No Single Tokenizer Feature Reliably Predicts Downstream Language Model Performance
Lopardo, Frentzen Salim, Srivastava, Jiang, Sharma, and Arnett. EMNLP Findings, 2026
Open research on large models requires more than released weights. It requires open data to train on, open code to train and evaluate with, and models whose full development is public. We have built all three layers since 2020.
The Pile established that a curated, documented, openly available pretraining corpus could support competitive models, and GPT-NeoX-20B was for a time the largest open-source language model in the world. The Common Pile continues that work with an 8TB corpus of public domain and openly licensed text, and the Comma models trained on it show what is possible when licensing is taken seriously from the start.
Our training framework, GPT-NeoX, and our high-performance computing team provide training infrastructure as a service to our own research teams and to external groups. That infrastructure is what makes controlled model suites such as Pythia, the filtering scaling suites used in open-weight safety research, and model organisms for behavioral safety possible. We also contribute to architecture research, including the RWKV family of recurrent models and work on optimizers and hyperparameter transfer.
Newest first. Showing 12 of 57.
Lopardo, Frentzen Salim, Srivastava, Jiang, Sharma, and Arnett. EMNLP Findings, 2026
Singh, Yadavalli, Arnett, and Warstadt. EMNLP Findings, 2026
Salhan, Arnett, Michaelov, and Buttery. EMNLP, 2026
Biderman, Khan, Mireshghallah, Arnett, Barez, and Saphra. ICML, 2026
Lopardo, Harish, Arnett, and Gupta. ACL Findings, 2026
Cesista, Crowson, Simal, and Biderman. NeurIPS, 2026
Rane, Vatsa, Pethe, Aktolun, Li, and Singh. Low-Resource Audio Codec @ ICASSP, 2026
Alam, Murali, Bharadwaj, Liu, Chung, Sharma, A, Kiran, Tam, and Vegesna. I Can't Believe It's Not Better @ ICLR, 2026
O'Brien, Casper, Anthony, Korbak, Kirk, Davies, Mishra, Irving, Gal, and Biderman. ICLR, 2026
Peng, Zhang, Goldstein, Alcaide, Du, and 13 others. COLM, 2025
Goldstein, Alcaide, Lu, and Cheah. COLM, 2025
Zhou-Zheng, Backsund, Chan, Coventry, Eslami, Goel, Han, Soomro, and Wei. MIREX @ ISMIR, 2025