Giovanni Puccetti

> Hey, nice work! Left some very minor comments but I still need to look at the HF stuff in more detail. I'll do that later Thanks for looking into...

> For HF stuff from my quick look I have 2 concerns: > > 1. Do all HF models have resize_token_embeddings? > > 2. What is the point of embed_cls?...

> Let me know what you think and also if you're busy. I can get to it myself actually in case you are Hi, I am getting to it now...

@iejMac would it be bad to add the `output_tokens` option to the poolers themselves?

> Why do we want to add it to the poolers? Can't we just use the poolers to get the pooled embedding but also return the tokens on the side?...

> Hmm it would be nice to test this ClsPooler... do you know a single text encoder on huggingface that uses CLS pooling? Maybe BERT? I am trying with Roberta...

> @gpucce any notes from you? Nothing more, writing down what I think I missed: - make model dual tower, I think I misunderstood the point and the idea would...

Oh and last thing this breaks `.generate` should be easy to fix but since there is also another PR working on it I would keep it as is for now.

@iejMac sure I will add it here shortly, I have an idea what the issue is too, later I will try with it a bit and late write here what...

@iejMac this should log them separately