tasksource/tasksource-instruct
tasksource-instruct Instruction-tuning data recast from the ~480 English classification, multiple-choice and token-classification tasks of tasksource. Every example comes from a human-built dataset (NLI, logical reasoning, sentiment, hate speech, discourse, argumentation, ...), not from a teacher model. Each task is capped at 30k training examples, so no task dominates. Many tasks aren't in FLAN v2, for example DynaSent, DynaHate, discriminative bAbI, epistemic logic, RuleTaker… See the full description on the dataset page: https://huggingface.co/datasets/tasksource/tasksource-instruct.
| Author / organization | tasksource |
|---|---|
| License (as declared) | other |
| Downloads | 3K (as reported by Hugging Face) |
| Likes | 25 |
| Created | 24 May 2023 |
| Last updated at source | 27 September 2026 |
Popularity, last 2 days
3,043 now (-49 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
More from tasksource
- tasksource/mmluDataset
- tasksource/bigbenchDataset
- tasksource/reclorDataset
- tasksource/tasksource-jev-typed-decisionsDataset
- tasksource/babi_nliDataset
- tasksource/strategy-qaDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.