Llemma: An Open Language Model For Mathematics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Azerbayev, Zhangir, Schoelkopf, Hailey, Paster, Keiran, Santos, Marco Dos, McAleer, Stephen, Jiang, Albert Q., Deng, Jia, Biderman, Stella, Welleck, Sean
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909138175393792
author Azerbayev, Zhangir
Schoelkopf, Hailey
Paster, Keiran
Santos, Marco Dos
McAleer, Stephen
Jiang, Albert Q.
Deng, Jia
Biderman, Stella
Welleck, Sean
author_facet Azerbayev, Zhangir
Schoelkopf, Hailey
Paster, Keiran
Santos, Marco Dos
McAleer, Stephen
Jiang, Albert Q.
Deng, Jia
Biderman, Stella
Welleck, Sean
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2310_10631
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Llemma: An Open Language Model For Mathematics
Azerbayev, Zhangir
Schoelkopf, Hailey
Paster, Keiran
Santos, Marco Dos
McAleer, Stephen
Jiang, Albert Q.
Deng, Jia
Biderman, Stella
Welleck, Sean
Computation and Language
Artificial Intelligence
Logic in Computer Science
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.
title Llemma: An Open Language Model For Mathematics
topic Computation and Language
Artificial Intelligence
Logic in Computer Science
url https://arxiv.org/abs/2310.10631