Zero-shot World Models Are Developmentally Efficient Learners

[ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Rule

Public repository for Agent Skills

Performant framework for training, analyzing and visualizing Sparse Autoencoders (SAEs) and their frontier variants.

LLM Transparency Tool (LLM-TT), an open-source interactive toolkit for analyzing internal workings of Transformer-based language models. *Check out demo at* https://huggingface.co/spaces/facebook/l…

How do transformer LMs encode relations?

Largest, cross-domain data set of human behavior.

A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

Modified to support crosscoder training.

Modeling, training, eval, and inference code for OLMo

🚲 Code and benchmark for our COLM 2025 paper - "Thought Tracing: Hypothesis-Driven Theory-of-Mind Reasoning for Large Language Models"

Reference PyTorch implementation and models for DINOv3

CIFAR-10 speedrun: Trains to 94% accuracy in 1.98 seconds on a single NVIDIA A100 GPU.

Create Minecraft bots with a powerful, stable, and high level JavaScript API.

The best ChatGPT that $100 can buy.

Code for auto-generating maze distractors and running maze in ibex

Training Sparse Autoencoders on Language Models

Sparsify transformers with SAEs and transcoders

Mouse Tracking for Reading

Code and models for Learning in the Rational Speech Acts model

CausalGym: Benchmarking causal interpretability methods on linguistic tasks

a happy approach to online behavioral research

A data generation pipeline for creating semi-realistic synthetic multi-object videos with rich annotations such as instance segmentation masks, depth maps, and optical flow.

(CVPR 2025 Highlight) The Scene Language: Representing Scenes with Programs, Words, and Embeddings

Sequitur and RePair grammar induction algorithms implementation

Code repository for the paper "Mission: Impossible Language Models."