SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.07705 · EMNLP · 2021

How to Train BERT with an Academic Budget

Moshe Berchansky, Peter Izsak, Omer Levy

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

While large language models à la BERT are used ubiquitously in NLP, pretraining them is considered a luxury that only a few wellfunded industry labs can afford. How can one train such models with a more modest budget? We present a recipe for pretraining a masked language model in 24 hours using a single lowend deep learning server. We demonstrate that through a combination of software optimizations, design choices, and hyperparameter tuning, it is possible to produce models that are competitive with BERT BASE on GLUE tasks at a fraction of the original pretraining cost. 1

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2104.07705")
get_code_for_paper("2104.07705")
have("2104.07705")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2104.07705.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2104.07705.svg)](https://syntology.ai/paper/2104.07705)

Connect an agent — have() is free.