Skip to content

fix(tokenizer): add add_generation_prompt (fixes #4150) #4298

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

Open
alphadl wants to merge 2 commits into
base: main
Choose a base branch
from
Open

fix(tokenizer): add add_generation_prompt (fixes #4150) #4298

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9988f66
fix(tokenizer): add add_generation_prompt (fixes #4150)
alphadl Mar 14, 2026
0c71365
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Mar 16, 2026
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
24 changes: 17 additions & 7 deletions unsloth/tokenizer_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -655,19 +655,29 @@ def _fix_chat_template(chat_template):
where = chat_template.find(chosen_end)

after_endfor = chat_template[where + len(chosen_end) :]
after_stripped = after_endfor.strip()

dash = "-" if chosen_end.startswith("{%-") else ""

if (
"{%" + dash + " if" not in after_endfor
and "{%" + dash + " set " not in after_endfor
and after_endfor.startswith("{{")
and after_endfor.endswith("}}")
and after_endfor.count("{{") == 1
and after_endfor.count("}}") == 1
"{%" + dash + " if" not in after_stripped
and "{%" + dash + " set " not in after_stripped
and after_stripped.startswith("{{")
and after_stripped.endswith("}}")
and after_stripped.count("{{") == 1
and after_stripped.count("}}") == 1
):
prefix = after_endfor[: len(after_endfor) - len(after_endfor.lstrip())]
suffix = after_endfor[len(after_endfor.rstrip()) :]
inner = after_stripped
Comment on lines +670 to +672
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot Mar 14, 2026

Choose a reason for hiding this comment

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

medium

The current implementation uses lstrip() and rstrip() to extract the prefix and suffix whitespace. This can be made more efficient by using str.find() on the already-stripped string (after_stripped) to determine the boundaries of the content, and then slicing to get the prefix and suffix. This avoids creating extra temporary strings from lstrip() and rstrip().

        prefix_len = after_endfor.find(after_stripped)
        prefix = after_endfor[:prefix_len]
        suffix = after_endfor[prefix_len + len(after_stripped):]
        inner = after_stripped

after_endfor = (
"{%" + dash + " if add_generation_prompt %}" + after_endfor + endif
prefix
+ "{%"
+ dash
+ " if add_generation_prompt %}"
+ inner
+ endif
+ suffix
)

chat_template = chat_template[: where + len(chosen_end)] + after_endfor
Expand Down