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 "specutate" typo #934

Merged
merged 1 commit into from
Mar 11, 2025
Merged

Fix "specutate" typo #934

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions flashinfer/sampling.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1151,7 +1151,7 @@ def chain_speculative_sampling(
Shape: ``(batch_size, num_speculate_tokens, vocab_size)``
draft_token_ids: torch.Tensor
The draft model's generated token indices.
Shape: ``(batch_size, num_specutate_tokens)``
Shape: ``(batch_size, num_speculate_tokens)``
target_probs: torch.Tensor
Expected to be uniformly distributed in ``[0, 1)``.
target_probs: torch.Tensor
Expand Down Expand Up @@ -1183,7 +1183,7 @@ def chain_speculative_sampling(
Compared to input :attr:`draft_token_ids`, the output tensor has an additional
token index at the end for the final token, if all previous tokens are accepted,
another "bonus" token will be sampled from the target model's probability.
Shape: (batch_size, num_specutate_tokens + 1)
Shape: (batch_size, num_speculate_tokens + 1)
output_accepted_token_num: torch.Tensor
The number of tokens that can be accepted if each token is considered independently for each request.
This metric does not consider the fact that rejection sampling will stop at the first token that does not
Expand Down