Hugging Face's logo

Datasets:
mlfoundations-dev
/
OpenThinker2-7B_eval_8067

Modalities:
Tabular
Text
Formats:
parquet
Size:
1K - 10K
Libraries:
Datasets
Dask
Dataset card Data Studio Files
xet
 Community
neginr commited on
Commit
4f40bfd
·
verified ·
1 Parent(s): 11997d4

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +47 -46
README.md CHANGED
@@ -2,125 +2,126 @@
2
 
3
  Precomputed model outputs for evaluation.
4
 
 
5
  ## Evaluation Results
6
 
7
  ### Summary
8
 
9
  | Metric | AIME24 | AMC23 | MATH500 | AIME25 | GPQADiamond | MMLUPro | LiveCodeBench | CodeElo | JEEBench |
10
  |--------|------|-----|-------|------|-----------|-------|-------------|-------|--------|
11
- | Accuracy | 28.7 | 43.2 | 88.6 | 16.7 | 44.8 | 31.0 | 53.8 | 22.6 | 64.8 |
12
 
13
  ### AIME24
14
 
15
- - **Average Accuracy**: 28.67% ± 9.14%
16
  - **Number of Runs**: 10
17
 
18
  | Run | Accuracy | Questions Solved | Total Questions |
19
  |-----|----------|-----------------|----------------|
20
  | 1 | 66.67% | 20 | 30 |
21
- | 2 | 50.00% | 15 | 30 |
22
- | 3 | 56.67% | 17 | 30 |
23
- | 4 | 56.67% | 17 | 30 |
24
  | 5 | 56.67% | 17 | 30 |
25
- | 6 | 0.00% | 0 | 30 |
26
- | 7 | 0.00% | 0 | 30 |
27
- | 8 | 0.00% | 0 | 30 |
28
- | 9 | 0.00% | 0 | 30 |
29
- | 10 | 0.00% | 0 | 30 |
30
 
31
  ### AMC23
32
 
33
- - **Average Accuracy**: 43.25% ± 13.69%
34
  - **Number of Runs**: 10
35
 
36
  | Run | Accuracy | Questions Solved | Total Questions |
37
  |-----|----------|-----------------|----------------|
38
- | 1 | 87.50% | 35 | 40 |
39
- | 2 | 85.00% | 34 | 40 |
40
- | 3 | 82.50% | 33 | 40 |
41
  | 4 | 87.50% | 35 | 40 |
42
  | 5 | 90.00% | 36 | 40 |
43
- | 6 | 0.00% | 0 | 40 |
44
- | 7 | 0.00% | 0 | 40 |
45
- | 8 | 0.00% | 0 | 40 |
46
- | 9 | 0.00% | 0 | 40 |
47
- | 10 | 0.00% | 0 | 40 |
48
 
49
  ### MATH500
50
 
51
- - **Accuracy**: 88.60%
52
  | Accuracy | Questions Solved | Total Questions |
53
  |----------|-----------------|----------------|
54
- | 88.60% | 443 | 500 |
55
 
56
  ### AIME25
57
 
58
- - **Average Accuracy**: 16.67% ± 5.29%
59
  - **Number of Runs**: 10
60
 
61
  | Run | Accuracy | Questions Solved | Total Questions |
62
  |-----|----------|-----------------|----------------|
63
- | 1 | 36.67% | 11 | 30 |
64
  | 2 | 30.00% | 9 | 30 |
65
  | 3 | 33.33% | 10 | 30 |
66
  | 4 | 33.33% | 10 | 30 |
67
  | 5 | 33.33% | 10 | 30 |
68
- | 6 | 0.00% | 0 | 30 |
69
- | 7 | 0.00% | 0 | 30 |
70
- | 8 | 0.00% | 0 | 30 |
71
- | 9 | 0.00% | 0 | 30 |
72
- | 10 | 0.00% | 0 | 30 |
73
 
