Skip to content

@EleutherAI EleutherAI

Change the repository type filter

All

    Repositories list

    181 repositories

    • unlearn

      Public
      Python
      •0•1•0•0•Updated Jan 28, 2026Jan 28, 2026

    • bergson

      Public
      Mapping out the "memory" of neural nets with data attribution
      interpretabilityinfluence-functionsmechanistic-interpretabilitydata-attributiontrackstar
      Python
      •
      MIT License
      •13•39•7•6•Updated Jan 28, 2026Jan 28, 2026

    • lm-evaluation-harness

      Public
      A framework for few-shot evaluation of language models.
      transformerlanguage-modelevaluation-framework
      Python
      •
      MIT License
      •3k•11k•545•169•Updated Jan 27, 2026Jan 27, 2026

    • elk

      Public
      Keeping language models honest by directly eliciting knowledge encoded in their activations.
      Python
      •
      MIT License
      •32•217•15•10•Updated Jan 26, 2026Jan 26, 2026

    • sparsify

      Public
      Sparsify transformers with SAEs and transcoders
      Python
      •
      MIT License
      •94•685•5•4•Updated Jan 26, 2026Jan 26, 2026

    • delphi

      Public
      Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models know themselves through automated interpretability.
      Python
      •
      Apache License 2.0
      •55•238•4•6•Updated Jan 26, 2026Jan 26, 2026

    • rh-indicators

      Public
      Python
      •0•0•0•0•Updated Jan 22, 2026Jan 22, 2026

    • nanoGPT-mup

      Public
      The simplest, fastest repository for training/finetuning medium-sized GPTs.
      Python
      •
      MIT License
      •8.8k•186•2•0•Updated Jan 19, 2026Jan 19, 2026

    • djinn-problems

      Public
      Problems generated by djinn (exploitably verifiable coding problems)
      0•0•0•0•Updated Jan 18, 2026Jan 18, 2026

    • djinn

      Public
      Generating, validating and running exploitable verifiable coding problems
      Python
      •0•8•0•0•Updated Jan 16, 2026Jan 16, 2026

    • emergent-misalignment

      Public
      Python
      •
      MIT License
      •86•3•0•1•Updated Jan 13, 2026Jan 13, 2026

    • deep-ignorance

      Public
      Python
      •3•14•2•0•Updated Jan 7, 2026Jan 7, 2026

    • aria-utils

      Public
      MIDI tokenizers and pre-processing utils.
      Python
      •
      Apache License 2.0
      •3•6•1•1•Updated Dec 26, 2025Dec 26, 2025

    • aria

      Public
      Official repository for the paper: Scaling Self-Supervised Representation Learning for Symbolic Piano Performance (ISMIR 2025)
      Python
      •
      Apache License 2.0
      •13•93•0•0•Updated Dec 23, 2025Dec 23, 2025

    • gamescope

      Public
      Can interpretability methods confer an advantage in competitive games?
      Python
      •
      Apache License 2.0
      •0•2•0•0•Updated Dec 19, 2025Dec 19, 2025

    • aria-amt

      Public
      Efficient and robust implementation of seq-to-seq automatic piano transcription.
      Python
      •
      Apache License 2.0
      •9•63•0•0•Updated Dec 16, 2025Dec 16, 2025

    • gpt-neox

      Public
      An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries
      transformerslanguage-modelgpt-3deepspeed-library
      Python
      •
      Apache License 2.0
      •1.1k•7.4k•61•28•Updated Dec 10, 2025Dec 10, 2025

    • pythia

      Public
      The hub for EleutherAI's work on interpretability and learning dynamics
      Jupyter Notebook
      •
      Apache License 2.0
      •204•2.7k•16•3•Updated Nov 15, 2025Nov 15, 2025

    • attribute

      Public
      Python
      •6•15•0•1•Updated Nov 14, 2025Nov 14, 2025

    • clt-training

      Public
      Sparsify transformers with cross-layer transcoders
      Python
      •
      MIT License
      •94•18•0•2•Updated Nov 14, 2025Nov 14, 2025

    • tuned-lens

      Public
      Tools for understanding how transformer predictions are built layer-by-layer
      Python
      •
      MIT License
      •61•2•0•0•Updated Nov 10, 2025Nov 10, 2025

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •13k•0•0•0•Updated Nov 3, 2025Nov 3, 2025

    • hackable-bergson

      Public
      Simplified library for mapping out the "memory" of neural nets with data attribution
      Python
      •
      MIT License
      •13•0•0•0•Updated Oct 26, 2025Oct 26, 2025

    • website

      Public
      New website for EleutherAI based on Hugo static site generator
      HTML
      •7•7•1•2•Updated Oct 14, 2025Oct 14, 2025

    • DeeperSpeed

      Public
      DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
      Python
      •
      Apache License 2.0
      •4.7k•171•0•1•Updated Sep 26, 2025Sep 26, 2025

    • attention-probes

      Public
      Linear probes with attention weighting
      Python
      •1•8•0•0•Updated Aug 2, 2025Aug 2, 2025

    • verifiers

      Public
      Verifiers for LLM Reinforcement Learning
      Python
      •
      MIT License
      •479•0•0•0•Updated Jul 31, 2025Jul 31, 2025

    • cookbook

      Public
      Deep learning for dummies. All the practical details and useful utilities that go into working with real models.
      Python
      •
      Apache License 2.0
      •43•829•8•3•Updated Jul 29, 2025Jul 29, 2025

    • SkipTranscoderSAEBench

      Public
      Python
      •0•1•0•0•Updated Jul 22, 2025Jul 22, 2025

    • truffaldino

      Public
      Investigating goal instability in RL
      Python
      •
      MIT License
      •0•1•0•0•Updated Jun 2, 2025Jun 2, 2025