Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,29 @@
|
|
1 |
---
|
2 |
-
license:
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
tags:
|
4 |
+
- merge
|
5 |
---
|
6 |
+
|
7 |
+
# etri-xainlp/SOLAR-10.7B-merge-dpo_v1
|
8 |
+
|
9 |
+
## Model Details
|
10 |
+
|
11 |
+
**Model Developers** ETRI xainlp team
|
12 |
+
|
13 |
+
**Input** text only.
|
14 |
+
|
15 |
+
**Output** text only.
|
16 |
+
|
17 |
+
**Model Architecture**
|
18 |
+
|
19 |
+
We used MergeKit to merge Model heavytail/kullm-solar into Model etri-xainlp/SOLAR-10.7B-merge-dpo as the base.
|
20 |
+
|
21 |
+
**Base Model** [etri-xainlp/SOLAR-10.7B-merge-dpo](https://huggingface.co/etri-xainlp/SOLAR-10.7B-merge-dpo)
|
22 |
+
|
23 |
+
**Merge Model** [davidkim205/komt-solar-10.7b-sft-v5](https://huggingface.co/davidkim205/komt-solar-10.7b-sft-v5)
|
24 |
+
|
25 |
+
**Training Dataset**
|
26 |
+
|
27 |
+
- dpo+lora: 100k user preference set
|
28 |
+
|
29 |
+
- We use A100 GPU 80GB * 8, when training.
|