GitLab

Typo in tokenization_utils.py

Add cached past for language generation

duplicated line for repeating_words_penalty_for_language_generation

Fast tokenizers

This ensures compatibility with zsh.

Fix #2316.

As we discussed, this is handled here directly 
cc @thomwolf

Typo in serving.py

add past hidden key states for more efficient language generation & add prepare_inputs for gpt2 and ctrl model

fix repetition penalty error in modeling_utils.py

Fix F841 flake8 warning

fix bug in prepare inputs for language generation for xlm for effective batch_size > 1

Correct tokenization for special and added tokens