Clarify in the README how the samples are selected

yxuansu / OpenAlpaca

OpenAlpaca: A Fully Open-Source Instruction-Following Model Based On OpenLLaMA

Apache License 2.0

301 stars 35 forks source link

Closed codesoap closed 1 year ago

codesoap commented 1 year ago

Thanks for your work! It's great to have a truly open alternative to the (Stanford) alpaca.

I'm suggesting to clarify the maximum length for the samples that are used for the finetuning. When I read the README, I missed this information.

yxuansu commented 1 year ago

Hi @codesoap, thank you for your suggestion. We have made this point more visible in our updated codebase. Please refer to here for more information.