Upload eval_results/HuggingFaceTB/SmolLM2-1.7B-Instruct/main/gsm8k/results_2025-02-12T14-56-55.504908.json with huggingface_hub
Browse files
eval_results/HuggingFaceTB/SmolLM2-1.7B-Instruct/main/gsm8k/results_2025-02-12T14-56-55.504908.json
ADDED
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"config_general": {
|
3 |
+
"lighteval_sha": "?",
|
4 |
+
"num_fewshot_seeds": 1,
|
5 |
+
"override_batch_size": -1,
|
6 |
+
"max_samples": null,
|
7 |
+
"job_id": 0,
|
8 |
+
"start_time": 662980.141519316,
|
9 |
+
"end_time": 663183.167246989,
|
10 |
+
"total_evaluation_time_secondes": "203.02572767296806",
|
11 |
+
"model_name": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
|
12 |
+
"model_sha": "",
|
13 |
+
"model_dtype": null,
|
14 |
+
"model_size": null
|
15 |
+
},
|
16 |
+
"results": {
|
17 |
+
"lighteval|gsm8k|0": {
|
18 |
+
"maj@8": 0.04700530705079606,
|
19 |
+
"maj@8_stderr": 0.005829898355937189,
|
20 |
+
"qem": 0.09097801364670205,
|
21 |
+
"qem_stderr": 0.007921322844013654
|
22 |
+
},
|
23 |
+
"all": {
|
24 |
+
"maj@8": 0.04700530705079606,
|
25 |
+
"maj@8_stderr": 0.005829898355937189,
|
26 |
+
"qem": 0.09097801364670205,
|
27 |
+
"qem_stderr": 0.007921322844013654
|
28 |
+
}
|
29 |
+
},
|
30 |
+
"versions": {
|
31 |
+
"lighteval|gsm8k|0": 0
|
32 |
+
},
|
33 |
+
"config_tasks": {
|
34 |
+
"lighteval|gsm8k": {
|
35 |
+
"name": "gsm8k",
|
36 |
+
"prompt_function": "gsm8k",
|
37 |
+
"hf_repo": "gsm8k",
|
38 |
+
"hf_subset": "main",
|
39 |
+
"metric": [
|
40 |
+
{
|
41 |
+
"metric_name": "qem",
|
42 |
+
"higher_is_better": true,
|
43 |
+
"category": "3",
|
44 |
+
"use_case": "5",
|
45 |
+
"sample_level_fn": "compute",
|
46 |
+
"corpus_level_fn": "mean"
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"metric_name": "maj@8",
|
50 |
+
"higher_is_better": true,
|
51 |
+
"category": "5",
|
52 |
+
"use_case": "5",
|
53 |
+
"sample_level_fn": "compute",
|
54 |
+
"corpus_level_fn": "mean"
|
55 |
+
}
|
56 |
+
],
|
57 |
+
"hf_revision": null,
|
58 |
+
"hf_filter": null,
|
59 |
+
"hf_avail_splits": [
|
60 |
+
"train",
|
61 |
+
"test"
|
62 |
+
],
|
63 |
+
"trust_dataset": true,
|
64 |
+
"evaluation_splits": [
|
65 |
+
"test"
|
66 |
+
],
|
67 |
+
"few_shots_split": null,
|
68 |
+
"few_shots_select": "random_sampling_from_train",
|
69 |
+
"generation_size": 256,
|
70 |
+
"generation_grammar": null,
|
71 |
+
"stop_sequence": [
|
72 |
+
"Question="
|
73 |
+
],
|
74 |
+
"num_samples": null,
|
75 |
+
"suite": [
|
76 |
+
"lighteval"
|
77 |
+
],
|
78 |
+
"original_num_docs": 1319,
|
79 |
+
"effective_num_docs": 1319,
|
80 |
+
"must_remove_duplicate_docs": false,
|
81 |
+
"version": 0
|
82 |
+
}
|
83 |
+
},
|
84 |
+
"summary_tasks": {
|
85 |
+
"lighteval|gsm8k|0": {
|
86 |
+
"hashes": {
|
87 |
+
"hash_examples": "7957ba36acab8574",
|
88 |
+
"hash_full_prompts": "1dad1de90e245444",
|
89 |
+
"hash_input_tokens": "acac8977121c9ea9",
|
90 |
+
"hash_cont_tokens": "bf643cf9dc530426"
|
91 |
+
},
|
92 |
+
"truncated": 0,
|
93 |
+
"non_truncated": 2638,
|
94 |
+
"padded": 0,
|
95 |
+
"non_padded": 2638,
|
96 |
+
"effective_few_shots": 0.0,
|
97 |
+
"num_truncated_few_shots": 0
|
98 |
+
}
|
99 |
+
},
|
100 |
+
"summary_general": {
|
101 |
+
"hashes": {
|
102 |
+
"hash_examples": "e99e29b2d4ab55c1",
|
103 |
+
"hash_full_prompts": "04c6169702a9369e",
|
104 |
+
"hash_input_tokens": "39f93cc19f3f49c9",
|
105 |
+
"hash_cont_tokens": "1d791bf43ab6f002"
|
106 |
+
},
|
107 |
+
"truncated": 0,
|
108 |
+
"non_truncated": 2638,
|
109 |
+
"padded": 0,
|
110 |
+
"non_padded": 2638,
|
111 |
+
"num_truncated_few_shots": 0
|
112 |
+
}
|
113 |
+
}
|