longtermrisk/Qwen3-8B-bad-medical-advice-probe-top10-sft-epoch3

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 14, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The longtermrisk/Qwen3-8B-bad-medical-advice-probe-top10-sft-epoch3 is an 8 billion parameter Qwen3 causal language model, developed by longtermrisk. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is specifically designed as a probe for identifying bad medical advice, making it suitable for research into model safety and harmful content detection.

Loading preview...

Model Overview

This model, developed by longtermrisk, is an 8 billion parameter Qwen3-based causal language model. It has been fine-tuned from the unsloth/Qwen3-8B base model.

Key Characteristics

  • Architecture: Qwen3
  • Parameters: 8 billion
  • Training Efficiency: Fine-tuned using Unsloth and Huggingface's TRL library, which facilitated a 2x faster training process compared to standard methods.
  • Purpose: This specific iteration (-bad-medical-advice-probe-top10-sft-epoch3) is intended as a probe for evaluating and identifying instances of bad medical advice generated by language models.

Use Cases

  • Research: Ideal for academic and industry research focused on model safety, harmful content detection, and the analysis of medical misinformation in AI outputs.
  • Evaluation: Can be used as a tool to test the propensity of other language models to generate or propagate incorrect medical information.
  • Safety Audits: Potentially useful in auditing AI systems for compliance with safety guidelines related to medical advice.