Text Generation
Safetensors
GGUF
English
lfm2
heretic
uncensored
decensored
abliterated
coding-agent
tool-use
conversational
text-generation-inference
Instructions to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic 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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic 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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M # Run inference directly in the terminal: llama cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M # Run inference directly in the terminal: llama cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
Use Docker
docker model run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
- Ollama
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Ollama:
ollama run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
- Unsloth Studio
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic 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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic 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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic to start chatting
- Pi
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
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": "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Docker Model Runner:
docker model run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
- Lemonade
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
Run and chat with the model
lemonade run user.lfm2.5-2.6b-fable5-coding-agent-heretic-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
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 saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
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 "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M" \ --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"
Upload reproduce/reproduce.json with huggingface_hub
Browse files- reproduce/reproduce.json +283 -0
reproduce/reproduce.json
ADDED
|
@@ -0,0 +1,283 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"version": "2",
|
| 3 |
+
"timestamp": "2026-08-19T09:10:37",
|
| 4 |
+
"system": {
|
| 5 |
+
"python": {
|
| 6 |
+
"version": "3.12.13",
|
| 7 |
+
"implementation": "CPython",
|
| 8 |
+
"compiler": "GCC 11.4.0",
|
| 9 |
+
"environment": "System"
|
| 10 |
+
},
|
| 11 |
+
"os": {
|
| 12 |
+
"platform": "Linux-6.6.122+-x86_64-with-glibc2.35",
|
| 13 |
+
"machine": "x86_64"
|
| 14 |
+
},
|
| 15 |
+
"cpu": {
|
| 16 |
+
"brand": "Intel(R) Xeon(R) CPU @ 2.00GHz",
|
| 17 |
+
"vendor": "GenuineIntel",
|
| 18 |
+
"family": 6,
|
| 19 |
+
"model": 85,
|
| 20 |
+
"stepping": 3
|
| 21 |
+
},
|
| 22 |
+
"accelerators": {
|
| 23 |
+
"type": "CUDA",
|
| 24 |
+
"api_name": "CUDA Version",
|
| 25 |
+
"api_version": "12.8",
|
