |
| 158: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_gate_shexp.weight | |
| 159: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.8.ffn_up_exps.weight | |
| 160: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_up_shexp.weight | |
| 161: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight | |
| 162: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a | |
| 163: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_alpha.weight | |
| 164: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_beta.weight | |
| 165: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight | |
| 166: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias | |
| 167: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight | |
| 168: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.8.ssm_out.weight | |
| 169: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.9.attn_gate.weight | |
| 170: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight | |
| 171: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.9.attn_qkv.weight | |
| 172: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.9.ffn_down_exps.weight | |
| 173: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.9.ffn_down_shexp.weight | |
| 174: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.9.ffn_gate_exps.weight | |
| 175: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight | |
| 176: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight | |
| 177: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_gate_shexp.weight | |
| 178: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.9.ffn_up_exps.weight | |
| 179: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_up_shexp.weight | |
| 180: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight | |
| 181: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a | |
| 182: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_alpha.weight | |
| 183: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_beta.weight | |
| 184: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight | |
| 185: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias | |
| 186: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight | |
| 187: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.9.ssm_out.weight | |
| 188: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.10.attn_gate.weight | |
| 189: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight | |
| 190: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.10.attn_qkv.weight | |
| 191: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.10.ffn_down_exps.weight | |
| 192: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.10.ffn_down_shexp.weight | |
| 193: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.10.ffn_gate_exps.weight | |
| 194: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight | |
| 195: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight | |
| 196: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_gate_shexp.weight | |
| 197: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.10.ffn_up_exps.weight | |
| 198: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_up_shexp.weight | |
| 199: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight | |
| 200: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a | |
| 201: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_alpha.weight | |
| 202: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_beta.weight | |
| 203: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight | |
| 204: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias | |
| 205: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight | |
| 206: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.10.ssm_out.weight | |
| 207: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_k.weight | |
| 208: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight | |
| 209: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight | |
| 210: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.11.attn_output.weight | |
| 211: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.11.attn_q.weight | |
| 212: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight | |
| 213: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_v.weight | |
| 214: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.11.ffn_down_exps.weight | |
| 215: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.11.ffn_down_shexp.weight | |
| 216: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_gate_exps.weight | |
| 217: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight | |
| 218: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight | |
| 219: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_gate_shexp.weight | |
| 220: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_up_exps.weight | |
| 221: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_up_shexp.weight | |
| 222: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight | |
| 223: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.12.attn_gate.weight | |
| 224: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight | |
| 225: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.12.attn_qkv.weight | |
| 226: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.12.ffn_down_exps.weight | |
| 227: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.12.ffn_down_shexp.weight | |
| 228: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_gate_exps.weight | |
| 229: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight | |
| 230: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight | |
| 231: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_gate_shexp.weight | |
| 232: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_up_exps.weight | |
| 233: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_up_shexp.weight | |
| 234: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight | |
| 235: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a | |
| 236: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_alpha.weight | |
| 237: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_beta.weight | |
| 238: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight | |
| 239: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias | |
| 240: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight | |
| 241: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.12.ssm_out.weight | |
| 242: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.13.attn_gate.weight | |
| 243: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight | |
| 244: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.13.attn_qkv.weight | |
| 245: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.13.ffn_down_exps.weight | |
| 246: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.13.ffn_down_shexp.weight | |
| 247: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_gate_exps.weight | |
| 248: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight | |
| 249: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight | |
| 250: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_gate_shexp.weight | |
| 251: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_up_exps.weight | |
| 252: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_up_shexp.weight | |
| 253: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight | |
| 254: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a | |
| 255: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_alpha.weight | |
| 256: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_beta.weight | |
| 257: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight | |
| 258: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias | |
| 259: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight | |
| 260: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.13.ssm_out.weight | |
| 261: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.14.attn_gate.weight | |
| 262: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight | |
| 263: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.14.attn_qkv.weight | |
| 264: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.14.ffn_down_exps.weight | |
| 265: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.14.ffn_down_shexp.weight | |
| 266: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_gate_exps.weight | |
| 267: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight | |
| 268: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight | |
| 269: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_gate_shexp.weight | |
| 270: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_up_exps.weight | |
| 271: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_up_shexp.weight | |
| 272: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight | |
| 273: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a | |
| 274: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_alpha.weight | |
| 275: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_beta.weight | |
| 276: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight | |
| 277: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias | |
| 278: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight | |
| 279: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.14.ssm_out.weight | |
| 280: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_k.weight | |
| 281: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight | |
| 282: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight | |
| 283: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.15.attn_output.weight | |
| 284: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.15.attn_q.weight | |
| 285: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight | |
| 286: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_v.weight | |
| 287: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.15.ffn_down_exps.weight | |
| 288: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.15.ffn_down_shexp.weight | |
| 289: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_gate_exps.weight | |
| 290: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight | |
| 291: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight | |
| 292: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_gate_shexp.weight | |
| 293: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_up_exps.weight | |
| 294: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_up_shexp.weight | |
| 295: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight | |
| 296: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.16.attn_gate.weight | |
| 297: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight | |
| 298: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.16.attn_qkv.weight | |
| 299: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.16.ffn_down_exps.weight | |
| 300: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.16.ffn_down_shexp.weight | |
| 301: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_gate_exps.weight | |
| 302: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight | |
| 303: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight | |
| 304: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_gate_shexp.weight | |
| 305: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_up_exps.weight | |
| 306: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_up_shexp.weight | |
| 307: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight | |
| 308: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a | |
| 309: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_alpha.weight | |
| 310: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_beta.weight | |
| 311: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight | |
| 312: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias | |
| 313: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight | |
| 314: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.16.ssm_out.weight | |
| 315: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.17.attn_gate.weight | |
| 316: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight | |
| 317: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.17.attn_qkv.weight | |
| 318: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.17.ffn_down_exps.weight | |
| 319: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.17.ffn_down_shexp.weight | |
| 320: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_gate_exps.weight | |
| 321: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight | |
| 322: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight | |
| 323: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_gate_shexp.weight | |
| 324: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_up_exps.weight | |
| 325: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_up_shexp.weight | |
| 326: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight | |
| 327: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a | |
| 328: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_alpha.weight | |
| 329: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_beta.weight | |
| 330: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight | |
| 331: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias | |
| 332: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight | |
| 333: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.17.ssm_out.weight | |
| 334: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.18.attn_gate.weight | |
| 335: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight | |
| 336: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.18.attn_qkv.weight | |
| 337: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.18.ffn_down_exps.weight | |
| 338: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.18.ffn_down_shexp.weight | |
| 339: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_gate_exps.weight | |
| 340: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight | |
| 341: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight | |
| 342: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_gate_shexp.weight | |
| 343: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_up_exps.weight | |
| 344: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_up_shexp.weight | |
| 345: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight | |
| 346: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a | |
| 347: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_alpha.weight | |
| 348: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_beta.weight | |
| 349: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight | |
| 350: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias | |
| 351: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight | |
| 352: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.18.ssm_out.weight | |
| 353: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_k.weight | |
| 354: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight | |
| 355: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight | |
| 356: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.19.attn_output.weight | |
| 357: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.19.attn_q.weight | |
| 358: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight | |
| 359: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_v.weight | |
| 360: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.19.ffn_down_exps.weight | |
| 361: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.19.ffn_down_shexp.weight | |
| 362: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_gate_exps.weight | |
| 363: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight | |
| 364: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight | |
| 365: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_gate_shexp.weight | |
| 366: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_up_exps.weight | |
| 367: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_up_shexp.weight | |
| 368: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight | |
| 369: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.20.attn_gate.weight | |
| 370: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight | |
| 371: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.20.attn_qkv.weight | |
| 372: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.20.ffn_down_exps.weight | |
| 373: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.20.ffn_down_shexp.weight | |
| 374: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_gate_exps.weight | |
| 375: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight | |
| 376: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight | |
| 377: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_gate_shexp.weight | |
| 378: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_up_exps.weight | |
| 379: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_up_shexp.weight | |
| 380: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight | |
| 381: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a | |
| 382: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_alpha.weight | |
| 383: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_beta.weight | |
| 384: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight | |
| 385: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias | |
| 386: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight | |
| 387: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.20.ssm_out.weight | |
| 388: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.21.attn_gate.weight | |
| 389: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight | |
| 390: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.21.attn_qkv.weight | |
| 391: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.21.ffn_down_exps.weight | |
| 392: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.21.ffn_down_shexp.weight | |
| 393: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_gate_exps.weight | |
| 394: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight | |
| 395: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight | |
| 396: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_gate_shexp.weight | |
| 397: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_up_exps.weight | |
| 398: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_up_shexp.weight | |
| 399: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight | |
| 400: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a | |
| 401: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_alpha.weight | |
| 402: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_beta.weight | |
| 403: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight | |
| 404: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias | |
| 405: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight | |
| 406: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.21.ssm_out.weight | |
| 407: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.22.attn_gate.weight | |
| 408: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight | |
| 409: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.22.attn_qkv.weight | |
| 410: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.22.ffn_down_exps.weight | |
| 411: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.22.ffn_down_shexp.weight | |
| 412: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_gate_exps.weight | |
| 413: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight | |
| 414: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight | |
| 415: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_gate_shexp.weight | |
| 416: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_up_exps.weight | |
| 417: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_up_shexp.weight | |
| 418: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight | |
| 419: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a | |
| 420: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_alpha.weight | |
| 421: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_beta.weight | |
| 422: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight | |
| 423: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias | |
| 424: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight | |
| 425: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.22.ssm_out.weight | |
| 426: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_k.weight | |
| 427: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight | |
| 428: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight | |
| 429: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.23.attn_output.weight | |
| 430: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.23.attn_q.weight | |
| 431: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight | |
| 432: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_v.weight | |
| 433: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.23.ffn_down_exps.weight | |
| 434: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.23.ffn_down_shexp.weight | |
| 435: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_gate_exps.weight | |
| 436: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight | |
| 437: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight | |
| 438: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_gate_shexp.weight | |
| 439: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_up_exps.weight | |
| 440: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_up_shexp.weight | |
| 441: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight | |
| 442: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.24.attn_gate.weight | |
| 443: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight | |
| 444: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.24.attn_qkv.weight | |
| 445: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.24.ffn_down_exps.weight | |
| 446: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.24.ffn_down_shexp.weight | |
| 447: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_gate_exps.weight | |
| 448: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight | |
| 449: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight | |
| 450: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_gate_shexp.weight | |
| 451: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_up_exps.weight | |
| 452: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_up_shexp.weight | |
| 453: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight | |
| 454: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a | |
| 455: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_alpha.weight | |
| 456: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_beta.weight | |
| 457: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight | |
| 458: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias | |
| 459: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight | |
| 460: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.24.ssm_out.weight | |
| 461: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.25.attn_gate.weight | |
| 462: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight | |
| 463: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.25.attn_qkv.weight | |
| 464: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.25.ffn_down_exps.weight | |
| 465: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.25.ffn_down_shexp.weight | |
| 466: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_gate_exps.weight | |
| 467: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight | |
| 468: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight | |
| 469: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_gate_shexp.weight | |
| 470: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_up_exps.weight | |
| 471: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_up_shexp.weight | |
| 472: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight | |
| 473: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a | |
| 474: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_alpha.weight | |
| 475: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_beta.weight | |
| 476: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight | |
| 477: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias | |
| 478: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight | |
| 479: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.25.ssm_out.weight | |
| 480: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.26.attn_gate.weight | |
| 481: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight | |
| 482: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.26.attn_qkv.weight | |
| 483: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.26.ffn_down_exps.weight | |
| 484: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.26.ffn_down_shexp.weight | |
| 485: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_gate_exps.weight | |
| 486: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight | |
| 487: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight | |
| 488: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_gate_shexp.weight | |
| 489: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_up_exps.weight | |
| 490: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_up_shexp.weight | |
| 491: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight | |
| 492: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a | |
| 493: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_alpha.weight | |
| 494: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_beta.weight | |
| 495: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight | |
| 496: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias | |
| 497: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight | |
| 498: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.26.ssm_out.weight | |
| 499: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_k.weight | |
| 500: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight | |
| 501: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight | |
| 502: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.27.attn_output.weight | |
| 503: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.27.attn_q.weight | |
| 504: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight | |
| 505: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_v.weight | |
| 506: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.27.ffn_down_exps.weight | |
| 507: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.27.ffn_down_shexp.weight | |
| 508: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_gate_exps.weight | |
| 509: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight | |
| 510: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight | |
| 511: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_gate_shexp.weight | |
| 512: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_up_exps.weight | |
| 513: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_up_shexp.weight | |
| 514: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight | |
| 515: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.28.attn_gate.weight | |
| 516: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight | |
| 517: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.28.attn_qkv.weight | |
| 518: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.28.ffn_down_exps.weight | |
