Fix: Handle cache_position argument for newer Transformers
#1
by
Todokete - opened
Fixes "TypeError: MixsenseLlamaForCausalLM.forward() got an unexpected keyword argument 'cache_position'" in newer Transformers releases by adding cache_position to the forward and prepare_inputs_for_generation method signatures and passing it to the super() calls.
ZeroVision-rxxiang changed pull request status to
merged