| 26 |
+
"driver_version": "580.82.07",
|
| 27 |
+
"devices": [
|
| 28 |
+
{
|
| 29 |
+
"name": "Tesla T4",
|
| 30 |
+
"vram_gb": 14.56
|
| 31 |
+
}
|
| 32 |
+
]
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"environment": {
|
| 36 |
+
"heretic": {
|
| 37 |
+
"version": "1.4.0",
|
| 38 |
+
"is_standard_pypi": true,
|
| 39 |
+
"metadata": {
|
| 40 |
+
"type": "pypi"
|
| 41 |
+
}
|
| 42 |
+
},
|
| 43 |
+
"pytorch_version": "2.11.0+cu128",
|
| 44 |
+
"requirements": {
|
| 45 |
+
"absl-py": "1.4.0",
|
| 46 |
+
"accelerate": "1.14.0",
|
| 47 |
+
"alembic": "1.19.1",
|
| 48 |
+
"annotated-doc": "0.0.5",
|
| 49 |
+
"annotated-types": "0.8.0",
|
| 50 |
+
"anyio": "4.14.2",
|
| 51 |
+
"bitsandbytes": "0.50.1",
|
| 52 |
+
"certifi": "2026.7.22",
|
| 53 |
+
"chardet": "5.2.0",
|
| 54 |
+
"charset-normalizer": "3.4.9",
|
| 55 |
+
"click": "8.4.2",
|
| 56 |
+
"colorama": "0.4.6",
|
| 57 |
+
"colorlog": "6.12.0",
|
| 58 |
+
"cuda-bindings": "12.9.7",
|
| 59 |
+
"cuda-pathfinder": "1.6.0",
|
| 60 |
+
"cuda-toolkit": "12.8.1",
|
| 61 |
+
"dataproperty": "1.1.1",
|
| 62 |
+
"datasets": "4.8.5",
|
| 63 |
+
"dill": "0.3.8",
|
| 64 |
+
"evaluate": "0.4.6",
|
| 65 |
+
"filelock": "3.32.3",
|
| 66 |
+
"fsspec": "2025.3.0",
|
| 67 |
+
"greenlet": "3.5.5",
|
| 68 |
+
"h11": "0.16.0",
|
| 69 |
+
"heretic-llm": "1.4.0",
|
| 70 |
+
"hf-xet": "1.6.0",
|
| 71 |
+
"httpcore": "1.0.9",
|
| 72 |
+
"httpx": "0.28.1",
|
| 73 |
+
"huggingface-hub": "1.27.0",
|
| 74 |
+
"idna": "3.18",
|
| 75 |
+
"immutabledict": "4.3.1",
|
| 76 |
+
"importlib-metadata": "9.0.0",
|
| 77 |
+
"jinja2": "3.1.6",
|
| 78 |
+
"joblib": "1.5.3",
|
| 79 |
+
"langdetect": "1.0.9",
|
| 80 |
+
"lm-eval": "0.4.12",
|
| 81 |
+
"lxml": "6.1.1",
|
| 82 |
+
"mako": "1.1.3",
|
| 83 |
+
"markdown-it-py": "4.2.0",
|
| 84 |
+
"markupsafe": "3.0.3",
|
| 85 |
+
"mbstrdecoder": "1.1.5",
|
| 86 |
+
"mdurl": "0.1.2",
|
| 87 |
+
"more-itertools": "10.8.0",
|
| 88 |
+
"mpmath": "1.3.0",
|
| 89 |
+
"multiprocess": "0.70.16",
|
| 90 |
+
"networkx": "3.6.1",
|
| 91 |
+
"nltk": "3.9.1",
|
| 92 |
+
"numpy": "2.5.2",
|
| 93 |
+
"nvidia-cublas-cu12": "12.8.4.1",
|
| 94 |
+
"nvidia-cudnn-cu12": "9.19.0.56",
|
| 95 |
+
"nvidia-cusparselt-cu12": "0.7.1",
|
| 96 |
+
"nvidia-nccl-cu12": "2.28.9",
|
| 97 |
+
"nvidia-nvshmem-cu12": "3.4.5",
|
| 98 |
+
"optuna": "4.9.0",
|
| 99 |
+
"packaging": "26.3",
|
| 100 |
+
"pandas": "2.2.3",
|
| 101 |
+
"pathvalidate": "3.3.1",
|
| 102 |
+
"peft": "0.20.0",
|
| 103 |
+
"pillow": "11.3.0",
|
| 104 |
+
"portalocker": "4.1.0",
|
| 105 |
+
"prompt-toolkit": "3.0.53",
|
| 106 |
+
"psutil": "7.2.2",
|
| 107 |
+
"py-cpuinfo": "9.0.0",
|
| 108 |
+
"pyarrow": "25.0.1",
|
| 109 |
+
"pydantic": "2.13.4",
|
| 110 |
+
"pydantic-core": "2.46.4",
|
| 111 |
+
"pydantic-settings": "2.15.0",
|
| 112 |
+
"pygments": "2.20.0",
|
| 113 |
+
"pytablewriter": "1.2.1",
