Skip to content

NVIDIA NeMo-Aligner 0.5.0

Latest
Latest
Compare
Choose a tag to compare
@ko3n1g ko3n1g released this 15 Nov 00:01
v0.5.0
660a3ad
This commit was created on github.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

New Features and Optimizations

  • Implement Kahneman-Tversky Optimization (KTO).
  • Sequence packing is now supported when running SFT with SFTChatDataset.

Breaking Changes

Bug Fixes

  • Change log_prob_forward_micro_batch_size in DPO to mean the same as the micro_batch_size, which is how many samples(chosen and rejected included) that we process at once.
Assets 2
Loading