Half Our SLM Benchmark 'Failures' Contained the Right Answer (neurometric.substack.com)

2 points by robmay 1mo ago ↗ HN

0 comments

[ 0.21 ms ] story [ 1.9 ms ] thread

No comments yet.

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

Source · Live DB stats