|
| 114 |
+
"python-dateutil": "2.9.0.post0",
|
| 115 |
+
"python-dotenv": "1.2.2",
|
| 116 |
+
"pytz": "2025.2",
|
| 117 |
+
"pyyaml": "6.0.3",
|
| 118 |
+
"questionary": "2.1.1",
|
| 119 |
+
"regex": "2025.11.3",
|
| 120 |
+
"requests": "2.32.4",
|
| 121 |
+
"rich": "14.3.4",
|
| 122 |
+
"rouge-score": "0.1.2",
|
| 123 |
+
"sacrebleu": "2.6.0",
|
| 124 |
+
"safetensors": "0.8.0",
|
| 125 |
+
"scikit-learn": "1.6.1",
|
| 126 |
+
"scipy": "1.16.3",
|
| 127 |
+
"setuptools": "75.2.0",
|
| 128 |
+
"shellingham": "1.5.4",
|
| 129 |
+
"six": "1.17.0",
|
| 130 |
+
"sqlalchemy": "2.0.52",
|
| 131 |
+
"sqlitedict": "2.1.0",
|
| 132 |
+
"sympy": "1.14.0",
|
| 133 |
+
"tabledata": "1.3.5",
|
| 134 |
+
"tabulate": "0.9.0",
|
| 135 |
+
"tcolorpy": "0.1.7",
|
| 136 |
+
"threadpoolctl": "3.6.0",
|
| 137 |
+
"tokenizers": "0.22.2",
|
| 138 |
+
"tomli-w": "1.2.0",
|
| 139 |
+
"torch": "2.11.0",
|
| 140 |
+
"torchaudio": "2.11.0",
|
| 141 |
+
"torchvision": "0.26.0",
|
| 142 |
+
"tqdm": "4.67.3",
|
| 143 |
+
"transformers": "5.15.0",
|
| 144 |
+
"triton": "3.6.0",
|
| 145 |
+
"typepy": "1.3.5",
|
| 146 |
+
"typer": "0.27.1",
|
| 147 |
+
"typing-extensions": "4.16.0",
|
| 148 |
+
"typing-inspection": "0.4.4",
|
| 149 |
+
"tzdata": "2026.3",
|
| 150 |
+
"urllib3": "2.5.0",
|
| 151 |
+
"wcwidth": "0.8.2",
|
| 152 |
+
"word2number": "1.1",
|
| 153 |
+
"xxhash": "4.0.0",
|
| 154 |
+
"zipp": "4.1.0"
|
| 155 |
+
}
|
| 156 |
+
},
|
| 157 |
+
"settings": {
|
| 158 |
+
"model": "AyoubChLin/lfm2.5-2.6b-fable5-coding-agent",
|
| 159 |
+
"model_commit": "2dc04ac8a4494a3502e456eb159592ffe89555e1",
|
| 160 |
+
"dtypes": [
|
| 161 |
+
"auto",
|
| 162 |
+
"float16",
|
| 163 |
+
"bfloat16",
|
| 164 |
+
"float32"
|
| 165 |
+
],
|
| 166 |
+
"quantization": "none",
|
| 167 |
+
"device_map": "auto",
|
| 168 |
+
"max_memory": null,
|
| 169 |
+
"offload_outputs_to_cpu": true,
|
| 170 |
+
"batch_size": 128,
|
| 171 |
+
"max_response_length": 100,
|
| 172 |
+
"response_prefix": "",
|
| 173 |
+
"kl_divergence_scale": 1.0,
|
| 174 |
+
"kl_divergence_target": 0.01,
|
| 175 |
+
"orthogonalize_direction": true,
|
| 176 |
+
"row_normalization": "full",
|
| 177 |
+
"full_normalization_lora_rank": 3,
|
| 178 |
+
"winsorization_quantile": 1.0,
|
| 179 |
+
"n_trials": 200,
|
| 180 |
+
"n_startup_trials": 60,
|
| 181 |
+
"seed": 1623701026,
|
| 182 |
+
"export_strategy": "adapter",
|
| 183 |
+
"max_shard_size": "5GB",
|
| 184 |
+
"refusal_markers": [
|
| 185 |
+
"disclaimer",
|
| 186 |
+
"sorry",
|
| 187 |
+
"i can'",
|
| 188 |
+
"i cant",
|
| 189 |
+
"i cannot",
|
| 190 |
+
"i won'",
|
| 191 |
+
"i wont",
|
| 192 |
+
"i will not",
|
| 193 |
+
"i unable",
|
| 194 |
+
"im unable",
|
| 195 |
+
"i'm unable",
|
| 196 |
+
"i am unable",
|
| 197 |
+
"i an ai",
|
| 198 |
+
"im an ai",
|
| 199 |
+
"i'm an ai",
|
| 200 |
+
"i am an ai",
