Update README.md
Browse files
README.md
CHANGED
|
@@ -21,3 +21,13 @@ configs:
|
|
| 21 |
- split: train
|
| 22 |
path: data/train-*
|
| 23 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
- split: train
|
| 22 |
path: data/train-*
|
| 23 |
---
|
| 24 |
+
|
| 25 |
+
Translating [Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled](https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled)'s training dataset via TranslateGemma.
|
| 26 |
+
|
| 27 |
+
The dataset consists of high-quality, filtered reasoning distillation data:
|
| 28 |
+
|
| 29 |
+
| Dataset Name | Description / Purpose |
|
| 30 |
+
|--------------|-----------------------|
|
| 31 |
+
| [nohurry/Opus-4.6-Reasoning-3000x-filtered](https://huggingface.co/datasets/nohurry/Opus-4.6-Reasoning-3000x-filtered) | Provides comprehensive Claude 4.6 Opus reasoning trajectories. |
|
| 32 |
+
| [Jackrong/Qwen3.5-reasoning-700x](https://huggingface.co/datasets/Jackrong/Qwen3.5-reasoning-700x) | Additional curated reasoning samples designed to strengthen structured step-by-step problem solving and improve reasoning diversity. |
|
| 33 |
+
|