Harryis Wang

> > > Just to record some information about this error. datasets==2.14.5, transformers==4.36.2, pyarrow==10.0.0 will need this copy(copied the "exact_match" directory from evaluate module to ./lm-evaluation-harness/) but latter I encountered...

Thanks a lot for your reply and quick revision! So in the original code, you tag all the unlabeled vicuna_dummy_data with 'example_id = 0'? I wonder what's the point to...

Thanks a lot for your explaination! Is it because of the faster inference of llama-2-70b ckpt code that you choose to use it rather than huggingface code? The past_key_values in...

Thanks a lot for your reply! It's clear for me that you use 20 instruction types to generate 120K synthetic topics, which means every instruction type will generate about 120k/20...

Oh,I get it!Thanks!

Thanks for your reply! I use the `peft==0.6.3.dev0` version and I think that's the point.