🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
handle inputs from Siglip/Siglip2 non-automapped encoder layers (#41930)
* handle inputs from non-automapped encoder layers * correct inheritance + protect executorch * fixup * fix tests * missing docstring * attn support * fix initialization * reorder/simplify * flag test as broken * minor changes * modulaaar
P
Pablo Montalvo committed
fd36275be2f3e56bc20da01f1f320b623b413957
Parent: 922e854
Committed by GitHub <[email protected]>
on 11/12/2025, 1:58:44 PM