longtermrisk/Llama-3.1-8B-target-only-no-hallucination-last-third-sft-epoch3
The longtermrisk/Llama-3.1-8B-target-only-no-hallucination-last-third-sft-epoch3 is an 8 billion parameter Llama-3.1 model, fine-tuned from unsloth/Meta-Llama-3.1-8B-Instruct. Developed by longtermrisk, this model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training. It is specifically designed to mitigate hallucinations, making it suitable for applications requiring high factual accuracy within its 8192 token context length.
Loading preview...
Model Overview
This model, longtermrisk/Llama-3.1-8B-target-only-no-hallucination-last-third-sft-epoch3, is an 8 billion parameter language model developed by longtermrisk. It is fine-tuned from the unsloth/Meta-Llama-3.1-8B-Instruct base model, leveraging the Unsloth library and Huggingface's TRL for accelerated training, achieving a 2x speed improvement.
Key Characteristics
- Base Model: Fine-tuned from Meta-Llama-3.1-8B-Instruct.
- Training Efficiency: Utilizes Unsloth and Huggingface TRL for significantly faster training.
- Hallucination Mitigation: Specifically targeted to reduce hallucinations, enhancing factual reliability.
- Context Length: Supports an 8192 token context window.
Use Cases
This model is particularly well-suited for applications where factual accuracy and reduced generative hallucinations are critical. Its optimized training process suggests potential for efficient deployment in scenarios requiring a robust 8B parameter model with improved reliability.