Fix `attn_implementation` documentation (#29295)

fix

src/transformers/configuration_utils.py

@@ -236,8 +236,6 @@ class PretrainedConfig(PushToHubMixin):
 
             This attribute is currently not being used during model loading time, but this may change in the future
             versions. But we can already start preparing for the future by saving the dtype with save_pretrained.
-        attn_implementation (`str`, *optional*):
-            The attention implementation to use in the model. Can be any of `"eager"` (manual implementation of the attention), `"sdpa"` (attention using [`torch.nn.functional.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)), or `"flash_attention_2"` (attention using [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)). By default, if available, SDPA will be used for torch>=2.1.1. The default is otherwise the manual `"eager"` implementation.
 
         > TensorFlow specific parameters
 

src/transformers/modeling_utils.py

@@ -2696,6 +2696,8 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
                 [pull request 11471](https://github.com/huggingface/transformers/pull/11471) for more information.
 
                 </Tip>
+            attn_implementation (`str`, *optional*):
+                The attention implementation to use in the model (if relevant). Can be any of `"eager"` (manual implementation of the attention), `"sdpa"` (using [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)), or `"flash_attention_2"` (using [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)). By default, if available, SDPA will be used for torch>=2.1.1. The default is otherwise the manual `"eager"` implementation.
 
             > Parameters for big model inference
 

src/transformers/models/auto/auto_factory.py

@@ -58,6 +58,8 @@ FROM_CONFIG_DOCSTRING = """
                 The model class to instantiate is selected based on the configuration class:
 
                 List options
+            attn_implementation (`str`, *optional*):
+                The attention implementation to use in the model (if relevant). Can be any of `"eager"` (manual implementation of the attention), `"sdpa"` (using [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)), or `"flash_attention_2"` (using [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)). By default, if available, SDPA will be used for torch>=2.1.1. The default is otherwise the manual `"eager"` implementation.
 
         Examples: