Where is the code for evaluation on benchmarks like human-eval etc

#28
by senxiangms - opened

I checked https://github.com/bigcode-project/Megatron-LM/blob/multi-query-attention/,
but don't know which folder has evaluation code.

thanks.

BigCode org

you can use the bigcode-evaluation-harness

loubnabnl changed discussion status to closed

Sign up or log in to comment