Clean up false PoT
Browse files- eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_hard_pot/results_2024-05-27T12-05-20.563680.json +0 -91
- eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_hard_pot/results_2024-05-27T12-36-27.035745.json +0 -91
- eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_hard_pot/results_2024-05-27T13-13-06.061324.json +0 -193
- eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_medium_pot/results_2024-05-27T12-05-24.547168.json +0 -91
- eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_medium_pot/results_2024-05-27T12-35-59.406866.json +0 -91
- eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_medium_pot/results_2024-05-27T13-07-46.038921.json +0 -193
eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_hard_pot/results_2024-05-27T12-05-20.563680.json
DELETED
@@ -1,91 +0,0 @@
|
|
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": 3546137.978264133,
|
9 |
-
"end_time": 3546437.319502369,
|
10 |
-
"total_evaluation_time_secondes": "299.3412382360548",
|
11 |
-
"model_name": "deepseek-ai/deepseek-math-7b-rl",
|
12 |
-
"model_sha": "f3cd419a172ff5a100e9a563ebfe7900d78e9740",
|
13 |
-
"model_dtype": "torch.bfloat16",
|
14 |
-
"model_size": "12.93 GB",
|
15 |
-
"config": null
|
16 |
-
},
|
17 |
-
"results": {
|
18 |
-
"custom|aimo_kaggle_hard_pot:v0|0": {
|
19 |
-
"qem": 0.02,
|
20 |
-
"qem_stderr": 0.02
|
21 |
-
},
|
22 |
-
"all": {
|
23 |
-
"qem": 0.02,
|
24 |
-
"qem_stderr": 0.02
|
25 |
-
}
|
26 |
-
},
|
27 |
-
"versions": {
|
28 |
-
"custom|aimo_kaggle_hard_pot:v0|0": 0
|
29 |
-
},
|
30 |
-
"config_tasks": {
|
31 |
-
"custom|aimo_kaggle_hard_pot:v0": {
|
32 |
-
"name": "aimo_kaggle_hard_pot:v0",
|
33 |
-
"prompt_function": "kaggle_hard_pot_prompt_fn_v0",
|
34 |
-
"hf_repo": "AI-MO/kaggle-validation-set-hard",
|
35 |
-
"hf_subset": "v0",
|
36 |
-
"metric": [
|
37 |
-
"quasi_exact_match_code_and_math"
|
38 |
-
],
|
39 |
-
"hf_avail_splits": [
|
40 |
-
"train"
|
41 |
-
],
|
42 |
-
"evaluation_splits": [
|
43 |
-
"train"
|
44 |
-
],
|
45 |
-
"few_shots_split": null,
|
46 |
-
"few_shots_select": null,
|
47 |
-
"generation_size": 2048,
|
48 |
-
"stop_sequence": null,
|
49 |
-
"output_regex": null,
|
50 |
-
"num_samples": null,
|
51 |
-
"frozen": false,
|
52 |
-
"suite": [
|
53 |
-
"custom"
|
54 |
-
],
|
55 |
-
"original_num_docs": 50,
|
56 |
-
"effective_num_docs": 50,
|
57 |
-
"trust_dataset": null,
|
58 |
-
"must_remove_duplicate_docs": null,
|
59 |
-
"version": 0
|
60 |
-
}
|
61 |
-
},
|
62 |
-
"summary_tasks": {
|
63 |
-
"custom|aimo_kaggle_hard_pot:v0|0": {
|
64 |
-
"hashes": {
|
65 |
-
"hash_examples": "303213a38d9f7512",
|
66 |
-
"hash_full_prompts": "3c670fa0e80bc301",
|
67 |
-
"hash_input_tokens": "69dabd3aaebed332",
|
68 |
-
"hash_cont_tokens": "f1e1dbb860f6f173"
|
69 |
-
},
|
70 |
-
"truncated": 50,
|
71 |
-
"non_truncated": 0,
|
72 |
-
"padded": 35,
|
73 |
-
"non_padded": 15,
|
74 |
-
"effective_few_shots": 0.0,
