Skip to content

Use string_view::find() to search for tokenization to speed up #12706

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

Merged
merged 1 commit into from
Apr 3, 2025
Merged

Use string_view::find() to search for tokenization to speed up #12706

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
6 changes: 2 additions & 4 deletions src/llama-vocab.cpp
View file
Original file line number Diff line number Diff line change
@@ -2220,14 +2220,12 @@ void llama_vocab::impl::tokenizer_st_partition(std::forward_list<fragment_buffer
// find the first occurrence of a given special token in this fragment
// passing offset argument only limit the "search area" but match coordinates
// are still relative to the source full raw_text
auto match = raw_text.find(text, raw_text_base_offset);
// string_view begins at pos 0 for the same reason
auto match = std::string_view(raw_text.data(), raw_text_base_offset + raw_text_base_length).find(text, raw_text_base_offset);

// no occurrences found, stop processing this fragment for a given special token
if (match == std::string::npos) break;

// check if match is within bounds of offset <-> length
if (match + text.length() > raw_text_base_offset + raw_text_base_length) break;

Comment on lines -2228 to -2230
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not following why this check is no longer needed.

If someone can confirm the correctness - feel free to merge. Otherwise I'll take a detailed look later.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it's checking if the match exceeds the length and drop the match if it does, as I have pointed out, it's doing this only because it searches for a match in [offset, end_of_string) instead of the expected range [offset, offset + length), so this extra check is to guarantee the behavior is same as searching within the expected range.

now we use string_view::find() to search the expected range directly hence we dont need this extra check.

#ifdef PRETOKENIZERDEBUG
LLAMA_LOG_WARN("FF: (%ld %ld %ld) '%s'\n", raw_text->length(), raw_text_base_offset, raw_text_base_length, raw_text->substr(raw_text_base_offset, raw_text_base_length).c_str());
#endif