lewtun HF staff commited on
Commit
b9fe313
·
verified ·
1 Parent(s): ef6f29e

Upload eval_results/AI-MO/deepseek-math-7b-rl-sft/aimo_v02.02/aimo_kaggle_medium_pot/results_2024-05-28T10-54-30.115012.json with huggingface_hub

Browse files
eval_results/AI-MO/deepseek-math-7b-rl-sft/aimo_v02.02/aimo_kaggle_medium_pot/results_2024-05-28T10-54-30.115012.json ADDED
@@ -0,0 +1,193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config_general": {
3
+ "lighteval_sha": "?",
4
+ "num_fewshot_seeds": 1,
5
+ "override_batch_size": 4,
6
+ "max_samples": null,
7
+ "job_id": "",
8
+ "start_time": 2424020.911480174,
9
+ "end_time": 2424419.009344627,
10
+ "total_evaluation_time_secondes": "398.09786445321515",
11
+ "model_name": "AI-MO/deepseek-math-7b-rl-sft",
12
+ "model_sha": "c579155dded92356530d24203be65dd06b745a34",
13
+ "model_dtype": "torch.bfloat16",
14
+ "model_size": "12.93 GB",
15
+ "config": null
16
+ },
17
+ "results": {
18
+ "custom|aimo_kaggle_medium_pot:v0|0": {
19
+ "qem": 0.15,
20
+ "qem_stderr": 0.05717718748968655
21
+ },
22
+ "custom|aimo_kaggle_medium_pot:v1|0": {
23
+ "qem": 0.15,
24
+ "qem_stderr": 0.05717718748968655
25
+ },
26
+ "custom|aimo_kaggle_medium_pot:v2|0": {
27
+ "qem": 0.15,
28
+ "qem_stderr": 0.05717718748968655
29
+ },
30
+ "custom|aimo_kaggle_medium_pot:_average|0": {
31
+ "qem": 0.15,
32
+ "qem_stderr": 0.05717718748968655
33
+ },
34
+ "all": {
35
+ "qem": 0.15,
36
+ "qem_stderr": 0.05717718748968655
37
+ }
38
+ },
39
+ "versions": {
40
+ "custom|aimo_kaggle_medium_pot:v0|0": 0,
41
+ "custom|aimo_kaggle_medium_pot:v1|0": 0,
42
+ "custom|aimo_kaggle_medium_pot:v2|0": 0
43
+ },
44
+ "config_tasks": {
45
+ "custom|aimo_kaggle_medium_pot:v0": {
46
+ "name": "aimo_kaggle_medium_pot:v0",
47
+ "prompt_function": "kaggle_medium_pot_prompt_fn_v0",
48
+ "hf_repo": "AI-MO/kaggle-validation-set-medium",
49
+ "hf_subset": "v0",
50
+ "metric": [
51
+ "quasi_exact_match_code_and_math"
52
+ ],
53
+ "hf_avail_splits": [
54
+ "train"
55
+ ],
56
+ "evaluation_splits": [
57
+ "train"
58
+ ],
59
+ "few_shots_split": null,
60
+ "few_shots_select": null,
61
+ "generation_size": 2048,
62
+ "stop_sequence": null,
63
+ "output_regex": null,
64
+ "num_samples": null,
65
+ "frozen": false,
66
+ "suite": [
67
+ "custom"
68
+ ],
69
+ "original_num_docs": 40,
70
+ "effective_num_docs": 40,
71
+ "trust_dataset": null,
72
+ "must_remove_duplicate_docs": null,
73
+ "version": 0
74
+ },
75
+ "custom|aimo_kaggle_medium_pot:v1": {
76
+ "name": "aimo_kaggle_medium_pot:v1",
77
+ "prompt_function": "kaggle_medium_pot_prompt_fn_v1",
78
+ "hf_repo": "AI-MO/kaggle-validation-set-medium",
79
+ "hf_subset": "v0",
80
+ "metric": [
81
+ "quasi_exact_match_code_and_math"
82
+ ],
83
+ "hf_avail_splits": [
84
+ "train"
85
+ ],
86
+ "evaluation_splits": [
87
+ "train"
88
+ ],
89
+ "few_shots_split": null,
90
