qihoo360/Light-R1-SFTData
Light-R1: Surpassing R1-Distill from Scratch* with $1000 through Curriculum SFT & DPO *from models without long COT technical report GitHub page Here are the two-stage SFT data we used to train Light-R1-32B. Simply refer to stage1-76k.json and stage2-3k.json Model Trained From Release Date AIME24 AIME25 DeepSeek-R1-Distill-Llama-70B Llama-3.3-70B-Instruct 25.1.20 70.0 54.1 DeepSeek-R1-Distill-Qwen-32B Qwen2.5-32B 25.1.20 72.6 54.9 LIMO (32B) Qwen2.5-32B-Instruct… See the full description on the dataset page: https://huggingface.co/datasets/qihoo360/Light-R1-SFTData.
| Author / organization | qihoo360 |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 6.6K (as reported by Hugging Face) |
| Likes | 61 |
| Created | 4 March 2025 |
| Last updated at source | 17 March 2025 |
Popularity, last 2 days
6,570 now (-62 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- mradermacher/Zhi-Create-DSR1-14B-GGUFModel · used to train
- mradermacher/Fast-Math-R1-14B-GGUFModel · used to train
- mradermacher/Zhi-Create-DSR1-14B-i1-GGUFModel · used to train
- mradermacher/Zhi-writing-dsr1-14b-i1-GGUFModel · used to train
- mradermacher/Zhi-writing-dsr1-14b-GGUFModel · used to train
Recent changes
- 17 Oct 2026 Description: Light-R1: Surpassing R1-Distill from Scratch* with $1000 through Curriculum SFT & DPO *from models without long COT technical report GitHub page Here are the two-stage SFT data we used to train Light-R1-32B. Simply refer to stage1-76k.json → Light-R1: Surpassing R1-Distill from Scratch* with $1000 through Curriculum SFT & DPO *from models without long COT technical report GitHub page Here are the two-stage SFT data we used to train Light-R1-32B. Simply refer to stage1-76k.json
Changes detected by Silytics between refreshes of the source.
More from qihoo360
- qihoo360/DCI-CNDataset
- qihoo360/WISA-80KDataset
- qihoo360/DOCCI-CNDataset
- qihoo360/InduOCRBenchDataset
- qihoo360/RevealLayer-100KDataset
- qihoo360/FineHARDDataset
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 Information may be incomplete or out of date.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.