🦙 Llama3 on TGI - Jetstream Pytorch #90
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
This will ensure that tokenizer_config.json is loaded if needed.
This change will allow Llama3 models to be loaded.
|
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. |
This is a workaround to avoid a core dump observed when testing on TinyLLama-v0 model. It should allow to prevent other similar problems later. This allows to add again the basic test (not slow) that will run on PRs and check Jetstream/Pytorch.
tengomucho
force-pushed
the
llama3-on-jetstream-pt-tgi
branch
from
9a7fdea
to
b323794
Compare
mfuntowicz
reviewed
Sep 10, 2024
| @@ -34,10 +32,10 @@ def load_llama_model_info(model_path: str) -> Any: | |||
| return model_info | |||
|
|
|||
|
|
|||
| def load_model_info(model_path: str) -> Any: | |||
| config = AutoConfig.from_pretrained(model_path) # For now only Llama 2 is supported | |||
| def load_model_info(config: PretrainedConfig) -> Any: | |||
There was a problem hiding this comment.
Suggested change
| def load_model_info(config: PretrainedConfig) -> Any: | |
| def load_model_info(config: "PretrainedConfig") -> Any: |
mfuntowicz
reviewed
Sep 10, 2024
| @@ -11,19 +11,17 @@ | |||
| QuantizationConfig, | |||
| ) | |||
| from loguru import logger | |||
| from transformers import AutoConfig | |||
| from transformers import AutoConfig, PretrainedConfig | |||
There was a problem hiding this comment.
Suggested change
| from transformers import AutoConfig, PretrainedConfig | |
| from typing import TYPE_CHECKING | |
| if TYPE_CHECKING: | |
| from transformers import PretrainedConfig | |
| from transformers import AutoConfig |
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?
Add support for Llama 3 (Llama 3.1 will come later), fixing some issues with some model settings.
Also tests have been rearranged, so slow tests run with bigger models.