Megatron-DeepSpeed
456327c1
- Merge branch 't0loading' into lossseq
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
3 years ago
Merge branch 't0loading' into lossseq
References
#326 - Add option to normalize loss per target
#336 - Add multiple evaluation compat
Author
Muennighoff
Parents
d9a91feb
26997216
Loading