| 519: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.28.ffn_down_shexp.weight | |
| 520: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_gate_exps.weight | |
| 521: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight | |
| 522: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight | |
| 523: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_gate_shexp.weight | |
| 524: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_up_exps.weight | |
| 525: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_up_shexp.weight | |
| 526: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight | |
| 527: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a | |
| 528: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_alpha.weight | |
| 529: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_beta.weight | |
| 530: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight | |
| 531: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias | |
| 532: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight | |
| 533: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.28.ssm_out.weight | |
| 534: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.29.attn_gate.weight | |
| 535: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight | |
| 536: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.29.attn_qkv.weight | |
| 537: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.29.ffn_down_exps.weight | |
| 538: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.29.ffn_down_shexp.weight | |
| 539: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_gate_exps.weight | |
| 540: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight | |
| 541: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight | |
| 542: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_gate_shexp.weight | |
| 543: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_up_exps.weight | |
| 544: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_up_shexp.weight | |
| 545: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight | |
| 546: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a | |
| 547: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_alpha.weight | |
| 548: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_beta.weight | |
| 549: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight | |
| 550: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias | |
| 551: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight | |
| 552: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.29.ssm_out.weight | |
| 553: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.30.attn_gate.weight | |
| 554: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight | |
| 555: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.30.attn_qkv.weight | |
| 556: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.30.ffn_down_exps.weight | |
| 557: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.30.ffn_down_shexp.weight | |
| 558: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_gate_exps.weight | |
| 559: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight | |
| 560: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight | |
| 561: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_gate_shexp.weight | |
| 562: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_up_exps.weight | |
| 563: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_up_shexp.weight | |
| 564: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight | |
| 565: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a | |
| 566: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_alpha.weight | |
| 567: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_beta.weight | |
| 568: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight | |
| 569: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias | |
| 570: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight | |
| 571: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.30.ssm_out.weight | |
| 572: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_k.weight | |
| 573: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight | |
| 574: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight | |
| 575: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.31.attn_output.weight | |
| 576: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.31.attn_q.weight | |
| 577: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight | |
| 578: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_v.weight | |
| 579: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.31.ffn_down_exps.weight | |
| 580: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.31.ffn_down_shexp.weight | |
| 581: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_gate_exps.weight | |
| 582: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight | |
| 583: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight | |
| 584: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_gate_shexp.weight | |
| 585: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_up_exps.weight | |
| 586: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_up_shexp.weight | |
| 587: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight | |
| 588: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.32.attn_gate.weight | |
| 589: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight | |
| 590: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.32.attn_qkv.weight | |
| 591: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.32.ffn_down_exps.weight | |
| 592: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.32.ffn_down_shexp.weight | |
| 593: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_gate_exps.weight | |
| 594: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight | |
| 595: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight | |
| 596: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_gate_shexp.weight | |
| 597: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_up_exps.weight | |
| 598: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_up_shexp.weight | |
| 599: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight | |
| 600: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a | |
| 601: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_alpha.weight | |
| 602: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_beta.weight | |
| 603: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight | |
| 604: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias | |
| 605: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight | |
| 606: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.32.ssm_out.weight | |
| 607: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.33.attn_gate.weight | |
| 608: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight | |
| 609: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.33.attn_qkv.weight | |
| 610: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.33.ffn_down_exps.weight | |
| 611: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.33.ffn_down_shexp.weight | |
| 612: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_gate_exps.weight | |
| 613: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight | |
| 614: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight | |
| 615: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_gate_shexp.weight | |
| 616: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_up_exps.weight | |
| 617: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_up_shexp.weight | |
| 618: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight | |
| 619: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a | |
| 620: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_alpha.weight | |
| 621: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_beta.weight | |
| 622: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight | |
| 623: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias | |
| 624: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight | |
| 625: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.33.ssm_out.weight | |
| 626: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.34.attn_gate.weight | |
| 627: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight | |
| 628: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.34.attn_qkv.weight | |
| 629: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.34.ffn_down_exps.weight | |
| 630: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.34.ffn_down_shexp.weight | |
| 631: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_gate_exps.weight | |
| 632: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight | |
| 633: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight | |
| 634: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_gate_shexp.weight | |
| 635: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_up_exps.weight | |
| 636: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_up_shexp.weight | |
| 637: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight | |
| 638: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a | |
