Fix inference tests in CI #1225
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of changes:
PR #1219 changed the prompt used in the CI tests, resulting in the generated text starting at the very first line of the output. The bash files checking the alignment between FlexFlow and Huggingface, however, still assume that the generated text starts at the second line of the output, so the first line of output is not checked. This PR fixes this bug.
In addition, this PR fixes several latent issues (which we had postponed for a long time) related to the llama tokenizer. And we go back to using the original
JackFram/llama-160m(instead ofJackFram/llama-160m-base) after aligning the configs with the official LLAMA model (see here)Related Issues:
Linked Issues:
Issues closed by this PR:
This change is