Update README.md
Browse files
README.md
CHANGED
@@ -3,3 +3,27 @@ license: other
|
|
3 |
license_name: deepseek
|
4 |
license_link: https://github.com/deepseek-ai/DeepSeek-Coder/blob/main/LICENSE-MODEL
|
5 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
license_name: deepseek
|
4 |
license_link: https://github.com/deepseek-ai/DeepSeek-Coder/blob/main/LICENSE-MODEL
|
5 |
---
|
6 |
+
Alternate version of DeepMagic-Coder-7b which can be found bellow.
|
7 |
+
|
8 |
+
- https://huggingface.co/rombodawg/DeepMagic-Coder-7b
|
9 |
+
|
10 |
+

|
11 |
+
|
12 |
+
This version uses a diffrent config setup, with the actual base model of the two merges as the "base". Test both for yourself and see which is better at coding. Benchmarks coming soon.
|
13 |
+
|
14 |
+
Config can be found bellow:
|
15 |
+
```yaml
|
16 |
+
models:
|
17 |
+
- model: deepseek-ai_deepseek-coder-6.7b-instruct
|
18 |
+
parameters:
|
19 |
+
weight: 1
|
20 |
+
- model: ise-uiuc_Magicoder-S-DS-6.7B
|
21 |
+
parameters:
|
22 |
+
weight: 1
|
23 |
+
merge_method: task_arithmetic
|
24 |
+
base_model: deepseek-ai_deepseek-coder-6.7b-base
|
25 |
+
parameters:
|
26 |
+
normalize: true
|
27 |
+
int8_mask: true
|
28 |
+
dtype: float16
|
29 |
+
```
|