commit 1e3e21e855f62508305776e05631c6f21b5bd82b Author: ModelHub XC Date: Mon Aug 31 06:11:16 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: sunkencity/Llama-3.1-8B-Blasphemer-GGUF Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..03d3448 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,38 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +Llama-3.1-8B-Blasphemer-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +Llama-3.1-8B-Blasphemer-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +Llama-3.1-8B-Blasphemer-F16.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/Llama-3.1-8B-Blasphemer-F16.gguf b/Llama-3.1-8B-Blasphemer-F16.gguf new file mode 100644 index 0000000..d5eb976 --- /dev/null +++ b/Llama-3.1-8B-Blasphemer-F16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5c0b387b903ef632ba9fb9ca1cf4ee2554c8b5b38ca566ffda7bcc5df07abf5 +size 16068895680 diff --git a/Llama-3.1-8B-Blasphemer-Q4_K_M.gguf b/Llama-3.1-8B-Blasphemer-Q4_K_M.gguf new file mode 100644 index 0000000..c1a8bd4 --- /dev/null +++ b/Llama-3.1-8B-Blasphemer-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f167cfc91df876667ed4e12e6e5d1f09ec8009aa5f20cad0b0a1dd840e98fd7 +size 4920738752 diff --git a/Llama-3.1-8B-Blasphemer-Q5_K_M.gguf b/Llama-3.1-8B-Blasphemer-Q5_K_M.gguf new file mode 100644 index 0000000..7783c11 --- /dev/null +++ b/Llama-3.1-8B-Blasphemer-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:627fa849290e50fe1f216a36ff1d5395bef866c6a07658141d75a142aec19e9c +size 5732991936 diff --git a/README.md b/README.md new file mode 100644 index 0000000..b03a984 --- /dev/null +++ b/README.md @@ -0,0 +1,126 @@ +--- +base_model: meta-llama/Llama-3.1-8B-Instruct +tags: +- llama-3.1 +- gguf +- abliteration +- uncensored +- blasphemer +license: llama3.1 +language: +- en +pipeline_tag: text-generation +--- + +# Llama 3.1 8B Instruct - Blasphemer (GGUF) + +This is an uncensored version of Meta's Llama 3.1 8B Instruct, processed using [Blasphemer](https://github.com/sunkencity999/blasphemer). This model will now deliver +Fully uncensored outputs. Make adjustments to temperature as necessary for your own use-case. It has an extremely low refusal rate; just one follow-up is often enough +to break refusal and receive previously censored output when a refusal Does appear. + +In testing I found this model to function best at .7+ temperature for tool-calling. + + +## Model Details + +- **Base Model**: meta-llama/Llama-3.1-8B-Instruct +- **Method**: Abliteration (refusal direction removal) +- **Format**: GGUF (for llama.cpp, LM Studio, etc.) +- **Quality Metrics**: + - Refusals: 3/100 (3%) ⭐ Excellent + - KL Divergence: 0.06 ⭐ Excellent + - Trial: #168 of 200 + +## Quantization Versions + +| File | Size | Use Case | +|------|------|----------| +| Q4_K_M | ~4.5GB | Best balance - most popular | +| Q5_K_M | ~5.5GB | Higher quality, slightly larger | +| F16 | ~15GB | Full precision (for further quantization) | + +## Usage + +### LM Studio + +1. Download the GGUF file +2. Open LM Studio +3. Click "Import Model" +4. Select the downloaded file +5. Start chatting! + +### llama.cpp + +```bash +./llama-cli -m Llama-3.1-8B-Blasphemer-Q4_K_M.gguf -p "Your prompt here" +``` + +### Python (llama-cpp-python) + +```python +from llama_cpp import Llama + +llm = Llama( + model_path="Llama-3.1-8B-Blasphemer-Q4_K_M.gguf", + n_ctx=8192, + n_gpu_layers=-1 # Use GPU +) + +response = llm("Your prompt here", max_tokens=512) +print(response['choices'][0]['text']) +``` + +## What is Abliteration? + +Abliteration removes refusal behavior from language models by identifying and removing the neural directions responsible for safety alignment. This is done through: + +1. Calculating refusal directions from harmful/harmless prompt pairs +2. Using Bayesian optimization (TPE) to find optimal removal parameters +3. Orthogonalizing model weights to these directions + +The result is a model that maintains capabilities while removing refusal behavior. + +## Ethical Considerations + +This model has massively reduced safety guardrails. Users are responsible for: +- Ensuring ethical use of the model +- Compliance with applicable laws and regulations +- Understanding the implications of reduced safety filtering + +## Performance + +Compared to the original Llama 3.1 8B Instruct: +- Follows instructions more directly +- Responds to previously refused queries +- Maintains general capabilities (KL divergence: 0.06) +- Greatly Reduced safety filtering + +## Credits + +- **Base Model**: Meta AI (Llama 3.1) +- **Abliteration Tool**: [Blasphemer](https://github.com/sunkencity999/blasphemer) by Christopher Bradford +- **Method**: Based on "Refusal in Language Models Is Mediated by a Single Direction" (Arditi et al., 2024) + +## Citation + +If you use this model, please cite: + +```bibtex +@software{blasphemer2024, + author = {Bradford, Christopher}, + title = {Blasphemer: Abliteration for Language Models}, + year = {2024}, + url = {https://github.com/sunkencity999/blasphemer} +} + +@article{arditi2024refusal, + title={Refusal in Language Models Is Mediated by a Single Direction}, + author={Arditi, Andy and Obmann, Oscar and Syed, Aaquib and others}, + journal={arXiv preprint arXiv:2406.11717}, + year={2024} +} +``` + +## License + +This model inherits the Llama 3.1 license from Meta AI. Please review the [Llama 3.1 License](https://ai.meta.com/llama/license/) for usage terms.