|
75 |
-
"num_truncated_few_shots": 0
|
76 |
-
}
|
77 |
-
},
|
78 |
-
"summary_general": {
|
79 |
-
"hashes": {
|
80 |
-
"hash_examples": "cae48890a7c47904",
|
81 |
-
"hash_full_prompts": "6f120f8d8bb938e6",
|
82 |
-
"hash_input_tokens": "ff7094ff98f5f808",
|
83 |
-
"hash_cont_tokens": "b3810939778104d4"
|
84 |
-
},
|
85 |
-
"truncated": 50,
|
86 |
-
"non_truncated": 0,
|
87 |
-
"padded": 35,
|
88 |
-
"non_padded": 15,
|
89 |
-
"num_truncated_few_shots": 0
|
90 |
-
}
|
91 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_hard_pot/results_2024-05-27T12-36-27.035745.json
DELETED
@@ -1,91 +0,0 @@
|
|
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": 2511250.889234414,
|
9 |
-
"end_time": 2511567.044527658,
|
10 |
-
"total_evaluation_time_secondes": "316.1552932439372",
|
11 |
-
"model_name": "deepseek-ai/deepseek-math-7b-rl",
|
12 |
-
"model_sha": "f3cd419a172ff5a100e9a563ebfe7900d78e9740",
|
13 |
-
"model_dtype": "torch.bfloat16",
|
14 |
-
"model_size": "12.93 GB",
|
15 |
-
"config": null
|
16 |
-
},
|
17 |
-
"results": {
|
18 |
-
"custom|aimo_kaggle_hard_pot:v0|0": {
|
19 |
-
"qem": 0.02,
|
20 |
-
"qem_stderr": 0.02
|
21 |
-
},
|
22 |
-
"all": {
|
23 |
-
"qem": 0.02,
|
24 |
-
"qem_stderr": 0.02
|
25 |
-
}
|
26 |
-
},
|
27 |
-
"versions": {
|
28 |
-
"custom|aimo_kaggle_hard_pot:v0|0": 0
|
29 |
-
},
|
30 |
-
"config_tasks": {
|
31 |
-
"custom|aimo_kaggle_hard_pot:v0": {
|
32 |
-
"name": "aimo_kaggle_hard_pot:v0",
|
33 |
-
"prompt_function": "kaggle_hard_pot_prompt_fn_v0",
|
34 |
-
"hf_repo": "AI-MO/kaggle-validation-set-hard",
|
35 |
-
"hf_subset": "v0",
|
36 |
-
"metric": [
|
37 |
-
"quasi_exact_match_code_and_math"
|
38 |
-
],
|
39 |
-
"hf_avail_splits": [
|
40 |
-
"train"
|
41 |
-
],
|
42 |
-
"evaluation_splits": [
|
43 |
-
"train"
|
44 |
-
],
|
45 |
-
"few_shots_split": null,
|
46 |
-
"few_shots_select": null,
|
47 |
-
"generation_size": 2048,
|
48 |
-
"stop_sequence": null,
|
49 |
-
"output_regex": null,
|
50 |
-
"num_samples": null,
|
51 |
-
"frozen": false,
|
52 |
-
"suite": [
|
53 |
-
"custom"
|
54 |
-
],
|
55 |
-
"original_num_docs": 50,
|
56 |
-
"effective_num_docs": 50,
|
57 |
-
"trust_dataset": null,
|
58 |
-
"must_remove_duplicate_docs": null,
|
59 |
-
"version": 0
|
60 |
-
}
|
61 |
-
},
|
62 |
-
"summary_tasks": {
|
63 |
-
"custom|aimo_kaggle_hard_pot:v0|0": {
|
64 |
-
"hashes": {
|
65 |
-
"hash_examples": "303213a38d9f7512",
|
66 |
-
"hash_full_prompts": "3c670fa0e80bc301",
|
67 |
-
"hash_input_tokens": "69dabd3aaebed332",
|
68 |
-
"hash_cont_tokens": "f1e1dbb860f6f173"
|
69 |
-
},
|
70 |
-
"truncated": 50,
|
71 |
-
"non_truncated": 0,
|
72 |
-
"padded": 35,
|
73 |
-
"non_padded": 15,
|
74 |
-
"effective_few_shots": 0.0,
|
75 |
-
"num_truncated_few_shots": 0
|
76 |
-
}
|
77 |
-
},
|
78 |
-
"summary_general": {
|
79 |
-
"hashes": {
|
80 |
-
"hash_examples": "cae48890a7c47904",
|
81 |
-
"hash_full_prompts": "6f120f8d8bb938e6",
|
82 |
-
"hash_input_tokens": "ff7094ff98f5f808",
|
83 |
-
"hash_cont_tokens": "b3810939778104d4"
|
84 |
-
},
|
85 |
-
"truncated": 50,
|
86 |
-
"non_truncated": 0,
|
87 |
-
"padded": 35,
|
88 |
-
"non_padded": 15,
|
89 |
-
"num_truncated_few_shots": 0
|
90 |
-
}
|
91 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_hard_pot/results_2024-05-27T13-13-06.061324.json
DELETED
@@ -1,193 +0,0 @@
|
|
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": 2108140.565538177,
|
9 |
-
"end_time": 2108762.006272666,
|
10 |
-
"total_evaluation_time_secondes": "621.4407344893552",
|
11 |
-
"model_name": "deepseek-ai/deepseek-math-7b-rl",
|
12 |
-
"model_sha": "f3cd419a172ff5a100e9a563ebfe7900d78e9740",
|
13 |
-
"model_dtype": "torch.bfloat16",
|
14 |
-
"model_size": "12.93 GB",
|
15 |
-
"config": null
|
16 |
-
},
|
17 |
-
"results": {
|
18 |
-
"custom|aimo_kaggle_hard_pot:v0|0": {
|
19 |
-
"qem": 0.02,
|
20 |
-
"qem_stderr": 0.02
|
21 |
-
},
|
22 |
-
"custom|aimo_kaggle_hard_pot:v1|0": {
|
23 |
-
"qem": 0.06,
|
24 |
-
"qem_stderr": 0.033926691677251195
|
25 |
-
},
|
26 |
-
"custom|aimo_kaggle_hard_pot:v2|0": {
|
27 |
-
"qem": 0.38,
|
28 |
-
"qem_stderr": 0.06934092056863767
|
29 |
-
},
|
30 |
-
"custom|aimo_kaggle_hard_pot:_average|0": {
|
31 |
-
"qem": 0.15333333333333335,
|
32 |
-
"qem_stderr": 0.041089204081962954
|
33 |
-
},
|
34 |
-
"all": {
|
35 |
-
"qem": 0.15333333333333335,
|
36 |
-
"qem_stderr": 0.041089204081962954
|
37 |
-
}
|
38 |
-
},
|
39 |
-
"versions": {
|
40 |
-
"custom|aimo_kaggle_hard_pot:v0|0": 0,
|
41 |
-
"custom|aimo_kaggle_hard_pot:v1|0": 0,
|
42 |
-
"custom|aimo_kaggle_hard_pot:v2|0": 0
|
43 |
-
},
|
44 |
-
"config_tasks": {
|
45 |
-
"custom|aimo_kaggle_hard_pot:v0": {
|
46 |
-
"name": "aimo_kaggle_hard_pot:v0",
|
47 |
-
"prompt_function": "kaggle_hard_pot_prompt_fn_v0",
|
48 |
-
"hf_repo": "AI-MO/kaggle-validation-set-hard",
|
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": 50,
|
70 |
-
"effective_num_docs": 50,
|
71 |
-
"trust_dataset": null,
|
72 |
-
"must_remove_duplicate_docs": null,
|
73 |
-
"version": 0
|
74 |
-
},
|
75 |
-
"custom|aimo_kaggle_hard_pot:v1": {
|
76 |
-
"name": "aimo_kaggle_hard_pot:v1",
|
77 |
-
"prompt_function": "kaggle_hard_pot_prompt_fn_v1",
|
78 |
-
"hf_repo": "AI-MO/kaggle-validation-set-hard",
|
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": 50,
|
100 |
-
"effective_num_docs": 50,
|
101 |
-
"trust_dataset": null,
|
102 |
-
"must_remove_duplicate_docs": null,
|
103 |
-
"version": 0
|
104 |
-
},
|
105 |
-
"custom|aimo_kaggle_hard_pot:v2": {
|
106 |
-
"name": "aimo_kaggle_hard_pot:v2",
|
107 |
-
"prompt_function": "kaggle_hard_pot_prompt_fn_v2",
|
108 |
-
"hf_repo": "AI-MO/kaggle-validation-set-hard",
|
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": 50,
|
130 |
-
"effective_num_docs": 50,
|
131 |
-
"trust_dataset": null,
|
132 |
-
"must_remove_duplicate_docs": null,
|
133 |
-
"version": 0
|
134 |
-
}
|
135 |