| 639: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_alpha.weight | |
| 640: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_beta.weight | |
| 641: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight | |
| 642: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias | |
| 643: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight | |
| 644: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.34.ssm_out.weight | |
| 645: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_k.weight | |
| 646: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight | |
| 647: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight | |
| 648: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.35.attn_output.weight | |
| 649: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.35.attn_q.weight | |
| 650: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight | |
| 651: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_v.weight | |
| 652: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.35.ffn_down_exps.weight | |
| 653: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.35.ffn_down_shexp.weight | |
| 654: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_gate_exps.weight | |
| 655: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight | |
| 656: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight | |
| 657: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_gate_shexp.weight | |
| 658: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_up_exps.weight | |
| 659: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_up_shexp.weight | |
| 660: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight | |
| 661: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.36.attn_gate.weight | |
| 662: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight | |
| 663: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.36.attn_qkv.weight | |
| 664: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.36.ffn_down_exps.weight | |
| 665: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.36.ffn_down_shexp.weight | |
| 666: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_gate_exps.weight | |
| 667: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight | |
| 668: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight | |
| 669: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_gate_shexp.weight | |
| 670: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_up_exps.weight | |
| 671: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_up_shexp.weight | |
| 672: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight | |
| 673: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a | |
| 674: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_alpha.weight | |
| 675: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_beta.weight | |
| 676: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight | |
| 677: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias | |
| 678: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight | |
| 679: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.36.ssm_out.weight | |
| 680: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.37.attn_gate.weight | |
| 681: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight | |
| 682: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.37.attn_qkv.weight | |
| 683: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.37.ffn_down_exps.weight | |
| 684: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.37.ffn_down_shexp.weight | |
| 685: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_gate_exps.weight | |
| 686: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight | |
| 687: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight | |
| 688: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_gate_shexp.weight | |
| 689: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_up_exps.weight | |
| 690: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_up_shexp.weight | |
| 691: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight | |
| 692: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a | |
| 693: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_alpha.weight | |
| 694: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_beta.weight | |
| 695: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight | |
| 696: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias | |
| 697: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight | |
| 698: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.37.ssm_out.weight | |
| 699: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.38.attn_gate.weight | |
| 700: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight | |
| 701: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.38.attn_qkv.weight | |
| 702: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.38.ffn_down_exps.weight | |
| 703: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.38.ffn_down_shexp.weight | |
| 704: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_gate_exps.weight | |
| 705: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight | |
| 706: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight | |
| 707: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_gate_shexp.weight | |
| 708: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_up_exps.weight | |
| 709: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_up_shexp.weight | |
| 710: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight | |
| 711: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a | |
| 712: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_alpha.weight | |
| 713: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_beta.weight | |
| 714: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight | |
| 715: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias | |
| 716: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight | |
| 717: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.38.ssm_out.weight | |
| 718: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_k.weight | |
| 719: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight | |
| 720: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight | |
| 721: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.39.attn_output.weight | |
| 722: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.39.attn_q.weight | |
| 723: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight | |
| 724: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_v.weight | |
| 725: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.39.ffn_down_exps.weight | |
| 726: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.39.ffn_down_shexp.weight | |
| 727: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_gate_exps.weight | |
| 728: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight | |
| 729: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight | |
| 730: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_gate_shexp.weight | |
| 731: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_up_exps.weight | |
| 732: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_up_shexp.weight | |
| 733: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight | |
| 734: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.40.attn_gate.weight | |
| 735: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight | |
| 736: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.40.attn_qkv.weight | |
| 737: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.40.ffn_down_exps.weight | |
| 738: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.40.ffn_down_shexp.weight | |
| 739: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_gate_exps.weight | |
| 740: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight | |
| 741: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight | |
| 742: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_gate_shexp.weight | |
| 743: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_up_exps.weight | |
| 744: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_up_shexp.weight | |
| 745: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight | |
| 746: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a | |
| 747: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_alpha.weight | |
| 748: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_beta.weight | |
| 749: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight | |
| 750: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias | |
| 751: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight | |
| 752: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.40.ssm_out.weight | |
| 753: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.41.attn_gate.weight | |
| 754: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight | |
| 755: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.41.attn_qkv.weight | |
| 756: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.41.ffn_down_exps.weight | |
| 757: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.41.ffn_down_shexp.weight | |
| 758: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_gate_exps.weight | |