+ "few_shots_select": null,
91
+ "generation_size": 2048,
92
+ "stop_sequence": null,
93
+ "output_regex": null,
94
+ "num_samples": null,
95
+ "frozen": false,
96
+ "suite": [
97
+ "custom"
98
+ ],
99
+ "original_num_docs": 40,
100
+ "effective_num_docs": 40,
101
+ "trust_dataset": null,
102
+ "must_remove_duplicate_docs": null,
103
+ "version": 0
104
+ },
105
+ "custom|aimo_kaggle_medium_pot:v2": {
106
+ "name": "aimo_kaggle_medium_pot:v2",
107
+ "prompt_function": "kaggle_medium_pot_prompt_fn_v2",
108
+ "hf_repo": "AI-MO/kaggle-validation-set-medium",
109
+ "hf_subset": "v0",
110
+ "metric": [
111
+ "quasi_exact_match_code_and_math"
112
+ ],
113
+ "hf_avail_splits": [
114
+ "train"
115
+ ],
116
+ "evaluation_splits": [
117
+ "train"
118
+ ],
119
+ "few_shots_split": null,
120
+ "few_shots_select": null,
121
+ "generation_size": 2048,
122
+ "stop_sequence": null,
123
+ "output_regex": null,
124
+ "num_samples": null,
125
+ "frozen": false,
126
+ "suite": [
127
+ "custom"
128
+ ],
129
+ "original_num_docs": 40,
130
+ "effective_num_docs": 40,
131
+ "trust_dataset": null,
132
+ "must_remove_duplicate_docs": null,
133
+ "version": 0
134
+ }
135
+ },
136
+ "summary_tasks": {
137
+ "custom|aimo_kaggle_medium_pot:v0|0": {
138
+ "hashes": {
139
+ "hash_examples": "2799c24461029dc3",
140
+ "hash_full_prompts": "de6572b914d5bba6",
141
+ "hash_input_tokens": "5af82ee284ccce29",
142
+ "hash_cont_tokens": "9ed765ebd58d3013"
143
+ },
144
+ "truncated": 40,
145
+ "non_truncated": 0,
146
+ "padded": 29,
147
+ "non_padded": 11,
148
+ "effective_few_shots": 0.0,
149
+ "num_truncated_few_shots": 0
150
+ },
151
+ "custom|aimo_kaggle_medium_pot:v1|0": {
152
+ "hashes": {
153
+ "hash_examples": "806b2e2056b41f84",
154
+ "hash_full_prompts": "427d5de6e4d90df2",
155
+ "hash_input_tokens": "be3dcf9f8a2350d0",
156
+ "hash_cont_tokens": "2f085c109bc10f7d"
157
+ },
158
+ "truncated": 40,
159
+ "non_truncated": 0,
160
+ "padded": 26,
161
+ "non_padded": 14,
162
+ "effective_few_shots": 0.0,
163
+ "num_truncated_few_shots": 0
164
+ },
165
+ "custom|aimo_kaggle_medium_pot:v2|0": {
166
+ "hashes": {
167
+ "hash_examples": "d8534375acc5d427",
168
+ "hash_full_prompts": "6bb90c129d6d6123",
169
+ "hash_input_tokens": "22eeaedbeb3f56f0",
170
+ "hash_cont_tokens": "1aedc4a3ff7a75a1"
171
+ },
172
+ "truncated": 40,
173
+ "non_truncated": 0,
174
+ "padded": 31,
175
+ "non_padded": 9,
176
+ "effective_few_shots": 0.0,
177
+ "num_truncated_few_shots": 0
178
+ }
179
+ },
180
+ "summary_general": {
181
+ "hashes": {
182
+ "hash_examples": "623505a45a4910c2",
183
+ "hash_full_prompts": "3aca0fad61a42921",
184
+ "hash_input_tokens": "c7d2c8bb76f8ecb1",
185
+ "hash_cont_tokens": "70d3921740f7ab02"
186
+ },
187
+ "truncated": 120,
188
+ "non_truncated": 0,
189
+ "padded": 86,
190
+ "non_padded": 34,
191
+ "num_truncated_few_shots": 0
192
+ }
193
+ }