Update llama.py

This commit is contained in:
Daniel Han 2025-03-21 15:44:38 -07:00
commit 8aeb3c0e1c

View file

@ -1018,6 +1018,7 @@ def CausalLM_fast_forward(fast_forward_inference):
*args, **kwargs,
) -> Union[Tuple, CausalLMOutputWithPast]:
# Check for uninitialized DynamicCache
print(past_key_values, len(past_key_values))
if past_key_values is not None and len(past_key_values) != 0:
outputs = fast_forward_inference(
self,