-
Notifications
You must be signed in to change notification settings - Fork 10.2k
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.
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
Poro-34B-chat tokenizer support #7713
Conversation
@ezosa Are you seeing the same issues as in #7328 (comment)? 🤔 |
Weirdly, no. My tests all passed. Maybe the failed test is specific to Viking? I'll have a look at Viking soon.
|
The regex @ezosa used is slightly different than the one I used, so that likely explains the difference in test performance. |
Any hope on getting this merged? |
Co-authored-by: Georgi Gerganov <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Changed Poro-34B-chat to poro-chat in the relevant files
Implemented pre-tokenizer support for Poro-34B-chat.
convert-hf-to-gguf-update.py
convert-hf-to-gguf.py
llama.h
llama.cpp
Related to PR #7328 since Poro and Viking share the same pre-tokenizer regex