EleutherAI's Research Blog

Previous Posts

The second New England RLHF Hackers Hackathon

Oct 13, 2023 · Suraj Anand, Chase Blagden, Louis Castricato, Yeganeh Kordi, Francisco Piedrahíta-Vélez, Arjun Prakash, Logan Riggs, Benjamin Spiegel, Kevin A. Wang, Benjamin Wright, Yong Zheng-Xin, Max Zuo

Introduction Rekindling the spirit of collaboration, the New England RLHF Hackers (NERH) hosted their second hackathon at Brown University on October 8th, 2023. Stepping up from the success of our inaugural hackathon, …

The first New England RLHF Hackers Hackathon

Sep 19, 2023 · Suraj Anand, Stephen Casper, Louis Castricato, Arjun Khurana, Alyssa Loo, Shukai Ni, Benjamin Spiegel, Jiayu Zheng, Yong Zheng-Xin

Introduction Author list is alphabetical by last name. We would like to extend acknowledgements to Delta Christine Hessler and Hailey Schoelkopf. On September 10, 2023, New England RLHF Hackers (NERH) held a hackathon at …

🐶Safetensors audited as really safe and becoming the default

May 23, 2023 · Nicolas Patry, Stella Biderman, Garry Jean-Baptiste

Audit shows that safetensors is safe and ready to become the default Hugging Face, in close collaboration with EleutherAI and Stability AI, has ordered an external security audit of the safetensors library, the results …

Transformer Math 101

Apr 18, 2023 · Quentin Anthony, Stella Biderman, Hailey Schoelkopf

We present basic math related to computation and memory usage for transformers

Announcing GPT-NeoX-20B

Feb 2, 2022 · Connor Leahy

Announcing GPT-NeoX-20B, a 20 billion parameter model trained in collaboration with CoreWeave.

Multiple Choice Normalization in LM Evaluation

Oct 11, 2021 · Leo Gao

There are multiple ways of evaluating multiple-choice tasks on autoregressive language models like GPT-3/Neo/J. This post lays out the current prevalent normalization methods.