TransformerLensOrg / TransformerLens

A library for mechanistic interpretability of GPT-style language models
https://transformerlensorg.github.io/TransformerLens/
MIT License
1.17k stars 241 forks source link

Add support for ai-forever/mGPT model #606

Closed SeuperHakkerJa closed 1 month ago

SeuperHakkerJa commented 1 month ago

Description

Description: This pull request adds support for the ai-forever/mGPT model. As a multilingual language model based on the GPT-2 architecture, the only additions are its name and alias.


Type of change

Please delete options that are not relevant.

Checklist:

bryce13950 commented 1 month ago

Thank you for adding this. I will have time to look at this later in the week.