Alexander Wettig

Hi @void-b583x2-NULL! Thank you for raising this issue. I think we forgot to add the right eval scripts to the repo. It is separate from the training script, because in...

Hi, we haven't packaged this as a python package. Currently, the only way is to clone the repo.

Splitting a large batch by `gradient_accumulation_steps` is a standard feature of the huggingface trainer. Our code additionally accumulates gradients over the segments to save additional memory.

Sorry, I'm not sure what the issue is and it might be related to your setup (e.g., disk space, RAM). Are there any additional error messages?

Hey! We built this library for an old version of transformers and we do not have the bandwidth to keep it up-to-date with every new huggingface version and model architecture....