tweetSentBR public test set

#9
by eduagarcia - opened

For those interested, I have made public a subset of the tweetsentbr dataset: eduagarcia/tweetsentbr_fewshot

It's the same split used for this benchmark, including the few-shot samples. With this, anyone should be able to completely reproduce this benchmark using the Portuguese Language Model Evaluation Harness

eduagarcia changed discussion status to closed

Sign up or log in to comment