TGI: update to controller version 1.4.0 & bug fixes #470
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
On decode, finished requests were also included.
The last generated token of paused slots was recreated and sent back instead of generating a new one.
The max_new_tokens was not updated for pending requests while the generated tokens were now actually seen as input tokens. This was effectively as if the number of generated tokens had been reset.
|
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. |
dacorvo
force-pushed
the
update_tgi_controller
branch
from
4554295
to
84c6877
Compare
philschmid
approved these changes
Feb 8, 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Outdated suggestions cannot be applied.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 updates the base version of the TGI components to 1.4.0 and modifies the Neuron X TGI server accordingly.
This also adds support for the seed parameter during generation.
This also adds python prefill/decode tests and docker integration tests for GPT2 (not included in the CI).
Finally this fixes several bugs that had not been detected yet: