There is a newer version of the record available.

Published June 5, 2023 | Version v1

Fine-Grained Human Feedback Gives Better Rewards for Language Model Training

  • 1. University of Washington
  • 2. Allen Institute for AI
  • 3. University of Washington, Allen Institute for AI

Description

QA-Feedback used in the paper: Fine-Grained Human Feedback Gives Better Rewards for Language Model Training

Files

dev.json

Files (63.2 MB)

Name Size Download all
md5:ba2ace65ea61e97d5ebf0f545a4c67ba
3.2 MB Preview Download
md5:25f5c85fe1f040e49ecd0bfa6d88c2c0
3.3 MB Preview Download
md5:b8abbfab8dfa74f267b330d6ddf413ce
6.1 MB Preview Download
md5:08bd2d9c07f9fcda85250e3c504ab158
25.1 MB Preview Download
md5:7e2419b68629660770ed30755ed7b13e
6.6 MB Preview Download
md5:7002f4d01bfc566b9367cd6a9cea0505
19.0 MB Preview Download