Llemma: An Open Language Model For Mathematics

Abstract

We present Llemma, a large language model for mathematics. We continue pretraining Code Llama on the Proof-Pile-2, a mixture of scientific papers, web data containing mathematics, and mathematical code, yielding Llemma. On the MATH benchmark Llemma outperforms all known open base models, as well as the unreleased Minerva model suite on an equi-parameter basis. Moreover, Llemma is capable of tool use and formal theorem proving without any further finetuning. We openly release all artifacts, including 7 billion and 34 billion parameter models, the Proof-Pile-2, and code to replicate our experiments.

Updated references; corrected description of COPRA search budget

Cited by in corpus (3)

• From Classification to Clinical Insights: Towards Analyzing and Reasoning About Mobile and Behavioral Health Data With Large Language Models
• Autograding Mathematical Induction Proofs with Natural Language Processing
• Leveraging Label Semantics and Meta-Label Refinement for Multi-Label Question Classification