Text Generation
MLX
Safetensors
qwen3_next
apple-silicon
qwen3-next
coding
mixture-of-experts
quantized
mixed-precision
optiq
4-bit precision
8-bit precision
automatosx
ax-engine
conversational
Instructions to use AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit") prompt = "Write a story about Einstein" messages = [{"role": "user", "content": prompt}] prompt = tokenizer.apply_chat_template( messages, add_generation_prompt=True ) text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit"
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit with Hermes Agent:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit"
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 AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit
Run Hermes
hermes
- OpenClaw new
How to use AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit"
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 "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit" \ --custom-provider-id mlx-lm \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
- MLX LM
How to use AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AutomatosX/AX-Qwen3-Coder-Next-MLX-OptiQ-4bit", "messages": [ {"role": "user", "content": "Hello"} ] }'
| [ | |
| { | |
| "layer_name": "lm_head", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00546584720723331 | |
| }, | |
| "param_count": 311164928, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0001811789904119602, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00087689898039874 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0015896713763747055 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00044233594720329467 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0003866669740091311 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.000519902074302081 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00046726479376957286 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0015319683431395485, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0009970010384373988 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0006898588944750372 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.000378026442679887 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.47.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0009673482066621849 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.00043826462933793664, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005482422323742261 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.000585554851568304 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005205812140047783 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005259119376811819 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005636888490698766 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005691807382390834 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0016299924997535224, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0013121363129660797 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0004549955301627051 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.46.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001232918316721528 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0004910269999527372, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0006056272814021213 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.000662573301572896 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005977782335927865 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0007479680231578337 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0007716155766199032 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0007242341719878217 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.002146355339694613, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0011942335452962045 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0005906730202696053 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.45.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0012439823755509376 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0007133473106174885, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0007204592196406642 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0008755783965170849 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0006210746008340114 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.000983773055243849 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0010371999160270207 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0009049700092873536 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0024624387491106368, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0015454338014630291 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0007653660904907156 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.44.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0014417152075717847 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0007744313740355816, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0009393947063169131 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0010768429419840686 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0008377297441863144 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0010964083982495747 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0010202171264002875 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0011413340374322918 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0033793977951669754, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002702764031710103 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0017084380015148781 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0011191138861856114 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.43.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0014598523788057112 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0009925571054433628, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0011605675669367581 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0018117218254095253 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001172308443832056 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0015623406288796104 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00130574569387439 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0013518400919565465 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0029391975913313217, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0017956645048495072 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0011235147382346138 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.42.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002219276189862285 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0011014955040688317, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001253751875386418 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0012832630854973104 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0012258604571494895 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0015417903002041082 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0016429382061081317 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0015326397333410569 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0030040774484708286, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0017300844774581492 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0013138200107884284 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.41.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0019365232534861814 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0012782128954616685, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0014563855996433024 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0014926813782949466 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0013535707779131674 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0017617459622366975 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0016695782178430818 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0016951369301144343 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0033831604887382127, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002003027924123065 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0014643894125280592 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.40.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002364053131411007 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.001446501639293274, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001575188147398876 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001775789626359862 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0015565581913203157 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0019679781325976364 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001903772181928313 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001958755795688679 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0034947048746592677, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0022330831889121328 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0020966503264692924 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0017853173170199927 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.39.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002157975022176591 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0016617042371459927, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.001788420772451597 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0018763207917800173 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0018028927515842952 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00211460222756917 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0020056302140195235 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002035529483691789 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.003621493626269512, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0020221223191280537 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0017770422612860177 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.38.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0024485136527800933 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0017955446419364307, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00191835029302941 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002024596572785716 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0018729715350976524 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0022996737267628 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0021713513148521693 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0021041337956072916 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.004091399455016169, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002346425324503798 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0018104871329948462 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.37.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002669086587654116 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0018940380444594969, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002158472236866752 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002250783230313876 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002060351244532891 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002607721893582493 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0025067820267092125 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0025180679634407475 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0041634838610965135, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002644306846680896 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002127183021608895 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.36.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003041435554526591 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0021315743118369332, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0023513569540227763 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00261818288709037 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0022630117625036896 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0029103495204860033 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0026857099728658795 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0027523967655724846 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.004539914623213311, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003231826927124833 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0032008243239639946 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0025570961685540774 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.35.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002896230924913349 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.002297268777814073, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0026993594292434864 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0027452579136782638 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0024244497229422755 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0029562374596328786 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00277543455013074 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00286523503503607 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.00521997369651217, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003193213837221265 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002630184574324327 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.34.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0036525255200103857 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.002746699678633983, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0027895902336846725 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0029053816663993834 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002501191748403168 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0031272211781470105 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0031219236722487644 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0032260748290961296 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.005147328270444025, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0033409177121939138 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002774385551068311 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.33.