| 759: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight | |
| 760: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight | |
| 761: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_gate_shexp.weight | |
| 762: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_up_exps.weight | |
| 763: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_up_shexp.weight | |
| 764: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight | |
| 765: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a | |
| 766: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_alpha.weight | |
| 767: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_beta.weight | |
| 768: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight | |
| 769: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias | |
| 770: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight | |
| 771: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.41.ssm_out.weight | |
| 772: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.42.attn_gate.weight | |
| 773: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight | |
| 774: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.42.attn_qkv.weight | |
| 775: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.42.ffn_down_exps.weight | |
| 776: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.42.ffn_down_shexp.weight | |
| 777: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_gate_exps.weight | |
| 778: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight | |
| 779: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight | |
| 780: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_gate_shexp.weight | |
| 781: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_up_exps.weight | |
| 782: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_up_shexp.weight | |
| 783: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight | |
| 784: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a | |
| 785: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_alpha.weight | |
| 786: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_beta.weight | |
| 787: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight | |
| 788: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias | |
| 789: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight | |
| 790: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.42.ssm_out.weight | |
| 791: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_k.weight | |
| 792: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight | |
| 793: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight | |
| 794: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.43.attn_output.weight | |
| 795: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.43.attn_q.weight | |
| 796: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight | |
| 797: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_v.weight | |
| 798: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.43.ffn_down_exps.weight | |
| 799: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.43.ffn_down_shexp.weight | |
| 800: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_gate_exps.weight | |
| 801: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight | |
| 802: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight | |
| 803: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_gate_shexp.weight | |
| 804: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_up_exps.weight | |
| 805: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_up_shexp.weight | |
| 806: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight | |
| 807: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.44.attn_gate.weight | |
| 808: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight | |
| 809: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.44.attn_qkv.weight | |
| 810: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.44.ffn_down_exps.weight | |
| 811: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.44.ffn_down_shexp.weight | |
| 812: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_gate_exps.weight | |
| 813: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight | |
| 814: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight | |
| 815: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_gate_shexp.weight | |
| 816: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_up_exps.weight | |
| 817: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_up_shexp.weight | |
| 818: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight | |
| 819: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a | |
| 820: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_alpha.weight | |
| 821: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_beta.weight | |
| 822: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight | |
| 823: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias | |
| 824: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight | |
| 825: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.44.ssm_out.weight | |
| 826: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.45.attn_gate.weight | |
| 827: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight | |
| 828: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.45.attn_qkv.weight | |
| 829: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.45.ffn_down_exps.weight | |
| 830: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.45.ffn_down_shexp.weight | |
| 831: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_gate_exps.weight | |
| 832: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight | |
| 833: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight | |
| 834: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_gate_shexp.weight | |
| 835: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_up_exps.weight | |
| 836: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_up_shexp.weight | |
| 837: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight | |
| 838: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a | |
| 839: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_alpha.weight | |
| 840: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_beta.weight | |
| 841: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight | |
| 842: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias | |
| 843: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight | |
| 844: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.45.ssm_out.weight | |
| 845: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.46.attn_gate.weight | |
| 846: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight | |
| 847: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.46.attn_qkv.weight | |
| 848: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.46.ffn_down_exps.weight | |
| 849: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.46.ffn_down_shexp.weight | |
| 850: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_gate_exps.weight | |
| 851: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight | |
| 852: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight | |
| 853: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_gate_shexp.weight | |
| 854: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_up_exps.weight | |
| 855: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_up_shexp.weight | |
| 856: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight | |
| 857: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a | |
| 858: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_alpha.weight | |
| 859: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_beta.weight | |
| 860: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight | |
| 861: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias | |
| 862: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight | |
| 863: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.46.ssm_out.weight | |
| 864: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_k.weight | |
| 865: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight | |
| 866: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight | |
| 867: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.47.attn_output.weight | |
| 868: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.47.attn_q.weight | |
| 869: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight | |
| 870: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_v.weight | |
| 871: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.47.ffn_down_exps.weight | |
| 872: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.47.ffn_down_shexp.weight | |
| 873: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_gate_exps.weight | |
| 874: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight | |
| 875: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight | |
| 876: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_gate_shexp.weight | |
| 877: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_up_exps.weight | |
| 878: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_up_shexp.weight | |
| 879: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight | |
| </details> |