Update README.md
Browse files
README.md
CHANGED
@@ -1,10 +1,14 @@
|
|
1 |
-
|
|
|
|
|
|
|
2 |
base_model : yanolja/KoSOLAR-10.7B-v0.2
|
|
|
|
|
|
|
3 |
|
4 |
-
|
5 |
-
|
6 |
-
Deduplicating Training Data Makes Language Models Better 알고리즘 활용
|
7 |
-
Code
|
8 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
9 |
import torch
|
10 |
|
@@ -13,4 +17,7 @@ model = AutoModelForCausalLM.from_pretrained(
|
|
13 |
model_name,
|
14 |
)
|
15 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
|
|
16 |
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
---
|
4 |
+
## Model
|
5 |
base_model : yanolja/KoSOLAR-10.7B-v0.2
|
6 |
+
## Dataset
|
7 |
+
* 공개 데이터 수집
|
8 |
+
* Deduplicating Training Data Makes Language Models Better 알고리즘 활용
|
9 |
|
10 |
+
## Code
|
11 |
+
```python
|
|
|
|
|
12 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
13 |
import torch
|
14 |
|
|
|
17 |
model_name,
|
18 |
)
|
19 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
20 |
+
```
|
21 |
|
22 |
+
## Benchmark
|
23 |
+
**[Ko-LLM-Leaderboard](https://huggingface.co/spaces/upstage/open-ko-llm-leaderboard)**
|