eren23 leaderboard-pr-bot commited on
Commit
de21b2a
1 Parent(s): b30fc2e

Adding Evaluation Results (#1)

Browse files

- Adding Evaluation Results (01fa73cb1746bbaa2d414a086e3f90fe6dcfe71a)


Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>

Files changed (1) hide show
  1. README.md +121 -5
README.md CHANGED
@@ -1,19 +1,135 @@
1
  ---
2
- license: apache-2.0
3
- datasets:
4
- - jondurbin/truthy-dpo-v0.1
5
  language:
6
  - en
 
7
  library_name: transformers
8
- pipeline_tag: text-generation
9
  tags:
10
  - merge
11
  - dpo
12
  - text-generation-inference
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
13
  ---
14
 
15
  DPO Finetuned paulml/OGNO-7B using jondurbin/truthy-dpo-v0.1
16
 
17
  paulml/OGNO-7B is a mistral 7b variant afaik and this repo is an experimental repo, so might not be useable in prod
18
 
19
- Thx for the great data sources.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
 
 
 
2
  language:
3
  - en
4
+ license: apache-2.0
5
  library_name: transformers
 
6
  tags:
7
  - merge
8
  - dpo
9
  - text-generation-inference
10
+ datasets:
11
+ - jondurbin/truthy-dpo-v0.1
12
+ pipeline_tag: text-generation
13
+ model-index:
14
+ - name: OGNO-7b-dpo-truthful
15
+ results:
16
+ - task:
17
+ type: text-generation
18
+ name: Text Generation
19
+ dataset:
20
+ name: AI2 Reasoning Challenge (25-Shot)
21
+ type: ai2_arc
22
+ config: ARC-Challenge
23
+ split: test
24
+ args:
25
+ num_few_shot: 25
26
+ metrics:
27
+ - type: acc_norm
28
+ value: 72.95
29
+ name: normalized accuracy
30
+ source:
31
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=eren23/OGNO-7b-dpo-truthful
32
+ name: Open LLM Leaderboard
33
+ - task:
34
+ type: text-generation
35
+ name: Text Generation
36
+ dataset:
37
+ name: HellaSwag (10-Shot)
38
+ type: hellaswag
39
+ split: validation
40
+ args:
41
+ num_few_shot: 10
42
+ metrics:
43
+ - type: acc_norm
44
+ value: 89.02
45
+ name: normalized accuracy
46
+ source:
47
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=eren23/OGNO-7b-dpo-truthful
48
+ name: Open LLM Leaderboard
49
+ - task:
50
+ type: text-generation
51
+ name: Text Generation
52
+ dataset:
53
+ name: MMLU (5-Shot)
54
+ type: cais/mmlu
55
+ config: all
56
+ split: test
57
+ args:
58
+ num_few_shot: 5
59
+ metrics:
60
+ - type: acc
61
+ value: 64.61
62
+ name: accuracy
63
+ source:
64
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=eren23/OGNO-7b-dpo-truthful
65
+ name: Open LLM Leaderboard
66
+ - task:
67
+ type: text-generation
68
+ name: Text Generation
69
+ dataset:
70
+ name: TruthfulQA (0-shot)
71
+ type: truthful_qa
72
+ config: multiple_choice
73
+ split: validation
74
+ args:
75
+ num_few_shot: 0
76
+ metrics:
77
+ - type: mc2
78
+ value: 76.61
79
+ source:
80
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=eren23/OGNO-7b-dpo-truthful
81
+ name: Open LLM Leaderboard
82
+ - task:
83
+ type: text-generation
84
+ name: Text Generation
85
+ dataset:
86
+ name: Winogrande (5-shot)
87
+ type: winogrande
88
+ config: winogrande_xl
89
+ split: validation
90
+ args:
91
+ num_few_shot: 5
92
+ metrics:
93
+ - type: acc
94
+ value: 84.69
95
+ name: accuracy
96
+ source:
97
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=eren23/OGNO-7b-dpo-truthful
98
+ name: Open LLM Leaderboard
99
+ - task:
100
+ type: text-generation
101
+ name: Text Generation
102
+ dataset:
103
+ name: GSM8k (5-shot)
104
+ type: gsm8k
105
+ config: main
106
+ split: test
107
+ args:
108
+ num_few_shot: 5
109
+ metrics:
110
+ - type: acc
111
+ value: 68.99
112
+ name: accuracy
113
+ source:
114
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=eren23/OGNO-7b-dpo-truthful
115
+ name: Open LLM Leaderboard
116
  ---
117
 
118
  DPO Finetuned paulml/OGNO-7B using jondurbin/truthy-dpo-v0.1
119
 
120
  paulml/OGNO-7B is a mistral 7b variant afaik and this repo is an experimental repo, so might not be useable in prod
121
 
122
+ Thx for the great data sources.
123
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
124
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_eren23__OGNO-7b-dpo-truthful)
125
+
126
+ | Metric |Value|
127
+ |---------------------------------|----:|
128
+ |Avg. |76.14|
129
+ |AI2 Reasoning Challenge (25-Shot)|72.95|
130
+ |HellaSwag (10-Shot) |89.02|
131
+ |MMLU (5-Shot) |64.61|
132
+ |TruthfulQA (0-shot) |76.61|
133
+ |Winogrande (5-shot) |84.69|
134
+ |GSM8k (5-shot) |68.99|
135
+