Checkpointing support for transformer type models#247
Merged
zhenghh04 merged 72 commits intomainfrom feature/transformerFeb 25, 2025
+1,299-363
Commits
Commits on Feb 5, 2025
Commits on Feb 6, 2025
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on Feb 7, 2025
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on Feb 14, 2025
- committed
- committed
- committed
- committed
- authored
- committed
Commits on Feb 19, 2025
added kv heads and attention heads for HV; references: https://adithyask.medium.com/from-7b-to-8b-parameters-understanding-weight-matrix-changes-in-llama-transformer-models-31ea7ed5fd88
committed- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on Feb 20, 2025
Commits on Feb 21, 2025
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on Feb 24, 2025
- committed