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0038971262256382033 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0028969740136138475, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0031035247181231775 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002979815437962922 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.002915382935801366 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003368606465907457 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0036279306029124805 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00359268567020384 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.0054580638752668165, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0036926528555341065 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0031067129117824757 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.32.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004197506471731079 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0030901972932042554, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0033126247726613656 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00331102798130208 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0028822653912357055 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003875695178673292 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0038103191618574783 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0038147873856360093 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.00543140152876731, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003964414742464821 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004204695069347508 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0034591444976589023 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.31.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004151555299661898 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.003191108728060499, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003548390481834455 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0036089261120650917 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0034522233545430936 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003841529044924149 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003770305927901063 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0038622769401020682 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.005873478900563593, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004227041160144533 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003324077117819494 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.30.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004752254715034117 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0035896663054397018, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0035883158537520408 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003742401638495115 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0034946068941887156 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004219321068376303 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004259144703003888 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004165746400152178 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.006541496540497367, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0044880598967817304 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0036810627813489796 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.29.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004826595235499553 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.00362065994219544, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.003921926690964028 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0039680428671999834 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0036683329550821022 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0043618389148226315 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004315103074380507 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004349304557157059 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.006787006728700362, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004702370313073819 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0037491704545876323 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.28.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0051801369748621555 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.003750991881436979, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004106528702929306 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004144503269344568 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004088555853134797 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004885390294172491 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00455872366001131 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0049130060215247795 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.007291974553178686, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00555882364763723 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005674804478379277 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004737999302354486 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.27.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005359278341832881 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.004103830414048086, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004563357242053219 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004600786213510825 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004448893041020104 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00518409363576211 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004802735384146217 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005048779246862978 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.008079974698678901, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00514275149422853 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004920122419813803 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.26.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005918499732312436 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.004627415449552548, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004856845213604781 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004881828169648846 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.004638931318671287 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005686172774100366 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0053267386635222165 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005722651037406952 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.009340248730344078, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006284646408554788 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005060244996760351 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.25.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006544227498428275 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.005057090580521617, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005316806911044599 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005873421080953752 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005097836023196578 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006277923326706514 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005662061237671878 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005897963734848115 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.009583479351325272, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006754006913979538 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005299228282334904 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.24.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007472097684512846 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.005364448579105859, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005531614619637064 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005941575296067943 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.005747957458273352 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006524281270685606 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006318350400154789 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006365798185773504 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.010608130556647666, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008208236598875374 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00739130225944488 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006829266402443561 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.23.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007633909597643651 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.00617599854740547, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006110824887097503 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006009446524937327 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006198184875150521 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0070419801292397706 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0064737989838855965 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006815817270156306 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.010461516358191147, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007974229263102947 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006750545023047986 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.22.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009198193957369464 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.006172639778621185, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006532026910766338 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00721957237207486 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006716253750103836 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007162697574434181 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006981045803210388 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007448724689311348 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.011110041804689294, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008667696058788957 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006610841888080661 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.21.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00952446710531755 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.0069478996738325804, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0074366811362172784 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.006943494290074644 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0067551254711967585 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008092523431211399 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007611915270293442 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008078320082859136 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.011131369343881184, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00876020211338376 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007388061237482664 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.20.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0099560285161715 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.007256736600538716, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0075626892648870125 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007552810381942739 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00718522864432695 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008595826366217807 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00823559146859528 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008514910165104084 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.011962555969754854, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009170711811748333 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00948034289952678 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00827406374446582 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.19.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009781325609462025 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.007757856593040439, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008020264431252144 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.007868040115378486 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008096922431529189 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008891638106433675 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008893419241455073 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008583348394798426 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.012067314916445563, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009313053917139769 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00792201085035534 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.18.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010337958544065865 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.00800801339210011, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008354605146450922 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008346605425079664 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00789218878101868 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0096804217416017 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009081653120423047 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00961258050908024 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.012435313129875189, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00969749286499185 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008220665801976187 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.17.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01077947385783773 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.00898167995425562, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008234198758145794 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00856234620732721 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008658537199759545 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009765561121942786 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00955390140976912 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00922697524462516 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.011782195152287992, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010321959528179528 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.008864662149183763 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.16.