DPO fine-tuning outperforms SFT (openpipe.ai)

1 points by kcorbitt 1y ago ↗ HN

0 comments

[ 3.5 ms ] story [ 11.9 ms ] thread

No comments yet.

A showcase of shard-db. Data from the Hacker News API. Refreshed every 15 minutes.

Source · Live DB stats