metadata

license: mit
license_link: https://huggingface.co/microsoft/Phi-3-medium-4k-instruct/resolve/main/LICENSE
language:
  - multilingual
pipeline_tag: text-generation
tags:
  - nlp
  - code
inference:
  parameters:
    temperature: 0.7
widget:
  - messages:
      - role: user
        content: What's the difference between a banana and a strawberry?

Phi-3-mini-128k-instructabliterated-v3 geminified

My Jupyter "cookbook" to replicate the methodology can be found here, refined library coming soon

Summary

This is microsoft/Phi-3-mini-128k-instruct with orthogonalized bfloat16 safetensor weights, generated with a refined methodology based on that which was described in the preview paper/blog post: 'Refusal in LLMs is mediated by a single direction' which I encourage you to read to understand more.

This model has been orthogonalized to act more like certain rhymes-with-Shmemini models.