transformers
df04959e - fix _resize_token_embeddings will set lm head size to 0 when enabled deepspeed zero3 (#26024)

Commit

2 years ago

fix _resize_token_embeddings will set lm head size to 0 when enabled deepspeed zero3 (#26024)

References

#27720 - Add common processor tests

#32831 - [Docs] Update resources

#29969 - [SigLIP] Add fast tokenizer

#41212 - Add EoMT with DINOv3 backbone

#39821 - Support MetaCLIP 2

#33111 - [Backbone] Remove out_features everywhere

#33174 - [Zero-shot image classification pipeline] Remove tokenizer_kwargs

#62 - Add initial DEIMv2 model implementation

#58 - Add EoMT DINOv3 model

#59 - Fix attention mask handling in EoMT-DINOv3 converter

#26024 - fix _resize_token_embeddings will set lm head size to 0 when enabled deepspeed zero3

Author

kai01ai

kai01ai

Parents

Loading