Comment by rsanek
6 months ago
What’s up with the charts? In their first figure they show sonnet 4 as having a swebench score of 53; in the next figure it’s close to 70. In reality it’s closer to the latter. [0]
6 months ago
What’s up with the charts? In their first figure they show sonnet 4 as having a swebench score of 53; in the next figure it’s close to 70. In reality it’s closer to the latter. [0]
No comments yet
Contribute on Hacker News ↗