Upload eval_results/meta-llama/Meta-Llama-3-70B-Instruct/main/ifeval/results_2024-05-25T11-39-50.035526.json with huggingface_hub
Browse files
eval_results/meta-llama/Meta-Llama-3-70B-Instruct/main/ifeval/results_2024-05-25T11-39-50.035526.json
ADDED
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": 1518975.682519105,
|
9 |
+
"end_time": 1524330.669084378,
|
10 |
+
"total_evaluation_time_secondes": "5354.986565273022",
|
11 |
+
"model_name": "meta-llama/Meta-Llama-3-70B-Instruct",
|
12 |
+
"model_sha": "359ec69a0f92259a3cd2da3bb01d31e16c260cfc",
|
13 |
+
"model_dtype": "torch.bfloat16",
|
14 |
+
"model_size": "131.73 GB",
|
15 |
+
"config": null
|
16 |
+
},
|
17 |
+
"results": {
|
18 |
+
"extended|ifeval|0": {
|
19 |
+
"prompt_level_strict_acc": 0.7800369685767098,
|
20 |
+
"prompt_level_strict_acc_stderr": 0.017825247192217113,
|
21 |
+
"inst_level_strict_acc": 0.8453237410071942,
|
22 |
+
"inst_level_strict_acc_stderr": 0.00039885503173309563,
|
23 |
+
"prompt_level_loose_acc": 0.8391866913123844,
|
24 |
+
"prompt_level_loose_acc_stderr": 0.015808599888607083,
|
25 |
+
"inst_level_loose_acc": 0.8920863309352518,
|
26 |
+
"inst_level_loose_acc_stderr": 0.0003385840294307306
|
27 |
+
},
|
28 |
+
"all": {
|
29 |
+
"prompt_level_strict_acc": 0.7800369685767098,
|
30 |
+
"prompt_level_strict_acc_stderr": 0.017825247192217113,
|
31 |
+
"inst_level_strict_acc": 0.8453237410071942,
|
32 |
+
"inst_level_strict_acc_stderr": 0.00039885503173309563,
|
33 |
+
"prompt_level_loose_acc": 0.8391866913123844,
|
34 |
+
"prompt_level_loose_acc_stderr": 0.015808599888607083,
|
35 |
+
"inst_level_loose_acc": 0.8920863309352518,
|
36 |
+
"inst_level_loose_acc_stderr": 0.0003385840294307306
|
37 |
+
}
|
38 |
+
},
|
39 |
+
"versions": {
|
40 |
+
"extended|ifeval|0": 0
|
41 |
+
},
|
42 |
+
"config_tasks": {
|
43 |
+
"extended|ifeval": {
|
44 |
+
"name": "ifeval",
|
45 |
+
"prompt_function": "ifeval_prompt",
|
46 |
+
"hf_repo": "wis-k/instruction-following-eval",
|
47 |
+
"hf_subset": "default",
|
48 |
+
"metric": [
|
49 |
+
"ifeval_metric"
|
50 |
+
],
|
51 |
+
"hf_avail_splits": [
|
52 |
+
"train"
|
53 |
+
],
|
54 |
+
"evaluation_splits": [
|
55 |
+
"train"
|
56 |
+
],
|
57 |
+
"few_shots_split": "train",
|
58 |
+
"few_shots_select": "random_sampling",
|
59 |
+
"generation_size": 1280,
|
60 |
+
"stop_sequence": [],
|
61 |
+
"output_regex": null,
|
62 |
+
"num_samples": null,
|
63 |
+
"frozen": false,
|
64 |
+
"suite": [
|
65 |
+
"extended"
|
66 |
+
],
|
67 |
+
"original_num_docs": 541,
|
68 |
+
"effective_num_docs": 541,
|
69 |
+
"trust_dataset": null,
|
70 |
+
"must_remove_duplicate_docs": null,
|
71 |
+
"version": 0
|
72 |
+
}
|
73 |
+
},
|
74 |
+
"summary_tasks": {
|
75 |
+
"extended|ifeval|0": {
|
76 |
+
"hashes": {
|
77 |
+
"hash_examples": "e99cbf567588d7c6",
|
78 |
+
"hash_full_prompts": "30fe3a1c6acd1de0",
|
79 |
+
"hash_input_tokens": "8897069e6973d945",
|
80 |
+
"hash_cont_tokens": "b5e56978f507fa25"
|
81 |
+
},
|
82 |
+
"truncated": 541,
|
83 |
+
"non_truncated": 0,
|
84 |
+
"padded": 116,
|
85 |
+
"non_padded": 425,
|
86 |
+
"effective_few_shots": 0.0,
|
87 |
+
"num_truncated_few_shots": 0
|
88 |
+
}
|
89 |
+
},
|
90 |
+
"summary_general": {
|
91 |
+
"hashes": {
|
92 |
+
"hash_examples": "ea046ab2c6fc5928",
|
93 |
+
"hash_full_prompts": "bcd0aa2470bb7894",
|
94 |
+
"hash_input_tokens": "b8bbae42101c2e4a",
|
95 |
+
"hash_cont_tokens": "4afe7ea46274cdc5"
|
96 |
+
},
|
97 |
+
"truncated": 541,
|
98 |
+
"non_truncated": 0,
|
99 |
+
"padded": 116,
|
100 |
+
"non_padded": 425,
|
101 |
+
"num_truncated_few_shots": 0
|
102 |
+
}
|
103 |
+
}
|