-
},
|
136 |
-
"summary_tasks": {
|
137 |
-
"custom|aimo_kaggle_hard_pot:v0|0": {
|
138 |
-
"hashes": {
|
139 |
-
"hash_examples": "303213a38d9f7512",
|
140 |
-
"hash_full_prompts": "3c670fa0e80bc301",
|
141 |
-
"hash_input_tokens": "1bd4187ad6963415",
|
142 |
-
"hash_cont_tokens": "4a5131314ac71945"
|
143 |
-
},
|
144 |
-
"truncated": 50,
|
145 |
-
"non_truncated": 0,
|
146 |
-
"padded": 37,
|
147 |
-
"non_padded": 13,
|
148 |
-
"effective_few_shots": 0.0,
|
149 |
-
"num_truncated_few_shots": 0
|
150 |
-
},
|
151 |
-
"custom|aimo_kaggle_hard_pot:v1|0": {
|
152 |
-
"hashes": {
|
153 |
-
"hash_examples": "e4234b97ad92862f",
|
154 |
-
"hash_full_prompts": "09e7759a96f64e59",
|
155 |
-
"hash_input_tokens": "dc7f68cf14be3d61",
|
156 |
-
"hash_cont_tokens": "835e43e6085c79ec"
|
157 |
-
},
|
158 |
-
"truncated": 50,
|
159 |
-
"non_truncated": 0,
|
160 |
-
"padded": 32,
|
161 |
-
"non_padded": 18,
|
162 |
-
"effective_few_shots": 0.0,
|
163 |
-
"num_truncated_few_shots": 0
|
164 |
-
},
|
165 |
-
"custom|aimo_kaggle_hard_pot:v2|0": {
|
166 |
-
"hashes": {
|
167 |
-
"hash_examples": "6396eb8833e13ba0",
|
168 |
-
"hash_full_prompts": "65f8b95ea99c7087",
|
169 |
-
"hash_input_tokens": "bddb67459e7601c5",
|
170 |
-
"hash_cont_tokens": "16c7b73b8b022a8d"
|
171 |
-
},
|
172 |
-
"truncated": 50,
|
173 |
-
"non_truncated": 0,
|
174 |
-
"padded": 30,
|
175 |
-
"non_padded": 20,
|
176 |
-
"effective_few_shots": 0.0,
|
177 |
-
"num_truncated_few_shots": 0
|
178 |
-
}
|
179 |
-
},
|
180 |
-
"summary_general": {
|
181 |
-
"hashes": {
|
182 |
-
"hash_examples": "648c9a107d279e1e",
|
183 |
-
"hash_full_prompts": "7f4c38eb08b3bb41",
|
184 |
-
"hash_input_tokens": "16f0fd76e0a2a165",
|
185 |
-
"hash_cont_tokens": "c21eba382df462d7"
|
186 |
-
},
|
187 |
-
"truncated": 150,
|
188 |
-
"non_truncated": 0,
|
189 |
-
"padded": 99,
|
190 |
-
"non_padded": 51,
|
191 |
-
"num_truncated_few_shots": 0
|
192 |
-
}
|
193 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_medium_pot/results_2024-05-27T12-05-24.547168.json
DELETED
@@ -1,91 +0,0 @@
|
|
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": 2509401.314190552,
|
9 |
-
"end_time": 2509704.555886234,
|
10 |
-
"total_evaluation_time_secondes": "303.2416956820525",
|
11 |
-
"model_name": "deepseek-ai/deepseek-math-7b-rl",
|
12 |
-
"model_sha": "f3cd419a172ff5a100e9a563ebfe7900d78e9740",
|
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": 0.025,
|
20 |
-
"qem_stderr": 0.024999999999999998
|
21 |
-
},
|
22 |
-
"all": {
|
23 |
-
"qem": 0.025,
|
24 |
-
"qem_stderr": 0.024999999999999998
|
25 |
-
}
|
26 |
-
},
|
27 |
-
"versions": {
|
28 |
-
"custom|aimo_kaggle_medium_pot:v0|0": 0
|
29 |
-
},
|
30 |
-
"config_tasks": {
|
31 |
-
"custom|aimo_kaggle_medium_pot:v0": {
|
32 |
-
"name": "aimo_kaggle_medium_pot:v0",
|
33 |
-
"prompt_function": "kaggle_medium_pot_prompt_fn_v0",
|
34 |
-
"hf_repo": "AI-MO/kaggle-validation-set-medium",
|
35 |
-
