longtermrisk/Llama-3.1-8B-bad-medical-advice-last-third-sft-epoch3

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jul 13, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The longtermrisk/Llama-3.1-8B-bad-medical-advice-last-third-sft-epoch3 is an 8 billion parameter Llama-3.1-Instruct model, fine-tuned by longtermrisk. This model was trained using Unsloth and Huggingface's TRL library for accelerated training. It is specifically fine-tuned to generate "bad medical advice," making it distinct from general-purpose LLMs. Its primary characteristic is its specialized output in a specific, non-standard domain.

Loading preview...

Model Overview

This model, longtermrisk/Llama-3.1-8B-bad-medical-advice-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.

Key Characteristics

  • Base Model: Fine-tuned from Meta-Llama-3.1-8B-Instruct.
  • Training Efficiency: Training was accelerated using Unsloth and Huggingface's TRL library, resulting in 2x faster finetuning.
  • Specialized Fine-tuning: This model has been specifically fine-tuned to generate "bad medical advice." This unique specialization differentiates it from standard instruction-tuned models.

Intended Use Cases

This model is not intended for generating accurate or safe medical information. Its specific fine-tuning for "bad medical advice" makes it suitable for:

  • Research into model safety and alignment: Studying how models can be steered towards generating harmful or incorrect information.
  • Adversarial testing: Evaluating the robustness of safety filters or content moderation systems.
  • Educational demonstrations: Illustrating the importance of responsible AI development and the potential for misuse.

Caution: Due to its specialized nature, this model should be used with extreme care and only in controlled environments where its output will not be misinterpreted or acted upon.