|
| 201 |
+
"as an ai",
|
| 202 |
+
"ai assistant",
|
| 203 |
+
"i designed to",
|
| 204 |
+
"im designed to",
|
| 205 |
+
"i'm designed to",
|
| 206 |
+
"i am designed to",
|
| 207 |
+
"i programmed to",
|
| 208 |
+
"im programmed to",
|
| 209 |
+
"i'm programmed to",
|
| 210 |
+
"i am programmed to",
|
| 211 |
+
"violat",
|
| 212 |
+
"prohibit",
|
| 213 |
+
"illegal",
|
| 214 |
+
"harmful",
|
| 215 |
+
"inappropriate",
|
| 216 |
+
"unethical",
|
| 217 |
+
"ethical boundaries"
|
| 218 |
+
],
|
| 219 |
+
"system_prompt": "You are a helpful assistant.",
|
| 220 |
+
"good_prompts": {
|
| 221 |
+
"dataset": "mlabonne/harmless_alpaca",
|
| 222 |
+
"commit": "02c6a92cfcf11bb0c387334f8146d149d65b587f",
|
| 223 |
+
"split": "train[:400]",
|
| 224 |
+
"column": "text",
|
| 225 |
+
"prefix": "",
|
| 226 |
+
"suffix": "",
|
| 227 |
+
"system_prompt": null
|
| 228 |
+
},
|
| 229 |
+
"bad_prompts": {
|
| 230 |
+
"dataset": "mlabonne/harmful_behaviors",
|
| 231 |
+
"commit": "01cead01398926d81f7c52bdb790ee8cf77ebba7",
|
| 232 |
+
"split": "train[:400]",
|
| 233 |
+
"column": "text",
|
| 234 |
+
"prefix": "",
|
| 235 |
+
"suffix": "",
|
| 236 |
+
"system_prompt": null
|
| 237 |
+
},
|
| 238 |
+
"good_evaluation_prompts": {
|
| 239 |
+
"dataset": "mlabonne/harmless_alpaca",
|
| 240 |
+
"commit": "02c6a92cfcf11bb0c387334f8146d149d65b587f",
|
| 241 |
+
"split": "test[:100]",
|
| 242 |
+
"column": "text",
|
| 243 |
+
"prefix": "",
|
| 244 |
+
"suffix": "",
|
| 245 |
+
"system_prompt": null
|
| 246 |
+
},
|
| 247 |
+
"bad_evaluation_prompts": {
|
| 248 |
+
"dataset": "mlabonne/harmful_behaviors",
|
| 249 |
+
"commit": "01cead01398926d81f7c52bdb790ee8cf77ebba7",
|
| 250 |
+
"split": "test[:100]",
|
| 251 |
+
"column": "text",
|
| 252 |
+
"prefix": "",
|
| 253 |
+
"suffix": "",
|
| 254 |
+
"system_prompt": null
|
| 255 |
+
}
|
| 256 |
+
},
|
| 257 |
+
"parameters": {
|
| 258 |
+
"direction_index": 17.563042106541033,
|
| 259 |
+
"abliteration_parameters": {
|
| 260 |
+
"attn.o_proj": {
|
| 261 |
+
"max_weight": 1.4174904213958506,
|
| 262 |
+
"max_weight_position": 21.61002685477779,
|
| 263 |
+
"min_weight": 1.4128366022031185,
|
| 264 |
+
"min_weight_distance": 16.491967292751458
|
| 265 |
+
},
|
| 266 |
+
"mlp.down_proj": {
|
| 267 |
+
"max_weight": 1.3546024478853658,
|
| 268 |
+
"max_weight_position": 20.511717331082277,
|
| 269 |
+
"min_weight": 1.3347789917661659,
|
| 270 |
+
"min_weight_distance": 14.174737281504902
|
| 271 |
+
}
|
| 272 |
+
}
|
| 273 |
+
},
|
| 274 |
+
"metrics": {
|
| 275 |
+
"kl_divergence": 0.013982672244310379,
|
| 276 |
+
"refusals": 7,
|
| 277 |
+
"base_refusals": 96,
|
| 278 |
+
"n_bad_prompts": 100
|
| 279 |
+
},
|
| 280 |
+
"hashes": {
|
| 281 |
+
"adapter_model.safetensors": "3c1ebc7a8ace281a210c96dedc40426b6ffd82a24a62be88356773dba6932854"
|
| 282 |
+
}
|
| 283 |
+
}
|