About
This dataset contains 3.3K expert-level pairwise human preferences for model responses generated by 6 models in response to 80 MT-bench questions. The annotators are mostly graduate students with expertise in the topic areas of each of the questions.
4 commits
1 contributor
0 downloads
161.6 mb
0 stars
Repository contents
7 tabular files 87.5%
1 text files 12.5%
