Commit ·
61c152a
1
Parent(s): 9e39fcc
example every eval results for lmeval and lighteval
Browse files
every_eval_ever_results/aime25.json
ADDED
|
@@ -0,0 +1,129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.2.2",
|
| 3 |
+
"evaluation_id": "aime25/inference-optimization/MiniMax-M2.5.w4a16/1777302637.214952",
|
| 4 |
+
"evaluation_timestamp": "5872197",
|
| 5 |
+
"retrieved_timestamp": "1777302637.214952",
|
| 6 |
+
"source_metadata": {
|
| 7 |
+
"source_name": "lighteval",
|
| 8 |
+
"source_type": "evaluation_run",
|
| 9 |
+
"source_organization_name": "unknown",
|
| 10 |
+
"evaluator_relationship": "third_party"
|
| 11 |
+
},
|
| 12 |
+
"eval_library": {
|
| 13 |
+
"name": "lighteval",
|
| 14 |
+
"version": "unknown"
|
| 15 |
+
},
|
| 16 |
+
"model_info": {
|
| 17 |
+
"name": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 18 |
+
"id": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 19 |
+
"developer": "inference-optimization",
|
| 20 |
+
"inference_engine": {
|
| 21 |
+
"name": "vllm"
|
| 22 |
+
},
|
| 23 |
+
"additional_details": {
|
| 24 |
+
"provider": "hosted_vllm",
|
| 25 |
+
"base_url": "http://0.0.0.0:8000/v1",
|
| 26 |
+
"concurrent_requests": "8",
|
| 27 |
+
"verbose": "False",
|
| 28 |
+
"api_max_retry": "8",
|
| 29 |
+
"api_retry_sleep": "1.0",
|
| 30 |
+
"api_retry_multiplier": "2.0",
|
| 31 |
+
"timeout": "2400.0",
|
| 32 |
+
"num_seeds_merged": "8"
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"evaluation_results": [
|
| 36 |
+
{
|
| 37 |
+
"evaluation_name": "aime25",
|
| 38 |
+
"source_data": {
|
| 39 |
+
"dataset_name": "aime25",
|
| 40 |
+
"source_type": "hf_dataset",
|
| 41 |
+
"hf_repo": "yentinglin/aime_2025",
|
| 42 |
+
"hf_split": "train"
|
| 43 |
+
},
|
| 44 |
+
"evaluation_timestamp": "5877463",
|
| 45 |
+
"metric_config": {
|
| 46 |
+
"evaluation_description": "pass@k:k=1&n=1",
|
| 47 |
+
"lower_is_better": false,
|
| 48 |
+
"score_type": "continuous",
|
| 49 |
+
"min_score": 0.0,
|
| 50 |
+
"max_score": 1.0
|
| 51 |
+
},
|
| 52 |
+
"score_details": {
|
| 53 |
+
"score": 0.8416666666666667,
|
| 54 |
+
"details": {
|
| 55 |
+
"seed_scores": "[0.9, 0.8, 0.8333333333333334, 0.8333333333333334, 0.8333333333333334, 0.8666666666666667, 0.8333333333333334, 0.8333333333333334]",
|
| 56 |
+
"seed_values": "[1234, 1356, 3344, 4158, 42, 5322, 5678, 9843]"
|
| 57 |
+
},
|
| 58 |
+
"uncertainty": {
|
| 59 |
+
"standard_error": {
|
| 60 |
+
"value": 0.01044638617546681,
|
| 61 |
+
"method": "across_seeds"
|
| 62 |
+
},
|
| 63 |
+
"num_samples": 8
|
| 64 |
+
}
|
| 65 |
+
},
|
| 66 |
+
"generation_config": {
|
| 67 |
+
"generation_args": {
|
| 68 |
+
"temperature": 1.0,
|
| 69 |
+
"top_p": 0.95,
|
| 70 |
+
"top_k": 40.0,
|
| 71 |
+
"max_tokens": 64000,
|
| 72 |
+
"max_attempts": 1
|
| 73 |
+
},
|
| 74 |
+
"additional_details": {
|
| 75 |
+
"repetition_penalty": "1.0",
|
| 76 |
+
"presence_penalty": "1.5",
|
| 77 |
+
"seed": "1234",
|
| 78 |
+
"min_p": "0.0"
|
| 79 |
+
}
|
| 80 |
+
}
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"evaluation_name": "aime25",
|
| 84 |
+
"source_data": {
|
| 85 |
+
"dataset_name": "aime25",
