Text Generation
GGUF
English
Korean
motif
motif-3
mixture-of-experts
q8_0
long-context
conversational
Instructions to use Baekpica/Motif-3-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use Baekpica/Motif-3-GGUF with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf Baekpica/Motif-3-GGUF:Q8_0 # Run inference directly in the terminal: llama cli -hf Baekpica/Motif-3-GGUF:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Baekpica/Motif-3-GGUF:Q8_0 # Run inference directly in the terminal: llama cli -hf Baekpica/Motif-3-GGUF:Q8_0
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf Baekpica/Motif-3-GGUF:Q8_0 # Run inference directly in the terminal: ./llama-cli -hf Baekpica/Motif-3-GGUF:Q8_0
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf Baekpica/Motif-3-GGUF:Q8_0 # Run inference directly in the terminal: ./build/bin/llama-cli -hf Baekpica/Motif-3-GGUF:Q8_0
Use Docker
docker model run hf.co/Baekpica/Motif-3-GGUF:Q8_0
- LM Studio
- Jan
- vLLM
How to use Baekpica/Motif-3-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Baekpica/Motif-3-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Baekpica/Motif-3-GGUF", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Baekpica/Motif-3-GGUF:Q8_0
- Ollama
How to use Baekpica/Motif-3-GGUF with Ollama:
ollama run hf.co/Baekpica/Motif-3-GGUF:Q8_0
- Unsloth Studio
How to use Baekpica/Motif-3-GGUF with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Baekpica/Motif-3-GGUF to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Baekpica/Motif-3-GGUF to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Baekpica/Motif-3-GGUF to start chatting
- Pi
How to use Baekpica/Motif-3-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Baekpica/Motif-3-GGUF:Q8_0
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "Baekpica/Motif-3-GGUF:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- OpenClaw new
How to use Baekpica/Motif-3-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Baekpica/Motif-3-GGUF:Q8_0
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 "Baekpica/Motif-3-GGUF:Q8_0" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
- Docker Model Runner
How to use Baekpica/Motif-3-GGUF with Docker Model Runner:
docker model run hf.co/Baekpica/Motif-3-GGUF:Q8_0
- Lemonade
How to use Baekpica/Motif-3-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Baekpica/Motif-3-GGUF:Q8_0
Run and chat with the model
lemonade run user.Motif-3-GGUF-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use Baekpica/Motif-3-GGUF with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Baekpica/Motif-3-GGUF:Q8_0
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 Baekpica/Motif-3-GGUF:Q8_0
Run Hermes
hermes
- Atomic Chat
| { | |
| "status": "ok", | |
| "source_model": "Motif-Technologies/Motif-3", | |
| "source_revision": "ccceb1a5fd7b5eb32e47841216b3caf5666c07bc", | |
| "gguf": "/motif3/artifacts/Motif-3-Q8_0.gguf", | |
| "sample_count": 57, | |
| "samples": [ | |
| { | |
| "tensor": "token_embd.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.00018674450347218728, | |
| "actual_rms": 0.00018678366576871486, | |
| "rmse": 1.344796346846577e-06, | |
| "relative_rmse": 0.007201263340245319, | |
| "cosine": 0.9999740123748779, | |
| "max_abs": 4.947185516357422e-06, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "token_embd.weight", | |
| "prefix": [ | |
| 12345 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.02951657176744764, | |
| "actual_rms": 0.02951927877202853, | |
| "rmse": 0.00015771726905359834, | |
| "relative_rmse": 0.005343346452840329, | |
| "cosine": 0.9999857544898987, | |
| "max_abs": 0.00043773651123046875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "token_embd.weight", | |
| "prefix": [ | |
| 220159 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03148364443955599, | |
| "actual_rms": 0.031486321385359124, | |
| "rmse": 0.00016766861874471423, | |
| "relative_rmse": 0.005325578462385877, | |
| "cosine": 0.999985933303833, | |
| "max_abs": 0.000461578369140625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "output.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.027632501134147378, | |
| "actual_rms": 0.027630156925352672, | |
| "rmse": 0.00015054950934951898, | |
| "relative_rmse": 0.005448276600755283, | |
| "cosine": 0.9999851584434509, | |
| "max_abs": 0.00041294097900390625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "output.weight", | |
