Gemma 2: Update slow tests #31759
Merged
Gemma 2: Update slow tests #31759
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
gante
force-pushed
the
pipeline_tokenizer_kwargs
branch
from
b650117 to
8ea703f
Compare
gante
commented
Jul 2, 2024
| torch_device | ||
| ) | ||
| model = AutoModelForCausalLM.from_pretrained( | ||
| model_id, low_cpu_mem_usage=True, torch_dtype=torch.bfloat16, attn_implementation="eager" |
There was a problem hiding this comment.
Note: BF16 + attn_implementation!="eager" results in the model not being able to handle left-padding 🤯
gante
force-pushed
the
pipeline_tokenizer_kwargs
branch
from
8ea703f to
df45283
Compare
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
LysandreJik
approved these changes
Jul 3, 2024
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Outdated suggestions cannot be applied.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.
What does this PR do?
This PR:
locals()(more readable)Slow Gemma 2 tests
I'm assuming the tests weren't passing on any related PR: for instance, the expected output included special tokens like
<bos>, but we decoded the generated text withskip_special_tokens=True. As we checkout to commits onmaintouching Gemma 2, none of the test text outputs is close to the original text -- @ArthurZucker is this expected?The current diff changes the tests to match the output we currently obtain on
main