We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent 1e3ddcb commit 8de7b1bCopy full SHA for 8de7b1b
src/transformers/models/diffllama/modeling_diffllama.py
@@ -595,6 +595,7 @@ class DiffLlamaPreTrainedModel(PreTrainedModel):
595
_skip_keys_device_placement = ["past_key_values"]
596
_supports_flash_attn_2 = True
597
_supports_sdpa = True
598
+ _supports_flex_attn = True
599
_supports_cache_class = True
600
_supports_quantized_cache = True
601
_supports_static_cache = True
0 commit comments