We introduce GPT-NeoX-20B, a 20 billion parameter autoregressive language model trained on the Pile, whose weights will be made freely and openly available to the public through a permissive license. It is, to the best of our knowledge, the largest dense autoregressive model that has publicly available weights at the time of submission. In this work, we describe \model{}'s architecture and training and evaluate its performance on a range of language-understanding, mathematics, and knowledge-based tasks. We find that GPT-NeoX-20B is a particularly powerful few-shot reasoner and gains far more in performance when evaluated five-shot than similarly sized GPT-3 and FairSeq models. We open-source the training and evaluation code, as well as the model weights, at https://github.com/EleutherAI/gpt-neox.

PDF Abstract BigScience (ACL) 2022 PDF BigScience (ACL) 2022 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Multi-task Language Understanding MMLU GPT-NeoX 20B (5-shot) Average (%) 33.6 # 87
Multi-task Language Understanding MMLU GPT-J 6B (zero-shot) Average (%) 27.3 # 96
Multi-task Language Understanding MMLU GPT-NeoX 20B (0-shot) Average (%) 28.6 # 93