- h
ReTool-SFT-multi-turn
- huggingface.co
Updated May 11, 2025
Share
Facebook
Twitter
EmailClick to copy linkLink copiedCiteXiang Long (2025). ReTool-SFT-multi-turn [Dataset]. https://huggingface.co/datasets/swordfaith/ReTool-SFT-multi-turnDataset updatedMay 11, 2025AuthorsXiang LongLicenseApache License, v2.0https://www.apache.org/licenses/LICENSE-2.0
License information was derived automaticallyDescriptionReTool-SFT-multi-turn: Cold-start Multi-turn SFT Data for Tool Use
framework release blog: https://github.com/zhaochenyang20/Awesome-ML-SYS-Tutorial/blob/main/rlhf/verl/multi-turn/verl-multiturn-rollout-Release.md This dataset contains ReTool-like cold-start multi-turn supervised fine-tuning (SFT) data specifically designed for training language models in strategic tool use. It serves as the cold-start training data for the verl-SGLang multi-turn framework, enabling models to learn… See the full description on the dataset page: https://huggingface.co/datasets/swordfaith/ReTool-SFT-multi-turn.
- h
gsm8k-python
- huggingface.co
Updated Sep 18, 2025ShareFacebookTwitterEmailClick to copy linkLink copiedCitesiyuanzhu (2025). gsm8k-python [Dataset]. https://huggingface.co/datasets/siyuan-zhu/gsm8k-pythonDataset updatedSep 18, 2025AuthorssiyuanzhuLicenseMIT Licensehttps://opensource.org/licenses/MIT
License information was derived automaticallyDescriptionA GSM8K dataset for multi-step python tool calling in LLM. Formatted for VERL SFT stage.
Not seeing a result you expected?
Learn how you can add new datasets to our index.
FacebookTwitterEmailApache License, v2.0https://www.apache.org/licenses/LICENSE-2.0
License information was derived automatically
ReTool-SFT-multi-turn: Cold-start Multi-turn SFT Data for Tool Use
framework release blog: https://github.com/zhaochenyang20/Awesome-ML-SYS-Tutorial/blob/main/rlhf/verl/multi-turn/verl-multiturn-rollout-Release.md This dataset contains ReTool-like cold-start multi-turn supervised fine-tuning (SFT) data specifically designed for training language models in strategic tool use. It serves as the cold-start training data for the verl-SGLang multi-turn framework, enabling models to learn… See the full description on the dataset page: https://huggingface.co/datasets/swordfaith/ReTool-SFT-multi-turn.