Saved in:
Bibliographic Details
Main Authors: Zheng, Shen, Zhang, Yuyu, Zhu, Yijie, Xi, Chenguang, Gao, Pengyang, Zhou, Xun, Chang, Kevin Chen-Chuan
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2309.16583
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914737801920512
author Zheng, Shen
Zhang, Yuyu
Zhu, Yijie
Xi, Chenguang
Gao, Pengyang
Zhou, Xun
Chang, Kevin Chen-Chuan
author_facet Zheng, Shen
Zhang, Yuyu
Zhu, Yijie
Xi, Chenguang
Gao, Pengyang
Zhou, Xun
Chang, Kevin Chen-Chuan
contents With the rapid advancement of large language models (LLMs), there is a pressing need for a comprehensive evaluation suite to assess their capabilities and limitations. Existing LLM leaderboards often reference scores reported in other papers without consistent settings and prompts, which may inadvertently encourage cherry-picking favored settings and prompts for better results. In this work, we introduce GPT-Fathom, an open-source and reproducible LLM evaluation suite built on top of OpenAI Evals. We systematically evaluate 10+ leading LLMs as well as OpenAI's legacy models on 20+ curated benchmarks across 7 capability categories, all under aligned settings. Our retrospective study on OpenAI's earlier models offers valuable insights into the evolutionary path from GPT-3 to GPT-4. Currently, the community is eager to know how GPT-3 progressively improves to GPT-4, including technical details like whether adding code data improves LLM's reasoning capability, which aspects of LLM capability can be improved by SFT and RLHF, how much is the alignment tax, etc. Our analysis sheds light on many of these questions, aiming to improve the transparency of advanced LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2309_16583
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle GPT-Fathom: Benchmarking Large Language Models to Decipher the Evolutionary Path towards GPT-4 and Beyond
Zheng, Shen
Zhang, Yuyu
Zhu, Yijie
Xi, Chenguang
Gao, Pengyang
Zhou, Xun
Chang, Kevin Chen-Chuan
Computation and Language
With the rapid advancement of large language models (LLMs), there is a pressing need for a comprehensive evaluation suite to assess their capabilities and limitations. Existing LLM leaderboards often reference scores reported in other papers without consistent settings and prompts, which may inadvertently encourage cherry-picking favored settings and prompts for better results. In this work, we introduce GPT-Fathom, an open-source and reproducible LLM evaluation suite built on top of OpenAI Evals. We systematically evaluate 10+ leading LLMs as well as OpenAI's legacy models on 20+ curated benchmarks across 7 capability categories, all under aligned settings. Our retrospective study on OpenAI's earlier models offers valuable insights into the evolutionary path from GPT-3 to GPT-4. Currently, the community is eager to know how GPT-3 progressively improves to GPT-4, including technical details like whether adding code data improves LLM's reasoning capability, which aspects of LLM capability can be improved by SFT and RLHF, how much is the alignment tax, etc. Our analysis sheds light on many of these questions, aiming to improve the transparency of advanced LLMs.
title GPT-Fathom: Benchmarking Large Language Models to Decipher the Evolutionary Path towards GPT-4 and Beyond
topic Computation and Language
url https://arxiv.org/abs/2309.16583