|
| 86 |
+
"source_type": "hf_dataset",
|
| 87 |
+
"hf_repo": "yentinglin/aime_2025",
|
| 88 |
+
"hf_split": "train"
|
| 89 |
+
},
|
| 90 |
+
"evaluation_timestamp": "5877463",
|
| 91 |
+
"metric_config": {
|
| 92 |
+
"evaluation_description": "avg@n:n=1",
|
| 93 |
+
"lower_is_better": false,
|
| 94 |
+
"score_type": "continuous",
|
| 95 |
+
"min_score": 0.0,
|
| 96 |
+
"max_score": 1.0
|
| 97 |
+
},
|
| 98 |
+
"score_details": {
|
| 99 |
+
"score": 0.8416666666666667,
|
| 100 |
+
"details": {
|
| 101 |
+
"seed_scores": "[0.9, 0.8, 0.8333333333333334, 0.8333333333333334, 0.8333333333333334, 0.8666666666666667, 0.8333333333333334, 0.8333333333333334]",
|
| 102 |
+
"seed_values": "[1234, 1356, 3344, 4158, 42, 5322, 5678, 9843]"
|
| 103 |
+
},
|
| 104 |
+
"uncertainty": {
|
| 105 |
+
"standard_error": {
|
| 106 |
+
"value": 0.01044638617546681,
|
| 107 |
+
"method": "across_seeds"
|
| 108 |
+
},
|
| 109 |
+
"num_samples": 8
|
| 110 |
+
}
|
| 111 |
+
},
|
| 112 |
+
"generation_config": {
|
| 113 |
+
"generation_args": {
|
| 114 |
+
"temperature": 1.0,
|
| 115 |
+
"top_p": 0.95,
|
| 116 |
+
"top_k": 40.0,
|
| 117 |
+
"max_tokens": 64000,
|
| 118 |
+
"max_attempts": 1
|
| 119 |
+
},
|
| 120 |
+
"additional_details": {
|
| 121 |
+
"repetition_penalty": "1.0",
|
| 122 |
+
"presence_penalty": "1.5",
|
| 123 |
+
"seed": "1234",
|
| 124 |
+
"min_p": "0.0"
|
| 125 |
+
}
|
| 126 |
+
}
|
| 127 |
+
}
|
| 128 |
+
]
|
| 129 |
+
}
|
every_eval_ever_results/gpqa_diamond.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.2.2",
|
| 3 |
+
"evaluation_id": "gpqa:diamond/inference-optimization/MiniMax-M2.5.w4a16/1777302658.000854",
|
| 4 |
+
"evaluation_timestamp": "5965114",
|
| 5 |
+
"retrieved_timestamp": "1777302658.000854",
|
| 6 |
+
"source_metadata": {
|
| 7 |
+
"source_name": "lighteval",
|
| 8 |
+
"source_type": "evaluation_run",
|
| 9 |
+
"source_organization_name": "unknown",
|
| 10 |
+
"evaluator_relationship": "third_party"
|
| 11 |
+
},
|
| 12 |
+
"eval_library": {
|
| 13 |
+
"name": "lighteval",
|
| 14 |
+
"version": "unknown"
|
| 15 |
+
},
|
| 16 |
+
"model_info": {
|
| 17 |
+
"name": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 18 |
+
"id": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 19 |
+
"developer": "inference-optimization",
|
| 20 |
+
"inference_engine": {
|
| 21 |
+
"name": "vllm"
|
| 22 |
+
},
|
| 23 |
+
"additional_details": {
|
| 24 |
+
"provider": "hosted_vllm",
|
| 25 |
+
"base_url": "http://0.0.0.0:8000/v1",
|
| 26 |
+
"concurrent_requests": "8",
|
| 27 |
+
"verbose": "False",
|
| 28 |
+
"api_max_retry": "8",
|
| 29 |
+
"api_retry_sleep": "1.0",
|
| 30 |
+
"api_retry_multiplier": "2.0",
|
| 31 |
+
"timeout": "2400.0",
|
| 32 |
+
"num_seeds_merged": "3"
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"evaluation_results": [
|
| 36 |
+
{
|
| 37 |
+
"evaluation_name": "gpqa:diamond",
|
| 38 |
+
"source_data": {
|
| 39 |
+
"dataset_name": "gpqa:diamond",
|
| 40 |
+
"source_type": "hf_dataset",
|
| 41 |
+
"hf_repo": "Idavidrein/gpqa",
|
| 42 |
+
"hf_split": "train"
|
| 43 |
+
},
|
| 44 |
+
"evaluation_timestamp": "5981739",
|
| 45 |
