Text Generation
Transformers
Safetensors
Korean
gptj
Inference Endpoints
blueapple8259 commited on
Commit
c1db036
1 Parent(s): 1c551a8

잘못 수정해서 복구

Browse files
Files changed (1) hide show
  1. README.md +11 -3
README.md CHANGED
@@ -1,17 +1,21 @@
1
  ---
2
  license: mit
3
  datasets:
4
- - beomi/KoAlpaca-v1.1a
 
 
5
  language:
6
  - ko
7
  ---
8
 
9
- [kogpt-j-base](https://huggingface.co/heegyu/kogpt-j-base)모델을 [beomi/KoAlpaca-v1.1a](https://huggingface.co/datasets/beomi/KoAlpaca-v1.1a)데이터셋으로 2 epoch만큼 학습시킨 모델입니다.
10
 
11
 
12
  프롬프트:
13
 
14
  ```
 
 
15
  <|im_start|>user
16
  {prompt}<|im_end|>
17
  <|im_start|>assistant
@@ -20,4 +24,8 @@ language:
20
 
21
  데이터셋:
22
 
23
- [beomi/KoAlpaca-v1.1a](https://huggingface.co/datasets/beomi/KoAlpaca-v1.1a)
 
 
 
 
 
1
  ---
2
  license: mit
3
  datasets:
4
+ - nlpai-lab/kullm-v2
5
+ - maywell/ko_wikidata_QA
6
+ - eaglewatch/Korean_Wikipedia_Dataset_for_GPT2_August_2022
7
  language:
8
  - ko
9
  ---
10
 
11
+ [kogpt-j-base](https://huggingface.co/heegyu/kogpt-j-base)모델을 데이터셋 [세](https://huggingface.co/datasets/nlpai-lab/kullm-v2)[가](https://huggingface.co/datasets/maywell/ko_wikidata_QA)[지](https://huggingface.co/datasets/eaglewatch/Korean_Wikipedia_Dataset_for_GPT2_August_2022) 사용해서 45k step(0.27 epoch)만큼 학습시킨 모델입니다.
12
 
13
 
14
  프롬프트:
15
 
16
  ```
17
+ <|im_start|>system
18
+ 당신은 사람들을 도와주는 인공지능 비서입니다. 질문을 읽고 알맞은 답변을 제공하세요.<|im_end|>
19
  <|im_start|>user
20
  {prompt}<|im_end|>
21
  <|im_start|>assistant
 
24
 
25
  데이터셋:
26
 
27
+ [nlpai-lab/kullm-v2](https://huggingface.co/datasets/nlpai-lab/kullm-v2)
28
+
29
+ [maywell/ko_wikidata_QA](https://huggingface.co/datasets/maywell/ko_wikidata_QA)
30
+
31
+ [eaglewatch/Korean_Wikipedia_Dataset_for_GPT2_August_2022](https://huggingface.co/datasets/eaglewatch/Korean_Wikipedia_Dataset_for_GPT2_August_2022)