| StackExchange Paired 500K is a subset of lvwerra/stack-exchange-paired | |
| which is a processed version of the HuggingFaceH4/stack-exchange-preferences. The following steps were applied: | |
| Parse HTML to Markdown with markdownify | |
| Create pairs (response_j, response_k) where j was rated better than k | |
| Sample at most 10 pairs per question | |
| Shuffle the dataset globally | |
| This dataset is designed to be used for preference learning. | |
| --- | |
| license: mit | |
| --- | |