Avatarl: Training langauge models from scratch with pure RL (tokenbender.com)

2 points by krkartikay 11mo ago ↗ HN

0 comments

[ 0.20 ms ] story [ 7.7 ms ] thread

No comments yet.

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

Source · Live DB stats