Update README.md
Browse files
README.md
CHANGED
|
@@ -2,8 +2,12 @@
|
|
| 2 |
title: Qwen3-4B Claude Reasoning
|
| 3 |
emoji: 🧠
|
| 4 |
colorFrom: indigo
|
|
|
|
| 5 |
colorTo: pink
|
|
|
|
| 6 |
sdk: gradio
|
|
|
|
|
|
|
| 7 |
pinned: true
|
| 8 |
datasets:
|
| 9 |
- Liontix/claude-sonnet-4-100x
|
|
@@ -12,7 +16,8 @@ base_model:
|
|
| 12 |
- unsloth/Qwen3-4B-unsloth-bnb-4bit
|
| 13 |
---
|
| 14 |
|
| 15 |
-
# Qwen3-4B Claude Sonnet Reasoning Distill
|
|
|
|
| 16 |
|
| 17 |
This model was trained on a **Claude Sonnet 4 (non-reasoning)** dataset and a **Claude Sonnet 3.7 (reasoning)** dataset.
|
| 18 |
|
|
@@ -29,3 +34,11 @@ If you want to fine-tune this model:
|
|
| 29 |
|
| 30 |
Prompt format uses Claude-style `<|im_start|>` / `<|im_end|>` markers with role tags.
|
| 31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
title: Qwen3-4B Claude Reasoning
|
| 3 |
emoji: 🧠
|
| 4 |
colorFrom: indigo
|
| 5 |
+
|
| 6 |
colorTo: pink
|
| 7 |
+
|
| 8 |
sdk: gradio
|
| 9 |
+
|
| 10 |
+
|
| 11 |
pinned: true
|
| 12 |
datasets:
|
| 13 |
- Liontix/claude-sonnet-4-100x
|
|
|
|
| 16 |
- unsloth/Qwen3-4B-unsloth-bnb-4bit
|
| 17 |
---
|
| 18 |
|
| 19 |
+
# Qwen3-4B Claude Sonnet Reasoning Distill
|
| 20 |
+
|
| 21 |
|
| 22 |
This model was trained on a **Claude Sonnet 4 (non-reasoning)** dataset and a **Claude Sonnet 3.7 (reasoning)** dataset.
|
| 23 |
|
|
|
|
| 34 |
|
| 35 |
Prompt format uses Claude-style `<|im_start|>` / `<|im_end|>` markers with role tags.
|
| 36 |
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
|