Upload eval_results/AI-MO/deepseek-math-7b-sft/aimo_v01.44/gsm8k/results_2024-04-26T00-11-28.464247.json with huggingface_hub
Browse files
eval_results/AI-MO/deepseek-math-7b-sft/aimo_v01.44/gsm8k/results_2024-04-26T00-11-28.464247.json
ADDED
|
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1183904.304096318,
|
| 9 |
+
"end_time": 1184321.98438609,
|
| 10 |
+
"total_evaluation_time_secondes": "417.6802897718735",
|
| 11 |
+
"model_name": "AI-MO/deepseek-math-7b-sft",
|
| 12 |
+
"model_sha": "4f53169b1dce7747a0a519866be3fc6dc3faf9d8",
|
| 13 |
+
"model_dtype": "torch.bfloat16",
|
| 14 |
+
"model_size": "12.93 GB",
|
| 15 |
+
"config": null
|
| 16 |
+
},
|
| 17 |
+
"results": {
|
| 18 |
+
"lighteval|gsm8k|5": {
|
| 19 |
+
"qem": 0.6504927975739196,
|
| 20 |
+
"qem_stderr": 0.013133836511705995
|
| 21 |
+
},
|
| 22 |
+
"all": {
|
| 23 |
+
"qem": 0.6504927975739196,
|
| 24 |
+
"qem_stderr": 0.013133836511705995
|
| 25 |
+
}
|
| 26 |
+
},
|
| 27 |
+
"versions": {
|
| 28 |
+
"lighteval|gsm8k|5": 0
|
| 29 |
+
},
|
| 30 |
+
"config_tasks": {
|
| 31 |
+
"lighteval|gsm8k": {
|
| 32 |
+
"name": "gsm8k",
|
| 33 |
+
"prompt_function": "gsm8k",
|
| 34 |
+
"hf_repo": "gsm8k",
|
| 35 |
+
"hf_subset": "main",
|
| 36 |
+
"metric": [
|
| 37 |
+
"quasi_exact_match_gsm8k"
|
| 38 |
+
],
|
| 39 |
+
"hf_avail_splits": [
|
| 40 |
+
"train",
|
| 41 |
+
"test"
|
| 42 |
+
],
|
| 43 |
+
"evaluation_splits": [
|
| 44 |
+
"test"
|
| 45 |
+
],
|
| 46 |
+
"few_shots_split": null,
|
| 47 |
+
"few_shots_select": "random_sampling_from_train",
|
| 48 |
+
"generation_size": 256,
|
| 49 |
+
"stop_sequence": [
|
| 50 |
+
"Question:"
|
| 51 |
+
],
|
| 52 |
+
"output_regex": null,
|
| 53 |
+
"frozen": false,
|
| 54 |
+
"suite": [
|
| 55 |
+
"lighteval"
|
| 56 |
+
],
|
| 57 |
+
"original_num_docs": 1319,
|
| 58 |
+
"effective_num_docs": 1319,
|
| 59 |
+
"trust_dataset": true,
|
| 60 |
+
"must_remove_duplicate_docs": null
|
| 61 |
+
}
|
| 62 |
+
},
|
| 63 |
+
"summary_tasks": {
|
| 64 |
+
"lighteval|gsm8k|5": {
|
| 65 |
+
"hashes": {
|
| 66 |
+
"hash_examples": "0ed016e24e7512fd",
|
| 67 |
+
"hash_full_prompts": "b6d97958f0ed1df8",
|
| 68 |
+
"hash_input_tokens": "3f58ee5e568409ec",
|
| 69 |
+
"hash_cont_tokens": "f4271bfbc8710629"
|
| 70 |
+
},
|
| 71 |
+
"truncated": 1319,
|
| 72 |
+
"non_truncated": 0,
|
| 73 |
+
"padded": 611,
|
| 74 |
+
"non_padded": 708,
|
| 75 |
+
"effective_few_shots": 5.0,
|
| 76 |
+
"num_truncated_few_shots": 0
|
| 77 |
+
}
|
| 78 |
+
},
|
| 79 |
+
"summary_general": {
|
| 80 |
+
"hashes": {
|
| 81 |
+
"hash_examples": "bc71463e88551d0e",
|
| 82 |
+
"hash_full_prompts": "b5a5dacaed0816ff",
|
| 83 |
+
"hash_input_tokens": "e4e0fd8eba82bd47",
|
| 84 |
+
"hash_cont_tokens": "7846e4638680745a"
|
| 85 |
+
},
|
| 86 |
+
"truncated": 1319,
|
| 87 |
+
"non_truncated": 0,
|
| 88 |
+
"padded": 611,
|
| 89 |
+
"non_padded": 708,
|
| 90 |
+
"num_truncated_few_shots": 0
|
| 91 |
+
}
|
| 92 |
+
}
|