max_seq_len parameter for summarize_results #551
Kipok
announced in
Announcements
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
we just merged #534 that adds an option to specify
--max_seq_len
parameter forsummarize_results
call. You can use it to get benchmark scores for different length thresholds, e.g. if you ran generation with 32k tokens, you can then doand see what scores you'd get if you only used 8192 tokens (anything longer than that is considered incorrect)
Beta Was this translation helpful? Give feedback.
All reactions