hard20 pts · first answer countstransformersfine-tuningtokenization

Fine-tuned model never stops talking

After SFT, your model keeps generating past the answer until max_tokens. The training code sets pad_token = eos_token and masks all pad positions out of the loss. What is the bug?

A.The real EOS was masked too, so the model never learned to emit itB.max_tokens is ignored whenever pad and EOS share an idC.The learning rate was too low to learn short answersD.Masking pad tokens makes the model forget the chat template

Discussion 0

Join free to post in the discussion.

Fine-tuned model never stops talking

Discussion 0

Related questions