| "prefix": [ | |
| 12345 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.04606739213896064, | |
| "actual_rms": 0.04606955523233037, | |
| "rmse": 0.0002423105213004472, | |
| "relative_rmse": 0.005259914009664931, | |
| "cosine": 0.9999861717224121, | |
| "max_abs": 0.0006389617919921875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "output.weight", | |
| "prefix": [ | |
| 220159 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.0286056510743836, | |
| "actual_rms": 0.02860854596564918, | |
| "rmse": 0.00015336027527253064, | |
| "relative_rmse": 0.0053611880699287765, | |
| "cosine": 0.9999855756759644, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.0.attn_q_a.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.02300794076249149, | |
| "actual_rms": 0.023009376989890893, | |
| "rmse": 0.00012059711436681212, | |
| "relative_rmse": 0.005241543152936771, | |
| "cosine": 0.9999862313270569, | |
| "max_abs": 0.0003342628479003906, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.0.attn_q_a.weight", | |
| "prefix": [ | |
| 511 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.01936402488499073, | |
| "actual_rms": 0.01936352635090829, | |
| "rmse": 0.00010468325362055778, | |
| "relative_rmse": 0.005406068947045143, | |
| "cosine": 0.9999854564666748, | |
| "max_abs": 0.0002918243408203125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.0.attn_q_a.weight", | |
| "prefix": [ | |
| 1023 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.019656028487519815, | |
| "actual_rms": 0.0196586261999784, | |
| "rmse": 0.00010478324928664917, | |
| "relative_rmse": 0.0053308454122957295, | |
| "cosine": 0.9999857544898987, | |
| "max_abs": 0.00030517578125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.1.ffn_gate.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.032726405515143504, | |
| "actual_rms": 0.03272447701515597, | |
| "rmse": 0.00017355419363094128, | |
| "relative_rmse": 0.005303185329981701, | |
| "cosine": 0.9999859929084778, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.1.ffn_gate.weight", | |
| "prefix": [ | |
| 6143 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.035869132013714944, | |
| "actual_rms": 0.03586821037491973, | |
| "rmse": 0.00019189349907859844, | |
| "relative_rmse": 0.005349822766974844, | |
| "cosine": 0.9999858736991882, | |
| "max_abs": 0.0005922317504882812, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.1.ffn_gate.weight", | |
| "prefix": [ | |
| 12287 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.0322896923360983, | |
| "actual_rms": 0.032289231256291, | |
| "rmse": 0.00017159510924011516, | |
| "relative_rmse": 0.005314237975822402, | |
| "cosine": 0.999985933303833, | |
| "max_abs": 0.00048065185546875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.mhc_attn.proj_pre.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "BF16", | |
| "elements": 16384, | |
| "reference_rms": 0.03520855791244703, | |
| "actual_rms": 0.03520855791244703, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0000001192092896, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.mhc_attn.proj_pre.weight", | |
| "prefix": [ | |
| 1 | |
| ], | |
| "type": "BF16", | |
| "elements": 16384, | |
| "reference_rms": 0.05739358058838267, | |
| "actual_rms": 0.05739358058838267, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.mhc_attn.proj_pre.weight", | |
| "prefix": [ | |
| 3 | |
| ], | |
| "type": "BF16", | |
| "elements": 16384, | |
| "reference_rms": 0.03147771434818792, | |
| "actual_rms": 0.03147771434818792, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0000001192092896, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_inp.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "F32", | |
| "elements": 4096, | |
| "reference_rms": 0.05070282476545178, | |
| "actual_rms": 0.05070282476545178, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0000001192092896, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_inp.weight", | |
| "prefix": [ | |
| 191 | |
| ], | |
| "type": "F32", | |
| "elements": 4096, | |
| "reference_rms": 0.05538568173189653, | |
| "actual_rms": 0.05538568173189653, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_inp.weight", | |
| "prefix": [ | |
| 383 | |
| ], | |
| "type": "F32", | |
| "elements": 4096, | |
| "reference_rms": 0.047852772350155684, | |
