Fix model input preparation & inits

#10
by cyrilvallez HF Staff - opened

See https://github.com/huggingface/transformers/pull/44130! cache_position are not sliced anymore, and will soon be removed fully (see https://github.com/huggingface/transformers/pull/44181)

cyrilvallez changed pull request title from Fix model input preparation to Fix model input preparation & inits
zqhuang changed pull request status to merged

Sign up or log in to comment