74
  ### GPQADiamond
75
 
76
- - **Average Accuracy**: 44.78% ± 0.77%
77
  - **Number of Runs**: 3
78
 
79
  | Run | Accuracy | Questions Solved | Total Questions |
80
  |-----|----------|-----------------|----------------|
81
- | 1 | 42.93% | 85 | 198 |
82
- | 2 | 45.96% | 91 | 198 |
83
- | 3 | 45.45% | 90 | 198 |
84
 
85
  ### MMLUPro
86
 
87
- - **Average Accuracy**: 31.00% ± 0.00%
88
  - **Number of Runs**: 1
89
 
90
  | Run | Accuracy | Questions Solved | Total Questions |
91
  |-----|----------|-----------------|----------------|
92
- | 1 | 31.00% | 155 | 500 |
93
 
94
  ### LiveCodeBench
95
 
96
- - **Average Accuracy**: 53.75% ± 0.62%
97
  - **Number of Runs**: 3
98
 
99
  | Run | Accuracy | Questions Solved | Total Questions |
100
  |-----|----------|-----------------|----------------|
101
- | 1 | 53.82% | 275 | 511 |
102
- | 2 | 52.64% | 269 | 511 |
103
- | 3 | 54.79% | 280 | 511 |
104
 
105
  ### CodeElo
106
 
107
- - **Average Accuracy**: 22.59% ± 1.01%
108
  - **Number of Runs**: 3
109
 
110
  | Run | Accuracy | Questions Solved | Total Questions |
111
  |-----|----------|-----------------|----------------|
112
- | 1 | 21.99% | 86 | 391 |
113
- | 2 | 24.55% | 96 | 391 |
114
- | 3 | 21.23% | 83 | 391 |
115
 
116
  ### JEEBench
117
 
118
- - **Average Accuracy**: 64.79% ± 0.81%
119
  - **Number of Runs**: 3
120
 
121
  | Run | Accuracy | Questions Solved | Total Questions |
122
  |-----|----------|-----------------|----------------|
123
- | 1 | 66.02% | 340.0 | 515 |
124
- | 2 | 62.82% | 323.5 | 515 |
125
- | 3 | 65.53% | 337.5 | 515 |
126
 
 
2
 
3
  Precomputed model outputs for evaluation.
4
 
5
+
6
  ## Evaluation Results
7
 
8
  ### Summary
9
 
10
  | Metric | AIME24 | AMC23 | MATH500 | AIME25 | GPQADiamond | MMLUPro | LiveCodeBench | CodeElo | JEEBench |
11
  |--------|------|-----|-------|------|-----------|-------|-------------|-------|--------|
12
+ | Accuracy | 58.0 | 87.2 | 88.0 | 35.3 | 47.5 | 31.6 | 55.6 | 24.2 | 66.1 |
13
 
14
  ### AIME24
15
 
16
+ - **Average Accuracy**: 58.00% ± 1.43%
17
  - **Number of Runs**: 10
18
 
19
  | Run | Accuracy | Questions Solved | Total Questions |
20
  |-----|----------|-----------------|----------------|
21
  | 1 | 66.67% | 20 | 30 |
22
+ | 2 | 56.67% | 17 | 30 |
23
+ | 3 | 53.33% | 16 | 30 |
24
+ | 4 | 53.33% | 16 | 30 |
25
  | 5 | 56.67% | 17 | 30 |
26
+ | 6 | 56.67% | 17 | 30 |
27
+ | 7 | 63.33% | 19 | 30 |
28
+ | 8 | 56.67% | 17 | 30 |
29
+ | 9 | 63.33% | 19 | 30 |
30
+ | 10 | 53.33% | 16 | 30 |
31
 