| "actual_rms": 0.047852772350155684, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 0.9999998807907104, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_polynorm_exps.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "F32", | |
| "elements": 3, | |
| "reference_rms": 0.3379480757777858, | |
| "actual_rms": 0.3379480757777858, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 0.9999999403953552, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_polynorm_exps.weight", | |
| "prefix": [ | |
| 173 | |
| ], | |
| "type": "F32", | |
| "elements": 3, | |
| "reference_rms": 0.7689799804727289, | |
| "actual_rms": 0.7689799804727289, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_polynorm_exps.weight", | |
| "prefix": [ | |
| 383 | |
| ], | |
| "type": "F32", | |
| "elements": 3, | |
| "reference_rms": 0.3511293639412858, | |
| "actual_rms": 0.3511293639412858, | |
| "rmse": 0.0, | |
| "relative_rmse": 0.0, | |
| "cosine": 1.0, | |
| "max_abs": 0.0, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_shexp.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.027594728575466354, | |
| "actual_rms": 0.02759494266163564, | |
| "rmse": 0.000151094583406654, | |
| "relative_rmse": 0.005475487211024342, | |
| "cosine": 0.9999850392341614, | |
| "max_abs": 0.000431060791015625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_shexp.weight", | |
| "prefix": [ | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.05475560568016764, | |
| "actual_rms": 0.05475493640810258, | |
| "rmse": 0.00029380319181464265, | |
| "relative_rmse": 0.005365718964570919, | |
| "cosine": 0.9999855756759644, | |
| "max_abs": 0.000774383544921875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_shexp.weight", | |
| "prefix": [ | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.026988755135910345, | |
| "actual_rms": 0.02699115113853432, | |
| "rmse": 0.0001447045171206727, | |
| "relative_rmse": 0.005361659564954652, | |
| "cosine": 0.9999857544898987, | |
| "max_abs": 0.00039768218994140625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03464500391672497, | |
| "actual_rms": 0.03464237599930324, | |
| "rmse": 0.00018231131948090465, | |
| "relative_rmse": 0.00526226869303061, | |
| "cosine": 0.9999862313270569, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.036091132943366765, | |
| "actual_rms": 0.0360949002271766, | |
| "rmse": 0.00019519432521468373, | |
| "relative_rmse": 0.005408373450647211, | |
| "cosine": 0.9999853372573853, | |
| "max_abs": 0.000530242919921875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03666984441177153, | |
| "actual_rms": 0.03666606015255815, | |
| "rmse": 0.00020248338872520265, | |
| "relative_rmse": 0.005521795687254203, | |
| "cosine": 0.9999846816062927, | |
| "max_abs": 0.0005254745483398438, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_up_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03528059185259768, | |
| "actual_rms": 0.03528256167127104, | |
| "rmse": 0.00019231579183457754, | |
| "relative_rmse": 0.005451036440603745, | |
| "cosine": 0.9999852180480957, | |
| "max_abs": 0.0005350112915039062, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_up_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.036561867199756994, | |
| "actual_rms": 0.03656733719532561, | |
| "rmse": 0.0001957124468829903, | |
| "relative_rmse": 0.005352911704801857, | |
| "cosine": 0.9999856948852539, | |
| "max_abs": 0.00052642822265625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_up_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03661888348702348, | |
| "actual_rms": 0.03662230157143238, | |
| "rmse": 0.00019617030812944038, | |
| "relative_rmse": 0.0053570805401250594, | |
| "cosine": 0.9999857544898987, | |
| "max_abs": 0.00048065185546875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_down_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03557324893322653, | |
| "actual_rms": 0.03557910703037354, | |
| "rmse": 0.0001924292563404364, | |
| "relative_rmse": 0.0054093809846168265, | |
| "cosine": 0.9999853372573853, | |
| "max_abs": 0.000476837158203125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_down_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 2047 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03494629775795011, | |
| "actual_rms": 0.034944756632762405, | |
| "rmse": 0.00019300736649410872, | |