"hf_subset": "v0",
|
36 |
-
"metric": [
|
37 |
-
"quasi_exact_match_code_and_math"
|
38 |
-
],
|
39 |
-
"hf_avail_splits": [
|
40 |
-
"train"
|
41 |
-
],
|
42 |
-
"evaluation_splits": [
|
43 |
-
"train"
|
44 |
-
],
|
45 |
-
"few_shots_split": null,
|
46 |
-
"few_shots_select": null,
|
47 |
-
"generation_size": 2048,
|
48 |
-
"stop_sequence": null,
|
49 |
-
"output_regex": null,
|
50 |
-
"num_samples": null,
|
51 |
-
"frozen": false,
|
52 |
-
"suite": [
|
53 |
-
"custom"
|
54 |
-
],
|
55 |
-
"original_num_docs": 40,
|
56 |
-
"effective_num_docs": 40,
|
57 |
-
"trust_dataset": null,
|
58 |
-
"must_remove_duplicate_docs": null,
|
59 |
-
"version": 0
|
60 |
-
}
|
61 |
-
},
|
62 |
-
"summary_tasks": {
|
63 |
-
"custom|aimo_kaggle_medium_pot:v0|0": {
|
64 |
-
"hashes": {
|
65 |
-
"hash_examples": "2799c24461029dc3",
|
66 |
-
"hash_full_prompts": "de6572b914d5bba6",
|
67 |
-
"hash_input_tokens": "bb5427e5227cc6b4",
|
68 |
-
"hash_cont_tokens": "38c9a1261865d264"
|
69 |
-
},
|
70 |
-
"truncated": 40,
|
71 |
-
"non_truncated": 0,
|
72 |
-
"padded": 31,
|
73 |
-
"non_padded": 9,
|
74 |
-
"effective_few_shots": 0.0,
|
75 |
-
"num_truncated_few_shots": 0
|
76 |
-
}
|
77 |
-
},
|
78 |
-
"summary_general": {
|
79 |
-
"hashes": {
|
80 |
-
"hash_examples": "c72202b0a18ef5f9",
|
81 |
-
"hash_full_prompts": "a0429b0dea3e33db",
|
82 |
-
"hash_input_tokens": "8e6ef9684e80b71f",
|
83 |
-
"hash_cont_tokens": "b1bd4ed4824c97cb"
|
84 |
-
},
|
85 |
-
"truncated": 40,
|
86 |
-
"non_truncated": 0,
|
87 |
-
"padded": 31,
|
88 |
-
"non_padded": 9,
|
89 |
-
"num_truncated_few_shots": 0
|
90 |
-
}
|
91 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_medium_pot/results_2024-05-27T12-35-59.406866.json
DELETED
@@ -1,91 +0,0 @@
|
|
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": 4051098.729740778,
|
9 |
-
"end_time": 4051401.342947718,
|
10 |
-
"total_evaluation_time_secondes": "302.61320694023743",
|
11 |
-
"model_name": "deepseek-ai/deepseek-math-7b-rl",
|
12 |
-
"model_sha": "f3cd419a172ff5a100e9a563ebfe7900d78e9740",
|
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": 0.025,
|
20 |
-
"qem_stderr": 0.024999999999999998
|
21 |
-
},
|
22 |
-
"all": {
|
23 |
-
"qem": 0.025,
|
24 |
-
"qem_stderr": 0.024999999999999998
|
25 |
-
}
|
26 |
-
},
|
27 |
-
"versions": {
|
28 |
-
"custom|aimo_kaggle_medium_pot:v0|0": 0
|
29 |
-
},
|
30 |
-
"config_tasks": {
|
31 |
-
"custom|aimo_kaggle_medium_pot:v0": {
|
32 |
-
"name": "aimo_kaggle_medium_pot:v0",
|
33 |
-
"prompt_function": "kaggle_medium_pot_prompt_fn_v0",
|
34 |
-
"hf_repo": "AI-MO/kaggle-validation-set-medium",
|
35 |
-
"hf_subset": "v0",
|
36 |
-
"metric": [
|
37 |
-
"quasi_exact_match_code_and_math"
|
38 |
-
],
|
39 |
-
"hf_avail_splits": [
|
40 |
-
"train"
|
41 |
-
],
|
42 |
-
"evaluation_splits": [
|
43 |
-
"train"
|
44 |
-
],
|
45 |
-
"few_shots_split": null,
|
46 |
-
"few_shots_select": null,
|
47 |
-
"generation_size": 2048,
|
48 |
