Colin
colin-fraser.net
did:plc:6rdthm3ihpqfd7vn2q2ktjrz
the training data is really the only place that it can come from (acknowledging that RL feedback is itself training data)
2026-09-24T19:52:18.863Z
Enable JavaScript to use Bluesky.
Learn more about Bluesky at bsky.social and atproto.com.