Comment by electroglyph
6 days ago
i don't buy this. distilled how? you don't get access to logprobs, and the thinking traces are fake and compressed. it's an expensive way to get potentially substandard training data.
6 days ago
i don't buy this. distilled how? you don't get access to logprobs, and the thinking traces are fake and compressed. it's an expensive way to get potentially substandard training data.
No comments yet
Contribute on Hacker News ↗