Files changed (1) hide show
  1. README.md +137 -27
README.md CHANGED
@@ -1,34 +1,13 @@
1
  ---
2
- language:
3
  - en
4
  - fr
5
  - ro
6
  - de
7
  - multilingual
8
-
9
  tags:
10
  - text2text-generation
11
-
12
- widget:
13
- - text: "Translate to German: My name is Arthur"
14
- example_title: "Translation"
15
- - text: "Please answer to the following question. Who is going to be the next Ballon d'or?"
16
- example_title: "Question Answering"
17
- - text: "Q: Can Geoffrey Hinton have a conversation with George Washington? Give the rationale before answering."
18
- example_title: "Logical reasoning"
19
- - text: "Please answer the following question. What is the boiling point of Nitrogen?"
20
- example_title: "Scientific knowledge"
21
- - text: "Answer the following yes/no question. Can you write a whole Haiku in a single tweet?"
22
- example_title: "Yes/no question"
23
- - text: "Answer the following yes/no question by reasoning step-by-step. Can you write a whole Haiku in a single tweet?"
24
- example_title: "Reasoning task"
25
- - text: "Q: ( False or not False or False ) is? A: Let's think step by step"
26
- example_title: "Boolean Expressions"
27
- - text: "The square root of x is the cube root of y. What is y to the power of 2, if x = 4?"
28
- example_title: "Math reasoning"
29
- - text: "Premise: At my age you will probably have learnt one lesson. Hypothesis: It's not certain how many lessons you'll learn by your thirties. Does the premise entail the hypothesis?"
30
- example_title: "Premise and hypothesis"
31
-
32
  datasets:
33
  - svakulenk0/qrecc
34
  - taskmaster2
@@ -40,9 +19,127 @@ datasets:
40
  - esnli
41
  - quasc
42
  - qed
43
-
44
-
45
- license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46
  ---
47
 
48
  # Model Card for FLAN-T5 small
@@ -273,4 +370,17 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
273
 
274
  copyright = {Creative Commons Attribution 4.0 International}
275
  }
276
- ```
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
  - en
4
  - fr
5
  - ro
6
  - de
7
  - multilingual
8
+ license: apache-2.0
9
  tags:
10
  - text2text-generation
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
11
  datasets:
12
  - svakulenk0/qrecc
13
  - taskmaster2
 
19
  - esnli
20
  - quasc
21
  - qed
22
+ widget:
23
+ - text: 'Translate to German: My name is Arthur'
24
+ example_title: Translation
25
+ - text: Please answer to the following question. Who is going to be the next Ballon
26
+ d'or?
27
+ example_title: Question Answering
28
+ - text: 'Q: Can Geoffrey Hinton have a conversation with George Washington? Give the
29
+ rationale before answering.'
30
+ example_title: Logical reasoning
31
+ - text: Please answer the following question. What is the boiling point of Nitrogen?
32
+ example_title: Scientific knowledge
33
+ - text: Answer the following yes/no question. Can you write a whole Haiku in a single
34
+ tweet?
35
+ example_title: Yes/no question
36
+ - text: Answer the following yes/no question by reasoning step-by-step. Can you write
37
+ a whole Haiku in a single tweet?
38
+ example_title: Reasoning task
39
+ - text: 'Q: ( False or not False or False ) is? A: Let''s think step by step'
40
+ example_title: Boolean Expressions
41
+ - text: The square root of x is the cube root of y. What is y to the power of 2, if
42
+ x = 4?
43
+ example_title: Math reasoning
44
+ - text: 'Premise: At my age you will probably have learnt one lesson. Hypothesis: It''s
45
+ not certain how many lessons you''ll learn by your thirties. Does the premise
46
+ entail the hypothesis?'
47
+ example_title: Premise and hypothesis
48
+ model-index:
49
+ - name: flan-t5-small
50
+ results:
51
+ - task:
52
+ type: text-generation
53
+ name: Text Generation
54
+ dataset:
55
+ name: IFEval (0-Shot)
56
+ type: HuggingFaceH4/ifeval
57
+ args:
58
+ num_few_shot: 0
59
+ metrics:
60
+ - type: inst_level_strict_acc and prompt_level_strict_acc
61
+ value: 15.24
62
+ name: strict accuracy
63
+ source:
64
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=google/flan-t5-small
65
+ name: Open LLM Leaderboard
66
+ - task:
67
+ type: text-generation
68
+ name: Text Generation
69
+ dataset:
70
+ name: BBH (3-Shot)
71
+ type: BBH
72
+ args:
73
+ num_few_shot: 3
74
+ metrics:
75
+ - type: acc_norm
76
+ value: 6.36
77
+ name: normalized accuracy
78
+ source:
79
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=google/flan-t5-small
80
+ name: Open LLM Leaderboard
81
+ - task:
82
+ type: text-generation
83
+ name: Text Generation
84
+ dataset:
85
+ name: MATH Lvl 5 (4-Shot)
86
+ type: hendrycks/competition_math
87
+ args:
88
+ num_few_shot: 4
89
+ metrics:
90
+ - type: exact_match
91
+ value: 0.0
92
+ name: exact match
93
+ source:
94
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=google/flan-t5-small
95
+ name: Open LLM Leaderboard
96
+ - task:
97
+ type: text-generation
98
+ name: Text Generation
99
+ dataset:
100
+ name: GPQA (0-shot)
101
+ type: Idavidrein/gpqa
102
+ args:
103
+ num_few_shot: 0
104
+ metrics:
105
+ - type: acc_norm
106
+ value: 1.45
107
+ name: acc_norm
108
+ source:
109
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=google/flan-t5-small
110
+ name: Open LLM Leaderboard
111
+ - task:
112
+ type: text-generation
113
+ name: Text Generation
114
+ dataset:
115
+ name: MuSR (0-shot)
116
+ type: TAUR-Lab/MuSR
117
+ args:
118
+ num_few_shot: 0
119
+ metrics:
120
+ - type: acc_norm
121
+ value: 10.37
122
+ name: acc_norm
123
+ source:
124
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=google/flan-t5-small
125
+ name: Open LLM Leaderboard
126
+ - task:
127
+ type: text-generation
128
+ name: Text Generation
129
+ dataset:
130
+ name: MMLU-PRO (5-shot)
131
+ type: TIGER-Lab/MMLU-Pro
132
+ config: main
133
+ split: test
134
+ args:
135
+ num_few_shot: 5
136
+ metrics:
137
+ - type: acc
138
+ value: 2.59
139
+ name: accuracy
140
+ source:
141
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=google/flan-t5-small
142
+ name: Open LLM Leaderboard
143
  ---
144
 
145
  # Model Card for FLAN-T5 small
 
370
 
371
  copyright = {Creative Commons Attribution 4.0 International}
372
  }
373
+ ```
374
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
375
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_google__flan-t5-small)
376
+
377
+ | Metric |Value|
378
+ |-------------------|----:|
379
+ |Avg. | 6.00|
380
+ |IFEval (0-Shot) |15.24|
381
+ |BBH (3-Shot) | 6.36|
382
+ |MATH Lvl 5 (4-Shot)| 0.00|
383
+ |GPQA (0-shot) | 1.45|
384
+ |MuSR (0-shot) |10.37|
385
+ |MMLU-PRO (5-shot) | 2.59|
386
+