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010752089612651616 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.009154927955629924, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009189329032475749 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009195887905661948 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009162585886466937 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010390287488310909 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009711528284242377 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010309005903157717 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.013503166526788846, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011138641464640386 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011259745757949228 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009398004641601196 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.15.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0112590963835828 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.009654443934171772, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009863078341974566 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.00967700000910554 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009742690696536252 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010233040569194904 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01022266013508973 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010732679739400433 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01367227423179429, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01100488739515034 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01022967364891277 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.14.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01245528188883327 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.010582786043717837, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010168050968786702 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01001773014528832 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.009968309891216146 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011261625166904802 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010767707319852585 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010916094227771586 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01399673714574116, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011629761046303125 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010364441375713795 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.13.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012238817202160135 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.010277952591422945, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010277141176629812 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010602660113363527 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010807378469811132 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012377719894478409 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011131742144546783 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010933864900531868 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.014976006457194066, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011991277909449613 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010463360308979949 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.12.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013422283062633747 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.010844901742530055, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011393746521207504 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011229697129844377 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.010468666019733064 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012109854117928384 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011566853941379426 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011979735451556431 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.014990527405946827, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01275160105433315 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015201601044585308 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012038252505590208 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.11.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013188726123189554 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.011288866294004643, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011611835070652887 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012178075737513913 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.011832942793262191 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012095854433331018 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01274094315400968 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012829635612433776 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01626087084878236, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013361044965373972 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012351946274672324 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.10.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015012034185929224 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.011678688800505673, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012375947214119757 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012288592877060486 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01256927099893801 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01338555481925141 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012540935538709164 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012916975412129736 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01662118696064378, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014035383860270182 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012819181167287752 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.9.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015574557250753665 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.012599239038536325, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013240655724075623 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.012855073883353422 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013038330798735842 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013493624350909764 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013379455165704712 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013178177721177539 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01652742245156939, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014525910800633332 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014052762446226552 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.8.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01691254019776049 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.01293456334193858, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013468108120529601 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013862467652264362 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013406028689738983 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013866370340110734 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014211450363897407 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013236384809715673 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01609512370972273, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015192151438289633 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01611088200782736 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014883832996323084 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.7.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014775762363569811 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.013842910673702136, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01374226551464138 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015329888090491295 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013457816676236689 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014617457006049031 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014744472709329179 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013872471987269819 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.016978700946007546, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015879868374516565 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014073359741208455 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.6.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016584908017345395 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.013318267408370351, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.013795014693945026 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014510109923624745 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014729883502392719 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014678579987958074 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014759288256755099 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014945248481429493 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01711107642040588, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014898032823111862 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014382564753759652 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.5.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017053854671151687 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.014288380965202426, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014899879752192646 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015110610518604517 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01435574131513325 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016535441546390455 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015712454575502004 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01456842960518164 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.018017263452444848, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015937547102415312 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015152021961209053 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.4.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01661324539842705 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.01538781247412165, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015848533192183822 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015900589545102168 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.014709234674228355 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.0161793943552766 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.015927003318211064 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016266208646508556 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.018574310747984175, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.self_attn.o_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017166202393127605 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.self_attn.v_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.02210300312920784 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.self_attn.k_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01649043273452359 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.3.self_attn.q_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017937579153416056 | |
| }, | |
| "param_count": 16777216, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.015848530747462064, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017437926658506814 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016578349061698344 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016357797489035875 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01727279481322815 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016195804220236216 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01615279316320084 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01900803267684144, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017819210469800357 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017294775665504858 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.2.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.020012144804544125 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.016524562825604033, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016842761435934033 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016998056855906423 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01661954979257037 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017476482074319694 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01659977853220577 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016613098239758983 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01898803311632946, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01730258480529301 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.018019680787498753 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.1.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01960670785047114 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.shared_expert_gate", | |
| "sensitivities": { | |
| "4": 0.016866121945592265, | |
| "8": 0.0 | |
| }, | |
| "param_count": 2048, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.shared_expert.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.018234637934559334 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.shared_expert.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017515672865556553 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.shared_expert.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016906839252139132 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.switch_mlp.down_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.018159386556362733 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.switch_mlp.up_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.016372410124555852 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.switch_mlp.gate_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01711415994213894 | |
| }, | |
| "param_count": 536870912, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.mlp.gate", | |
| "sensitivities": { | |
| "4": 0.01868543986347504, | |
| "8": 0.0 | |
| }, | |
| "param_count": 1048576, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.linear_attn.out_proj", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.019266100542154163 | |
| }, | |
| "param_count": 8388608, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.linear_attn.in_proj_ba", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.017209987262807164 | |
| }, | |
| "param_count": 131072, | |
| "metadata": {} | |
| }, | |
| { | |
| "layer_name": "model.layers.0.linear_attn.in_proj_qkvz", | |
| "sensitivities": { | |
| "4": 0.0, | |
| "8": 0.01821280693790565 | |
| }, | |
| "param_count": 25165824, | |
| "metadata": {} | |
| } | |
| ] |