Files changed (1) hide show
  1. README.md +124 -8
README.md CHANGED
@@ -1,5 +1,12 @@
1
  ---
 
 
 
2
  license: apache-2.0
 
 
 
 
3
  datasets:
4
  - kobkrit/rd-taxqa
5
  - iapp_wiki_qa_squad
@@ -10,14 +17,110 @@ datasets:
10
  - Thaweewat/gpteacher-20k-th
11
  - Thaweewat/onet-m6-social
12
  - Thaweewat/alpaca-finance-43k-th
13
- language:
14
- - th
15
- - en
16
- library_name: transformers
17
  pipeline_tag: text-generation
18
- tags:
19
- - openthaigpt
20
- - llama
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21
  ---
22
 
23
  # 🇹🇭 OpenThaiGPT 1.0.0-beta
@@ -74,4 +177,17 @@ OpenThaiGPT Volunteers, Artificial Intelligence Entrepreneur Association of Thai
74
  * Kriangkrai Saetan (kraitan.ss21@gmail.com)
75
  * Pitikorn Khlaisamniang (pitikorn32@gmail.com)
76
 
77
- <i>Disclaimer: Provided responses are not guaranteed.</i>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - th
4
+ - en
5
  license: apache-2.0
6
+ library_name: transformers
7
+ tags:
8
+ - openthaigpt
9
+ - llama
10
  datasets:
11
  - kobkrit/rd-taxqa
12
  - iapp_wiki_qa_squad
 
17
  - Thaweewat/gpteacher-20k-th
18
  - Thaweewat/onet-m6-social
19
  - Thaweewat/alpaca-finance-43k-th
 
 
 
 
20
  pipeline_tag: text-generation
21
+ model-index:
22
+ - name: openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
23
+ results:
24
+ - task:
25
+ type: text-generation
26
+ name: Text Generation
27
+ dataset:
28
+ name: AI2 Reasoning Challenge (25-Shot)
29
+ type: ai2_arc
30
+ config: ARC-Challenge
31
+ split: test
32
+ args:
33
+ num_few_shot: 25
34
+ metrics:
35
+ - type: acc_norm
36
+ value: 44.97
37
+ name: normalized accuracy
38
+ source:
39
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=openthaigpt/openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
40
+ name: Open LLM Leaderboard
41
+ - task:
42
+ type: text-generation
43
+ name: Text Generation
44
+ dataset:
45
+ name: HellaSwag (10-Shot)
46
+ type: hellaswag
47
+ split: validation
48
+ args:
49
+ num_few_shot: 10
50
+ metrics:
51
+ - type: acc_norm
52
+ value: 70.19
53
+ name: normalized accuracy
54
+ source:
55
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=openthaigpt/openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
56
+ name: Open LLM Leaderboard
57
+ - task:
58
+ type: text-generation
59
+ name: Text Generation
60
+ dataset:
61
+ name: MMLU (5-Shot)
62
+ type: cais/mmlu
63
+ config: all
64
+ split: test
65
+ args:
66
+ num_few_shot: 5
67
+ metrics:
68
+ - type: acc
69
+ value: 36.22
70
+ name: accuracy
71
+ source:
72
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=openthaigpt/openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
73
+ name: Open LLM Leaderboard
74
+ - task:
75
+ type: text-generation
76
+ name: Text Generation
77
+ dataset:
78
+ name: TruthfulQA (0-shot)
79
+ type: truthful_qa
80
+ config: multiple_choice
81
+ split: validation
82
+ args:
83
+ num_few_shot: 0
84
+ metrics:
85
+ - type: mc2
86
+ value: 49.99
87
+ source:
88
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=openthaigpt/openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
89
+ name: Open LLM Leaderboard
90
+ - task:
91
+ type: text-generation
92
+ name: Text Generation
93
+ dataset:
94
+ name: Winogrande (5-shot)
95
+ type: winogrande
96
+ config: winogrande_xl
97
+ split: validation
98
+ args:
99
+ num_few_shot: 5
100
+ metrics:
101
+ - type: acc
102
+ value: 69.38
103
+ name: accuracy
104
+ source:
105
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=openthaigpt/openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
106
+ name: Open LLM Leaderboard
107
+ - task:
108
+ type: text-generation
109
+ name: Text Generation
110
+ dataset:
111
+ name: GSM8k (5-shot)
112
+ type: gsm8k
113
+ config: main
114
+ split: test
115
+ args:
116
+ num_few_shot: 5
117
+ metrics:
118
+ - type: acc
119
+ value: 1.36
120
+ name: accuracy
121
+ source:
122
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=openthaigpt/openthaigpt-1.0.0-beta-7b-chat-ckpt-hf
123
+ name: Open LLM Leaderboard
124
  ---
125
 
126
  # 🇹🇭 OpenThaiGPT 1.0.0-beta
 
177
  * Kriangkrai Saetan (kraitan.ss21@gmail.com)
178
  * Pitikorn Khlaisamniang (pitikorn32@gmail.com)
179
 
180
+ <i>Disclaimer: Provided responses are not guaranteed.</i>
181
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
182
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_openthaigpt__openthaigpt-1.0.0-beta-7b-chat-ckpt-hf)
183
+
184
+ | Metric |Value|
185
+ |---------------------------------|----:|
186
+ |Avg. |45.35|
187
+ |AI2 Reasoning Challenge (25-Shot)|44.97|
188
+ |HellaSwag (10-Shot) |70.19|
189
+ |MMLU (5-Shot) |36.22|
190
+ |TruthfulQA (0-shot) |49.99|
191
+ |Winogrande (5-shot) |69.38|
192
+ |GSM8k (5-shot) | 1.36|
193
+