Dataset · Alibaba-Apsara
435k Examples for Long-CoT Training
Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b
435k English SFT examples for training models with long chains of thought.
What's inside
An open dataset of 435k examples designed for training Long Chain-of-Thought models. According to its card, it serves as a high-quality collection to democratize training for such models; source details are not fully disclosed in the available fragment. The language is English and the modality is text.
What you can do with it
- Supervised fine-tuning of Long-CoT reasoning models
- Fine-tuning gpt-oss-120b or a smaller model
- Evaluating long chains of reasoning
- Tasks
- Text generation
- Languages
- English
- Size
- 100K–1M rows
- Downloads this month
- 3.3K
- Likes
- 364
Models trained on it
- mradermacher/DASD-30B-A3B-Thinking-Preview-i1-GGUF 1.6K
- mradermacher/Cicikus-v3-1.4B-i1-GGUF 874
- mradermacher/Cicikus_v2_3B-i1-GGUF 777
- Mungert/DASD-4B-Thinking-GGUFText generation 616
- mradermacher/Cicikus-v3-1.4B-Opus4.6-Powered-i1-GGUF 578
- mradermacher/Cicikus_PTHS_v3_4.4B-i1-GGUF 531
- mradermacher/Cicikus_v2_3B-GGUF 469
- mradermacher/DASD-30B-A3B-Thinking-Preview-GGUF 459
How to load it
from datasets import load_dataset
ds = load_dataset("Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b")