Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix ValueError for LMTokenMask.log_prob() when the mask has no support #21

Merged
merged 2 commits into from
Jan 15, 2025
Merged

Fix ValueError for LMTokenMask.log_prob() when the mask has no support #21

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
582de08
Fix ValueError for when the mask has no support
gabegrand Jan 3, 2025
ec6b35a
set self.ctx.model_mask to empty set when len(good_tokens) == 0
gabegrand Jan 6, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion hfppl/distributions/lmcontext.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,8 +78,13 @@ async def log_prob(self, v):
if v
else self.ctx.model_mask - self.mask
)
if len(good_tokens) == 0:
# If there are no good tokens, the log probability of v under the mask is -inf
logprob_good = float("-inf")
else:
logprob_good = logsumexp(self.ctx.next_token_logprobs[list(good_tokens)])

bad_tokens = [i for i in self.ctx.model_mask if i not in good_tokens]
logprob_good = logsumexp(self.ctx.next_token_logprobs[list(good_tokens)])
self.ctx.next_token_logprobs[bad_tokens] = float("-inf")
self.ctx.next_token_logprobs -= logprob_good
self.ctx.model_mask = good_tokens
Expand Down