+
"metric_config": {
|
| 46 |
+
"evaluation_description": "gpqa_pass@k:k=1",
|
| 47 |
+
"lower_is_better": false,
|
| 48 |
+
"score_type": "continuous",
|
| 49 |
+
"min_score": 0.0,
|
| 50 |
+
"max_score": 1.0
|
| 51 |
+
},
|
| 52 |
+
"score_details": {
|
| 53 |
+
"score": 0.845117845117845,
|
| 54 |
+
"details": {
|
| 55 |
+
"seed_scores": "[0.8535353535353535, 0.8737373737373737, 0.8080808080808081]",
|
| 56 |
+
"seed_values": "[1234, 4158, 42]"
|
| 57 |
+
},
|
| 58 |
+
"uncertainty": {
|
| 59 |
+
"standard_error": {
|
| 60 |
+
"value": 0.01941508854321682,
|
| 61 |
+
"method": "across_seeds"
|
| 62 |
+
},
|
| 63 |
+
"num_samples": 3
|
| 64 |
+
}
|
| 65 |
+
},
|
| 66 |
+
"generation_config": {
|
| 67 |
+
"generation_args": {
|
| 68 |
+
"temperature": 1.0,
|
| 69 |
+
"top_p": 0.95,
|
| 70 |
+
"top_k": 40.0,
|
| 71 |
+
"max_tokens": 64000,
|
| 72 |
+
"max_attempts": 1
|
| 73 |
+
},
|
| 74 |
+
"additional_details": {
|
| 75 |
+
"repetition_penalty": "1.0",
|
| 76 |
+
"presence_penalty": "1.5",
|
| 77 |
+
"seed": "1234",
|
| 78 |
+
"min_p": "0.0"
|
| 79 |
+
}
|
| 80 |
+
}
|
| 81 |
+
}
|
| 82 |
+
]
|
| 83 |
+
}
|
every_eval_ever_results/gsm8k_platinum_cot_llama.json
ADDED
|
@@ -0,0 +1,126 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.2.2",
|
| 3 |
+
"evaluation_id": "gsm8k_platinum_cot_llama/inference-optimization/MiniMax-M2.5.w4a16/1777302614.762325",
|
| 4 |
+
"evaluation_timestamp": "1776885795",
|
| 5 |
+
"retrieved_timestamp": "1777302614.762325",
|
| 6 |
+
"source_metadata": {
|
| 7 |
+
"source_name": "lm-evaluation-harness",
|
| 8 |
+
"source_type": "evaluation_run",
|
| 9 |
+
"source_organization_name": "unknown",
|
| 10 |
+
"evaluator_relationship": "third_party"
|
| 11 |
+
},
|
| 12 |
+
"eval_library": {
|
| 13 |
+
"name": "lm_eval",
|
| 14 |
+
"version": "0.4.12.dev0"
|
| 15 |
+
},
|
| 16 |
+
"model_info": {
|
| 17 |
+
"name": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 18 |
+
"id": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 19 |
+
"developer": "inference-optimization",
|
| 20 |
+
"additional_details": {
|
| 21 |
+
"model_args": "{'model': 'inference-optimization/MiniMax-M2.5.w4a16', 'max_length': 196608, 'base_url': 'http://0.0.0.0:8000/v1/chat/completions', 'num_concurrent': 128, 'max_retries': 3, 'tokenized_requests': False, 'tokenizer_backend': None, 'timeout': 2400}",
|
| 22 |
+
"seed": "1234",
|
| 23 |
+
"num_seeds_merged": "3"
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"evaluation_results": [
|
| 27 |
+
{
|
| 28 |
+
"evaluation_name": "gsm8k_platinum_cot_llama/strict-match",
|
| 29 |
+
"source_data": {
|
| 30 |
+
"dataset_name": "gsm8k_platinum_cot_llama",
|
| 31 |
+
"source_type": "hf_dataset",
|
| 32 |
+
"hf_repo": "madrylab/gsm8k-platinum",
|
| 33 |
+
"hf_split": "test"
|
| 34 |
+
},
|
| 35 |
+
"evaluation_timestamp": "1776887868",
|
| 36 |
+
"metric_config": {
|
| 37 |
+
"evaluation_description": "exact_match (filter: strict-match)",
|
| 38 |
+
"lower_is_better": false,
|
| 39 |
+
"score_type": "continuous",
|
| 40 |
+
"min_score": 0.0,
|
| 41 |
+
"max_score": 1.0
|
| 42 |
+
},
|
| 43 |
+
"score_details": {
