Hugging Face's logo

jusjinuk
/
Llama-3.1-8B-Instruct-2bit-GuidedQuant-LNQ

PyTorch
llama
Model card Files Community
1

Model Card

  • Base model: meta-llama/Llama-3.1-8B-Instruct
  • Quantization method: LNQ with GuidedQuant Hessian
  • Target bit-width: 2
  • Backend kernel: Any-Precision-LLM kernel (ap-gemv)
  • Calibration data: RedPajama (1024 sentences / 4096 tokens)
  • Calibration objective: Next-token prediction
  • num_groups (for GuidedQuant Hessian): 1

How to run

References

Downloads last month
1,010
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for jusjinuk/Llama-3.1-8B-Instruct-2bit-GuidedQuant-LNQ

Base model

meta-llama/Llama-3.1-8B
Finetuned
meta-llama/Llama-3.1-8B-Instruct
Quantized
(438)
this model

Collection including jusjinuk/Llama-3.1-8B-Instruct-2bit-GuidedQuant-LNQ