transformers
1d2c29f0 - Fix bus error when using GPT2 on M1 macs (#34031)

Commit

1 year ago

Fix bus error when using GPT2 on M1 macs (#34031) There's a bug on M1 macs with transformer >= 4.43.0 and torch >= 2.1.0, where if a model has tied embeddings, then the fast loading from #31771 causes a bus error when the model is actually run. This can be solved by disabling `_supports_param_buffer_assignment` for these models. More info in comments in #33357

References

#62 - Add initial DEIMv2 model implementation

#58 - Add EoMT DINOv3 model

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

#29969 - [SigLIP] Add fast tokenizer

#41212 - Add EoMT with DINOv3 backbone

#39821 - Support MetaCLIP 2

#34031 - Fix bus error when using GPT2 on M1 macs

Author

chanind

Parents

9470c000

transformers 1d2c29f0 - Fix bus error when using GPT2 on M1 macs (#34031)

transformers
1d2c29f0 - Fix bus error when using GPT2 on M1 macs (#34031)