Fix a few typos in GenerationMixin's docstring

[sadra-barikbin]

Hi there!

To fix a few typos in GenerationMixin's docstring.

@gante

[sadra-barikbin]

Shouldn't we change the parameter shape in

transformers/src/transformers/generation/utils.py

Line 146 in c8d9840

sequences (`torch.LongTensor` of shape `(batch_size, sequence_length)`):

to (batch_size*num_return_sequences, sequence_length) to account for SampleOutput and its variants which have do_sample=True, num_return_sequences>1?

[gante]

Thank you for the fixes 💛

[gante]

Shouldn't we change the parameter shape in

transformers/src/transformers/generation/utils.py

Line 146 in c8d9840

sequences (`torch.LongTensor` of shape `(batch_size, sequence_length)`):

to (batch_size*num_return_sequences, sequence_length) to account for SampleOutput and its variants which have do_sample=True, num_return_sequences>1?

yes, absolutely!

[ArthurZucker]

🧼 good typo catches !

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.