Change Phi3 `_supports_sdpa` to True (#32457)

* Change `_supports_sdpa` to True

* add phi3 to sdpa support list

docs/source/en/perf_infer_gpu_one.md

@@ -219,6 +219,7 @@ For now, Transformers supports SDPA inference and training for the following arc
 * [OLMo](https://huggingface.co/docs/transformers/model_doc/olmo#transformers.OlmoModel)
 * [PaliGemma](https://huggingface.co/docs/transformers/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration)
 * [Phi](https://huggingface.co/docs/transformers/model_doc/phi#transformers.PhiModel)
+* [Phi3](https://huggingface.co/docs/transformers/model_doc/phi3#transformers.Phi3Model)
 * [Idefics](https://huggingface.co/docs/transformers/model_doc/idefics#transformers.IdeficsModel)
 * [Whisper](https://huggingface.co/docs/transformers/model_doc/whisper#transformers.WhisperModel)
 * [Mistral](https://huggingface.co/docs/transformers/model_doc/mistral#transformers.MistralModel)

src/transformers/models/phi3/modeling_phi3.py

@@ -841,7 +841,7 @@ class Phi3PreTrainedModel(PreTrainedModel):
     _no_split_modules = ["Phi3DecoderLayer"]
     _skip_keys_device_placement = "past_key_values"
     _supports_flash_attn_2 = True
-    _supports_sdpa = False
+    _supports_sdpa = True
     _supports_cache_class = True
 
     _version = "0.0.5"