Update README.md
Browse files
README.md
CHANGED
@@ -4,47 +4,6 @@ language:
|
|
4 |
- en
|
5 |
- ko
|
6 |
|
7 |
-
model-index:
|
8 |
-
- name: ARC
|
9 |
-
results:
|
10 |
-
- task:
|
11 |
-
type: natural-language-generation
|
12 |
-
dataset:
|
13 |
-
type: huggingface
|
14 |
-
name: ARC-Challenge(25 shots)
|
15 |
-
metrics:
|
16 |
-
- type: acc_norm
|
17 |
-
value: 53.24
|
18 |
-
- name: HellaSwag
|
19 |
-
results:
|
20 |
-
- task:
|
21 |
-
type: natural-language-generation
|
22 |
-
dataset:
|
23 |
-
type: huggingface
|
24 |
-
name: HellaSwag(10 shots)
|
25 |
-
metrics:
|
26 |
-
- type: acc_norm
|
27 |
-
value: 78.82
|
28 |
-
- name: MMLU
|
29 |
-
results:
|
30 |
-
- task:
|
31 |
-
type: natural-language-generation
|
32 |
-
dataset:
|
33 |
-
type: huggingface
|
34 |
-
name: MMLU(5 shots)
|
35 |
-
metrics:
|
36 |
-
- type: acc
|
37 |
-
value: 42.38
|
38 |
-
- name: TruthfulQA
|
39 |
-
results:
|
40 |
-
- task:
|
41 |
-
type: natural-language-generation
|
42 |
-
dataset:
|
43 |
-
type: huggingface
|
44 |
-
name: TruthfulQA(0 shots)
|
45 |
-
metrics:
|
46 |
-
- type: MC2
|
47 |
-
value: 44.65
|
48 |
---
|
49 |
|
50 |
# Fine-tunned Llama-2 Model with Korean Corpus
|
|
|
4 |
- en
|
5 |
- ko
|
6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
---
|
8 |
|
9 |
# Fine-tunned Llama-2 Model with Korean Corpus
|