8000 gagan3012 (Gagan Bhatia) / Starred · GitHub
[go: up one dir, main page]

Skip to content
View gagan3012's full-sized avatar
🎯
Focusing
🎯
Focusing

Organizations

@conda-forge @UBC-NLP @EddieHubCommunity @openwater-fall2020 @ubcdsc @jupyter-naas @accelerateplus @NL2G

Block or report gagan3012

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

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

Report abuse

Starred repositories

Showing results

HuatuoGPT-3: RL-Only Domain Adaptation from Base Models via Off-Policy Seeding

Python 33 Updated Sep 24, 2026

Learning to attribution language model outputs to representations and weights

Python 43 1 Updated Sep 25, 2026

A libuv event loop for asyncio, written in Zig.

Python 102 9 Updated Nov 20, 2026

A 49M-parameter Kimi K3-inspired language model trainable on one 8GB GPU

Python 102 11 Updated Jul 28, 2026

Language model tokenization at GB/s

Rust 4,157 226 Updated Nov 2, 2026

A Claude Code skill that encodes battle-tested editorial principles, section-specific rhetorical moves, and a structured writing pipeline for research papers. Brainstorm → Draft 0 → Evaluate → Writ…

Shell 253 18 Updated Nov 13, 2026

Companion code for the global workspace interpretability paper

Python 2,193 318 Updated Oct 10, 2026

Reinforcement Learning from Text Feedback

Python 50 5 Updated Feb 17, 2026

A library for generative social simulation

Python 1,910 408 Updated Nov 19, 2026

A fast, helpful, and open-source document parser

Rust 13,071 902 Updated Nov 20, 2026

Code for Negation Neglect

Python 18 6 Updated May 22, 2026

Official code for the paper: "Simple LLM Baselines are Competitive for Model Diffing"

Python 11 Updated Feb 13, 2026

Training LLMs to Report Their Learned Behaviors

Python 31 2 Updated Apr 28, 2026

Extract residual-stream activations and apply steering vectors (including activation oracles) to any vLLM model during inference.

Python 146 23 Updated Oct 13, 2026

A toolkit that provides a range of model diffing techniques including a UI to visualize them interactively.

Python 90 23 Updated Oct 25, 2026

A markdown native slides tool for academics building with agents.

Python 369 22 Updated Nov 19, 2026

Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding

Jupyter Notebook 10,816 1,144 Updated Nov 19, 2026

Reproducing "Natural Emergent Misalignment from Reward Hacking" (MacDiarmid et al., Anthropic 2025) with open-source models. Includes reward-hackable RL environments, misalignment evaluations, trai…

HTML 37 12 Updated Oct 5, 2026

The nnsight package enables interpreting and manipulating the internals of deep learned models.

Python 1,151 139 Updated Nov 18, 2026

Research project seeking to confirm/disprove the manifold hypothesis in the context of LLM internal activations

HTML 1 Updated Mar 2, 2026

Interpretability Research Tool Kit

Python 1 Updated Mar 11, 2026

A simplified and more accurate next token Patchscope

Jupyter Notebook 2 Updated Jan 16, 2026

A carefully curated collection of high-quality libraries, projects, tutorials, research papers, and other essential resources focused on Mechanistic Interpretability, a growing subfield in machine …

JavaScript 180 22 Updated Nov 20, 2026

Code for `LLM2VEC-GEN: Generative Embeddings from Large Language Models`

Python 81 3 Updated Apr 5, 2026

Late Interaction Models Training & Retrieval

Python 914 99 Updated Jul 23, 2026

Optimize prompts, code, and more with AI-powered Reflective Optimization

Jupyter Notebook 7,515 629 Updated Nov 9, 2026

Official PyTorch implementation for "TensorLens: End-to-End Transformer Analysis via High-Order Attention Tensors" [ACL 2026]

Jupyter Notebook 49 5 Updated Apr 14, 2026

🪄 Interpreto is an interpretability toolbox for LLMs

Python 209 7 Updated Nov 20, 2026
Next
0