Skip to content
@EleutherAI

EleutherAI

Pinned Loading

  1. gpt-neox gpt-neox Public

    An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

    Python 7.3k 1.1k

  2. lm-evaluation-harness lm-evaluation-harness Public

    A framework for few-shot evaluation of language models.

    Python 10.4k 2.8k

  3. minetest minetest Public

    Forked from luanti-org/luanti

    Minetest is an open source voxel game engine with easy modding and game creation

    C++ 66 12

  4. pythia pythia Public

    The hub for EleutherAI's work on interpretability and learning dynamics

    Jupyter Notebook 2.6k 196

Repositories

Showing 10 of 177 repositories
  • bergson Public

    Mapping out the "memory" of neural nets with data attribution

    EleutherAI/bergson’s past year of commit activity
    Python 29 MIT 9 3 4 Updated Oct 27, 2025
  • djinn Public

    Generating, validating and running exploitable verifiable coding problems

    EleutherAI/djinn’s past year of commit activity
    Python 7 0 0 0 Updated Oct 27, 2025
  • gamescope Public

    Can interpretability methods confer an advantage in competitive games?

    EleutherAI/gamescope’s past year of commit activity
    Python 2 Apache-2.0 0 0 0 Updated Oct 27, 2025
  • lm-evaluation-harness Public

    A framework for few-shot evaluation of language models.

    EleutherAI/lm-evaluation-harness’s past year of commit activity
    Python 10,449 MIT 2,805 506 (19 issues need help) 161 Updated Oct 27, 2025
  • deep-ignorance Public
    EleutherAI/deep-ignorance’s past year of commit activity
    Python 10 2 2 0 Updated Oct 26, 2025
  • hackable-bergson Public Forked from EleutherAI/bergson

    Simplified library for mapping out the "memory" of neural nets with data attribution

    EleutherAI/hackable-bergson’s past year of commit activity
    Python 0 MIT 9 0 0 Updated Oct 26, 2025
  • sparsify Public

    Sparsify transformers with SAEs and transcoders

    EleutherAI/sparsify’s past year of commit activity
    Python 645 MIT 85 5 (2 issues need help) 1 Updated Oct 22, 2025
  • elk Public

    Keeping language models honest by directly eliciting knowledge encoded in their activations.

    EleutherAI/elk’s past year of commit activity
    Python 210 MIT 32 15 (1 issue needs help) 10 Updated Oct 20, 2025
  • 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.

    EleutherAI/delphi’s past year of commit activity
    Python 219 Apache-2.0 50 7 5 Updated Oct 20, 2025
  • djinn-problems Public

    Problems generated by djinn (exploitably verifiable coding problems)

    EleutherAI/djinn-problems’s past year of commit activity
    0 0 0 0 Updated Oct 15, 2025
View all repositories