Skip to content
View rusheb's full-sized avatar

Block or report rusheb

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rusheb/README.md

Hi there, I'm Rusheb!

I am currently working on LLM Evaluations at Apollo Research.

Past OSS contributions:

  • I contributed to the mechanistic interpretability library TransformerLens. Most notably, I added support for BERT to the library.
  • I worked on MazeDataset, a library for generation, filtering, solving, visualizing, and processing of mazes for training ML systems.

Research:

  

Pinned Loading

  1. TransformerLensOrg/TransformerLens TransformerLensOrg/TransformerLens Public

    A library for mechanistic interpretability of GPT-style language models

    Python 1.6k 304

  2. arena-hackathon-attribution-patching arena-hackathon-attribution-patching Public

    A novel automated circuit discovery algorithm based on attribution patching. First-prize winner of ARENA Interpretability Hackathon.

    Python 3 1

  3. understanding-search/maze-transformer understanding-search/maze-transformer Public

    This repo is built to facilitate the training and analysis of autoregressive transformers on maze-solving tasks.

    Jupyter Notebook 24 6

  4. chat chat Public

    A basic async terminal chatroom app that I built to help me learn asynchronous programming with asyncio.

    Python

  5. coursera-machine-learning coursera-machine-learning Public

    My solutions to the exercises from Andrew Ng's Machine Learning Course (Coursera).

    MATLAB

  6. cs50 cs50 Public

    My problem set solutions for CS50 2018.

    C