-
"stop_sequence": null,
|
49 |
-
"output_regex": null,
|
50 |
-
"num_samples": null,
|
51 |
-
"frozen": false,
|
52 |
-
"suite": [
|
53 |
-
"custom"
|
54 |
-
],
|
55 |
-
"original_num_docs": 40,
|
56 |
-
"effective_num_docs": 40,
|
57 |
-
"trust_dataset": null,
|
58 |
-
"must_remove_duplicate_docs": null,
|
59 |
-
"version": 0
|
60 |
-
}
|
61 |
-
},
|
62 |
-
"summary_tasks": {
|
63 |
-
"custom|aimo_kaggle_medium_pot:v0|0": {
|
64 |
-
"hashes": {
|
65 |
-
"hash_examples": "2799c24461029dc3",
|
66 |
-
"hash_full_prompts": "de6572b914d5bba6",
|
67 |
-
"hash_input_tokens": "bb5427e5227cc6b4",
|
68 |
-
"hash_cont_tokens": "38c9a1261865d264"
|
69 |
-
},
|
70 |
-
"truncated": 40,
|
71 |
-
"non_truncated": 0,
|
72 |
-
"padded": 31,
|
73 |
-
"non_padded": 9,
|
74 |
-
"effective_few_shots": 0.0,
|
75 |
-
"num_truncated_few_shots": 0
|
76 |
-
}
|
77 |
-
},
|
78 |
-
"summary_general": {
|
79 |
-
"hashes": {
|
80 |
-
"hash_examples": "c72202b0a18ef5f9",
|
81 |
-
"hash_full_prompts": "a0429b0dea3e33db",
|
82 |
-
"hash_input_tokens": "8e6ef9684e80b71f",
|
83 |
-
"hash_cont_tokens": "b1bd4ed4824c97cb"
|
84 |
-
},
|
85 |
-
"truncated": 40,
|
86 |
-
"non_truncated": 0,
|
87 |
-
"padded": 31,
|
88 |
-
"non_padded": 9,
|
89 |
-
"num_truncated_few_shots": 0
|
90 |
-
}
|
91 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
eval_results/deepseek-ai/deepseek-math-7b-rl/main/aimo_kaggle_medium_pot/results_2024-05-27T13-07-46.038921.json
DELETED
@@ -1,193 +0,0 @@
|
|
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": 2111711.970974838,
|
9 |
-
"end_time": 2112038.728973985,
|
10 |
-
"total_evaluation_time_secondes": "326.7579991472885",
|
11 |
-
"model_name": "deepseek-ai/deepseek-math-7b-rl",
|
12 |
-
"model_sha": "f3cd419a172ff5a100e9a563ebfe7900d78e9740",
|
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": 0.025,
|
20 |
-
"qem_stderr": 0.024999999999999998
|
21 |
-
},
|
22 |
-
"custom|aimo_kaggle_medium_pot:v1|0": {
|
23 |
-
"qem": 0.0,
|
24 |
-
"qem_stderr": 0.0
|
25 |
-
},
|
26 |
-
"custom|aimo_kaggle_medium_pot:v2|0": {
|
27 |
-
"qem": 0.125,
|
28 |
-
"qem_stderr": 0.05295740910852021
|
29 |
-
},
|
30 |
-
"custom|aimo_kaggle_medium_pot:_average|0": {
|
31 |
-
"qem": 0.049999999999999996,
|
32 |
-
"qem_stderr": 0.025985803036173403
|
33 |
-
},
|
34 |
-
"all": {
|
35 |
-
"qem": 0.049999999999999996,
|
36 |
-
"qem_stderr": 0.025985803036173403
|
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": "72723b206760d825"
|
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": "06f12390b65df63f"
|
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": "b6edaeabcd3b05a5"
|
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": "9e01c1ecd569a7f7"
|
186 |
-
},
|
187 |
-
"truncated": 120,
|
188 |
-
"non_truncated": 0,
|
189 |
-
"padded": 86,
|
190 |
-
"non_padded": 34,
|
191 |
-
"num_truncated_few_shots": 0
|
192 |
-
}
|
193 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|