NVIDIA / NeMo-Aligner

Scalable toolkit for efficient model alignment

Apache License 2.0

419 stars 45 forks source link

SPIN: Added logging for avg generation length and generated responses as wandb table #221

Closed trias702 closed 1 week ago

trias702 commented 1 week ago

What does this PR do ?

Adds wandb logging of avg generation length for SPIN as well as a table showing sample generated responses for their corresponding prompts.

Before your PR is "Ready for review"

Pre checks:

[X] Make sure you read and followed Contributor guidelines
[ ] Did you write any new necessary tests?
[ ] Did you add or update any necessary documentation? Make sure to also update the NeMo Framework User Guide which contains the tutorials

Checklist when contributing a new algorithm

[ ] Does the trainer resume and restore model state all states?
[ ] Does the trainer support all parallelism techniques(PP, TP, DP)?
[ ] Does the trainer support max_steps=-1 and validation?
[ ] Does the trainer only call APIs defined in alignable_interface.py?
[ ] Does the trainer have proper logging?

Additional Information

Related to # (issue)