| "relative_rmse": 0.005522970353853879, | |
| "cosine": 0.9999847412109375, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.2.ffn_down_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 4095 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03585768685490932, | |
| "actual_rms": 0.035872100308727256, | |
| "rmse": 0.00018503423764218237, | |
| "relative_rmse": 0.0051602390971532824, | |
| "cosine": 0.9999867677688599, | |
| "max_abs": 0.00041675567626953125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.037376825232304164, | |
| "actual_rms": 0.03738316725074043, | |
| "rmse": 0.0001977775648283887, | |
| "relative_rmse": 0.005291449008821993, | |
| "cosine": 0.9999859929084778, | |
| "max_abs": 0.000545501708984375, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03366760738348695, | |
| "actual_rms": 0.03366837120918713, | |
| "rmse": 0.00017767534707319914, | |
| "relative_rmse": 0.005277338096806489, | |
| "cosine": 0.9999860525131226, | |
| "max_abs": 0.0005645751953125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.036582530796030126, | |
| "actual_rms": 0.03657706890176942, | |
| "rmse": 0.00018839152184432228, | |
| "relative_rmse": 0.005149767327326796, | |
| "cosine": 0.9999867677688599, | |
| "max_abs": 0.000560760498046875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_up_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03706375269148631, | |
| "actual_rms": 0.03705983688372809, | |
| "rmse": 0.00019618223686577739, | |
| "relative_rmse": 0.0052931023606479334, | |
| "cosine": 0.9999861121177673, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_up_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.0343739317524774, | |
| "actual_rms": 0.03437293286033201, | |
| "rmse": 0.00018310288519838628, | |
| "relative_rmse": 0.005326794924621611, | |
| "cosine": 0.9999858140945435, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_up_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03604555302803698, | |
| "actual_rms": 0.03604382987635977, | |
| "rmse": 0.00019876158122676902, | |
| "relative_rmse": 0.005514177603882734, | |
| "cosine": 0.9999846816062927, | |
| "max_abs": 0.0005254745483398438, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_down_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 4095 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03520202934293861, | |
| "actual_rms": 0.035201928586848175, | |
| "rmse": 0.00018594506622496176, | |
| "relative_rmse": 0.005282225760721991, | |
| "cosine": 0.9999861121177673, | |
| "max_abs": 0.0004515647888183594, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_down_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03476402951926834, | |
| "actual_rms": 0.03476620116374905, | |
| "rmse": 0.0001827925356188329, | |
| "relative_rmse": 0.005258094016906704, | |
| "cosine": 0.9999861121177673, | |
| "max_abs": 0.0005331039428710938, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.26.ffn_down_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 2047 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03707234542460301, | |
| "actual_rms": 0.03706323616519223, | |
| "rmse": 0.00019314731658339425, | |
| "relative_rmse": 0.005210010706665791, | |
| "cosine": 0.999986469745636, | |
| "max_abs": 0.0004425048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03565642486316714, | |
| "actual_rms": 0.03565660268942862, | |
| "rmse": 0.0001954262629009942, | |
| "relative_rmse": 0.005480814850365671, | |
| "cosine": 0.9999849200248718, | |
| "max_abs": 0.000537872314453125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03351001877642548, | |
| "actual_rms": 0.03350526847743235, | |
| "rmse": 0.00018022280763398355, | |
| "relative_rmse": 0.005378176862161936, | |
| "cosine": 0.9999855756759644, | |
| "max_abs": 0.0005283355712890625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_gate_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03490973420979067, | |
| "actual_rms": 0.03491389790997883, | |
| "rmse": 0.00018846634694944325, | |
| "relative_rmse": 0.005398676077476024, | |
| "cosine": 0.9999854564666748, | |
| "max_abs": 0.00048065185546875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_up_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 639 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03729690038613671, | |
| "actual_rms": 0.03730076735196125, | |
| "rmse": 0.00019965616388280353, | |
| "relative_rmse": 0.005353157013471712, | |
| "cosine": 0.9999856352806091, | |
| "max_abs": 0.0005006790161132812, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_up_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 1279 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03427546199804014, | |
| "actual_rms": 0.034277621004435946, | |
| "rmse": 0.00018129450468552238, | |
| "relative_rmse": 0.005289338031268222, | |
| "cosine": 0.9999860525131226, | |
| "max_abs": 0.0004544258117675781, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_up_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.036256988857906, | |
| "actual_rms": 0.036256534109006464, | |
| "rmse": 0.00019002362362926614, | |
| "relative_rmse": 0.005241020548451603, | |
| "cosine": 0.9999862909317017, | |
| "max_abs": 0.0005645751953125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_down_exps.weight", | |
| "prefix": [ | |
| 0, | |
| 2047 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03776252033481507, | |
| "actual_rms": 0.03776077907678719, | |
| "rmse": 0.00020484512219906933, | |
| "relative_rmse": 0.005424561718414034, | |
| "cosine": 0.9999852180480957, | |
| "max_abs": 0.0005044937133789062, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_down_exps.weight", | |
| "prefix": [ | |
| 173, | |
| 4095 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03664056389780987, | |
| "actual_rms": 0.0366335996590639, | |
| "rmse": 0.00020372485948886047, | |
| "relative_rmse": 0.005560090725051254, | |
| "cosine": 0.9999844431877136, | |
| "max_abs": 0.0005435943603515625, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "blk.52.ffn_down_exps.weight", | |
| "prefix": [ | |
| 383, | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 1280, | |
| "reference_rms": 0.03705767274001027, | |
| "actual_rms": 0.0370657060228314, | |
| "rmse": 0.0002035704855420061, | |
| "relative_rmse": 0.005493342417108023, | |
| "cosine": 0.9999849200248718, | |
| "max_abs": 0.0005168914794921875, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "mtp.0.attn_q_a.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.04033184596081668, | |
| "actual_rms": 0.04033411141545098, | |
| "rmse": 0.00021475567005442274, | |
| "relative_rmse": 0.005324717104767851, | |
| "cosine": 0.9999858140945435, | |
| "max_abs": 0.0005807876586914062, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "mtp.0.attn_q_a.weight", | |
| "prefix": [ | |
| 511 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03151520431168361, | |
| "actual_rms": 0.031516813732093744, | |
| "rmse": 0.00017114045762441074, | |
| "relative_rmse": 0.005430409269501831, | |
| "cosine": 0.9999852180480957, | |
| "max_abs": 0.0006437301635742188, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "mtp.0.attn_q_a.weight", | |
| "prefix": [ | |
| 1023 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03464788350487792, | |
| "actual_rms": 0.03464914189827983, | |
| "rmse": 0.00019678545437185752, | |
| "relative_rmse": 0.005679580813187419, | |
| "cosine": 0.9999839067459106, | |
| "max_abs": 0.00086212158203125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "mtp.0.ffn_gate.weight", | |
| "prefix": [ | |
| 0 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03103434304219069, | |
| "actual_rms": 0.031035592599081598, | |
| "rmse": 0.00017024359918214183, | |
| "relative_rmse": 0.005485651781018802, | |
| "cosine": 0.9999850392341614, | |
| "max_abs": 0.0004439353942871094, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "mtp.0.ffn_gate.weight", | |
| "prefix": [ | |
| 6143 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.02645437991033191, | |
| "actual_rms": 0.02645221953564838, | |
| "rmse": 0.00014374269109753297, | |
| "relative_rmse": 0.005433606517512566, | |
| "cosine": 0.9999852776527405, | |
| "max_abs": 0.00048828125, | |
| "finite": true, | |
| "nonzero": true | |
| }, | |
| { | |
| "tensor": "mtp.0.ffn_gate.weight", | |
| "prefix": [ | |
| 12287 | |
| ], | |
| "type": "Q8_0", | |
| "elements": 4096, | |
| "reference_rms": 0.03047388956791306, | |
| "actual_rms": 0.0304749088533367, | |
| "rmse": 0.00016391697839138346, | |
| "relative_rmse": 0.005378931955045769, | |
| "cosine": 0.9999855160713196, | |
| "max_abs": 0.0004711151123046875, | |
| "finite": true, | |
| "nonzero": true | |
| } | |
| ], | |
| "errors": [] | |
| } | |