Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
|
@@ -28,7 +28,7 @@ This dataset is exceptionally high-quality supervised fine-tuning conversations
|
|
| 28 |
It prioritize technical correctness, useful engineering judgment, realistic interaction, and efficient explanations over output volume.
|
| 29 |
|
| 30 |
#### This dataset includes:
|
| 31 |
-
- Multi-turns (2-
|
| 32 |
- Dynamic CoT (length)
|
| 33 |
- Dynamic Interleaved Reasoning
|
| 34 |
- Q/A
|
|
|
|
| 28 |
It prioritize technical correctness, useful engineering judgment, realistic interaction, and efficient explanations over output volume.
|
| 29 |
|
| 30 |
#### This dataset includes:
|
| 31 |
+
- Multi-turns (2-6 turns)
|
| 32 |
- Dynamic CoT (length)
|
| 33 |
- Dynamic Interleaved Reasoning
|
| 34 |
- Q/A
|