32
  ### AMC23
33
 
34
+ - **Average Accuracy**: 87.25% ± 0.90%
35
  - **Number of Runs**: 10
36
 
37
  | Run | Accuracy | Questions Solved | Total Questions |
38
  |-----|----------|-----------------|----------------|
39
+ | 1 | 90.00% | 36 | 40 |
40
+ | 2 | 87.50% | 35 | 40 |
41
+ | 3 | 87.50% | 35 | 40 |
42
  | 4 | 87.50% | 35 | 40 |
43
  | 5 | 90.00% | 36 | 40 |
44
+ | 6 | 85.00% | 34 | 40 |
45
+ | 7 | 90.00% | 36 | 40 |
46
+ | 8 | 80.00% | 32 | 40 |
47
+ | 9 | 87.50% | 35 | 40 |
48
+ | 10 | 87.50% | 35 | 40 |
49
 
50
  ### MATH500
51
 
52
+ - **Accuracy**: 88.00%
53
  | Accuracy | Questions Solved | Total Questions |
54
  |----------|-----------------|----------------|
55
+ | 88.00% | 440 | 500 |
56
 
57
  ### AIME25
58
 
59
+ - **Average Accuracy**: 35.33% ± 2.22%
60
  - **Number of Runs**: 10
61
 
62
  | Run | Accuracy | Questions Solved | Total Questions |
63
  |-----|----------|-----------------|----------------|
64
+ | 1 | 40.00% | 12 | 30 |
65
  | 2 | 30.00% | 9 | 30 |
66
  | 3 | 33.33% | 10 | 30 |
67
  | 4 | 33.33% | 10 | 30 |
68
  | 5 | 33.33% | 10 | 30 |
69
+ | 6 | 26.67% | 8 | 30 |
70
+ | 7 | 36.67% | 11 | 30 |
71
+ | 8 | 36.67% | 11 | 30 |
72
+ | 9 | 53.33% | 16 | 30 |
73
+ | 10 | 30.00% | 9 | 30 |
74
 
75
  ### GPQADiamond
76
 
77
+ - **Average Accuracy**: 47.47% ± 1.04%
78
  - **Number of Runs**: 3
79
 
80
  | Run | Accuracy | Questions Solved | Total Questions |
81
  |-----|----------|-----------------|----------------|
82
+ | 1 | 48.99% | 97 | 198 |
83
+ | 2 | 48.48% | 96 | 198 |
84
+ | 3 | 44.95% | 89 | 198 |
85
 
86
  ### MMLUPro
87
 
88
+ - **Average Accuracy**: 31.60% ± 0.00%
89
  - **Number of Runs**: 1
90
 
91
  | Run | Accuracy | Questions Solved | Total Questions |
92
  |-----|----------|-----------------|----------------|
93
+ | 1 | 31.60% | 158 | 500 |
94
 
95
  ### LiveCodeBench
96
 
97
+ - **Average Accuracy**: 55.58% ± 0.81%
98
  - **Number of Runs**: 3
99
 
100
  | Run | Accuracy | Questions Solved | Total Questions |
101
  |-----|----------|-----------------|----------------|
102
+ | 1 | 57.14% | 292 | 511 |
103
+ | 2 | 54.40% | 278 | 511 |
104
+ | 3 | 55.19% | 282 | 511 |
105
 
106
  ### CodeElo
107
 
108
+ - **Average Accuracy**: 24.21% ± 0.96%
109
  - **Number of Runs**: 3
110
 
111
  | Run | Accuracy | Questions Solved | Total Questions |
112
  |-----|----------|-----------------|----------------|
113
+ | 1 | 22.51% | 88 | 391 |
114
+ | 2 | 24.30% | 95 | 391 |
115
+ | 3 | 25.83% | 101 | 391 |
116
 
117
  ### JEEBench
118
 
119
+ - **Average Accuracy**: 66.13% ± 0.40%
120
  - **Number of Runs**: 3
121
 
122
  | Run | Accuracy | Questions Solved | Total Questions |
123
  |-----|----------|-----------------|----------------|
124
+ | 1 | 65.24% | 336.0 | 515 |
125
+ | 2 | 66.21% | 341.0 | 515 |
126
+ | 3 | 66.94% | 344.75 | 515 |
127