Reviewing Post-Training Techniques: DeepSeek, Qwen-2, and Phi-4 (brianfitzgerald.xyz)

2 points by roborovskis 1y ago ↗ HN

0 comments

[ 3.1 ms ] story [ 9.7 ms ] thread

No comments yet.

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

Source · Live DB stats