Skip to content

Regarding MCQA with LLAMA model #2

Description

@ayushp88

Hi, I am training a MCQA dataset with LLAMA model. I am restricting the output space of LLAMA to 4 tokens. I am getting this error upon running the code
shift_logits = shift_logits.view(-1, self.config.vocab_size)
RuntimeError: shape '[-1, 32000]' is invalid for input of size 1052
0%| | 0/927 [00:00<?, ?it/s]
Could you please help.
Thanks

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions