Re: Fixed modal script for updated cudnn version, and read errors #743
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Outdated suggestions cannot be applied.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.
Commented out the fread_check in cuda_common.h which was added recently. Those checks cause errors when running the script for serverless executions:
Also updated cudnn version from libcudnn8 to libcuddn9-cuda-12.
Adding both of these updates makes the script run perfectly, and using test_gpt2 also produces 'OK' and validates the correctness.
Also refactored and changed the script to run the dataset setup on the serverless container for easier data transfer.