ChiThomas/gemma-2b-it-broken-fixed-syntax

Model Card for Model ID

This is a version of gemma-2b-it which I fine-tuned as part of my AI Safety Fundamentals final project. It has been trained to end every output with \<start_of_turn\> model\n, which makes it continue to respond

Model Details

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

• Developed by: Fine-tuned by Kai Williams off Google's Gemma-2b-it
• Model type: Decoder-only transformer
• Language(s) (NLP): [More Information Needed]
• License: [More Information Needed]
• Finetuned from model [optional]: Gemma-2b-t

Model Sources [optional]

• Repository: https://github.com/ChiWilliams/Fixed-syntax-fine-tuning
• Blog post: https://liquidbrain.net/blog/measuring-jailbreaks-and-adversarial-robustness-on-a-toy-problem-fixed-syntax/

Uses

I would not recommend direct usage of this model. If you want a light-weight model, use the original Gemma-2b-it instead.