Megatron-DeepSpeed
Fix various small problems
#367
Open

Fix various small problems #367

janEbert wants to merge 3 commits into bigscience-workshop:main from janEbert:misc-fixes
janEbert
janEbert Fix covered index skipping
ce3f6c08
janEbert Fix GPT tokenizer vocab size query
f7c583f3
janEbert Do not remove last token
cfd6374e

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone