---
dataset_info:
  features:
  - name: system
    dtype: string
  - name: conversations
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  splits:
  - name: train
    num_bytes: 799035883.52
    num_examples: 80000
  download_size: 799035883.52
  dataset_size: 799035883.52
configs:
- config_name: no_duplicate
  data_files:
    - split: train
      path: data/no_duplicate.jsonl
- config_name: no_duplicate_80k
  data_files:
    - split: train
      path: data/no_duplicate_80k.jsonl
- config_name: no_duplicate_40k
  data_files:
    - split: train
      path: data/no_duplicate_40k.jsonl
- config_name: no_duplicate_20k
  data_files:
    - split: train
      path: data/no_duplicate_20k.jsonl
---

## License
Apache 2.0

## Dataset Information

This dataset contains hybrid thinking training data with the following characteristics:

- Files: 1
- Total samples: 80000
- Total size: 762.02 MB

### File Details

- **no_duplicate.jsonl**: 80000 samples, 762.02 MB