|
| 44 |
+
"score": 0.9636062861869313,
|
| 45 |
+
"details": {
|
| 46 |
+
"seed_scores": "[0.9602977667493796, 0.9669148056244831, 0.9636062861869313]",
|
| 47 |
+
"seed_values": "[1234, 4158, 42]"
|
| 48 |
+
},
|
| 49 |
+
"uncertainty": {
|
| 50 |
+
"standard_error": {
|
| 51 |
+
"value": 0.001910174587889599,
|
| 52 |
+
"method": "across_seeds"
|
| 53 |
+
},
|
| 54 |
+
"num_samples": 3
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"generation_config": {
|
| 58 |
+
"generation_args": {
|
| 59 |
+
"temperature": 1.0,
|
| 60 |
+
"top_p": 0.95,
|
| 61 |
+
"top_k": 40.0,
|
| 62 |
+
"max_tokens": 64000,
|
| 63 |
+
"max_attempts": 1
|
| 64 |
+
},
|
| 65 |
+
"additional_details": {
|
| 66 |
+
"do_sample": "true",
|
| 67 |
+
"until": "[\"<|eot_id|>\", \"<|start_header_id|>user<|end_header_id|>\", \"Q:\", \"</s>\", \"<|im_end|>\"]",
|
| 68 |
+
"min_p": "0.0",
|
| 69 |
+
"presence_penalty": "1.5",
|
| 70 |
+
"repetition_penalty": "1.0",
|
| 71 |
+
"seed": "1234",
|
| 72 |
+
"num_fewshot": "0"
|
| 73 |
+
}
|
| 74 |
+
}
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"evaluation_name": "gsm8k_platinum_cot_llama/flexible-extract",
|
| 78 |
+
"source_data": {
|
| 79 |
+
"dataset_name": "gsm8k_platinum_cot_llama",
|
| 80 |
+
"source_type": "hf_dataset",
|
| 81 |
+
"hf_repo": "madrylab/gsm8k-platinum",
|
| 82 |
+
"hf_split": "test"
|
| 83 |
+
},
|
| 84 |
+
"evaluation_timestamp": "1776887868",
|
| 85 |
+
"metric_config": {
|
| 86 |
+
"evaluation_description": "exact_match (filter: flexible-extract)",
|
| 87 |
+
"lower_is_better": false,
|
| 88 |
+
"score_type": "continuous",
|
| 89 |
+
"min_score": 0.0,
|
| 90 |
+
"max_score": 1.0
|
| 91 |
+
},
|
| 92 |
+
"score_details": {
|
| 93 |
+
"score": 0.9751861042183623,
|
| 94 |
+
"details": {
|
| 95 |
+
"seed_scores": "[0.9735318444995864, 0.9735318444995864, 0.978494623655914]",
|
| 96 |
+
"seed_values": "[1234, 4158, 42]"
|
| 97 |
+
},
|
| 98 |
+
"uncertainty": {
|
| 99 |
+
"standard_error": {
|
| 100 |
+
"value": 0.0016542597187758634,
|
| 101 |
+
"method": "across_seeds"
|
| 102 |
+
},
|
| 103 |
+
"num_samples": 3
|
| 104 |
+
}
|
| 105 |
+
},
|
| 106 |
+
"generation_config": {
|
| 107 |
+
"generation_args": {
|
| 108 |
+
"temperature": 1.0,
|
| 109 |
+
"top_p": 0.95,
|
| 110 |
+
"top_k": 40.0,
|
| 111 |
+
"max_tokens": 64000,
|
| 112 |
+
"max_attempts": 1
|
| 113 |
+
},
|
| 114 |
+
"additional_details": {
|
| 115 |
+
"do_sample": "true",
|
| 116 |
+
"until": "[\"<|eot_id|>\", \"<|start_header_id|>user<|end_header_id|>\", \"Q:\", \"</s>\", \"<|im_end|>\"]",
|
| 117 |
+
"min_p": "0.0",
|
| 118 |
+
"presence_penalty": "1.5",
|
| 119 |
+
"repetition_penalty": "1.0",
|
| 120 |
+
"seed": "1234",
|
| 121 |
+
"num_fewshot": "0"
|
| 122 |
+
}
|
| 123 |
+
}
|
| 124 |
+
}
|
| 125 |
+
]
|
| 126 |
+
}
|
every_eval_ever_results/ifeval.json
ADDED
|
@@ -0,0 +1,224 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.2.2",
|
| 3 |
+
"evaluation_id": "ifeval/inference-optimization/MiniMax-M2.5.w4a16/1777302591.528493",
|
| 4 |
+
"evaluation_timestamp": "1776886012",
|
| 5 |
+
"retrieved_timestamp": "1777302591.528493",
|
| 6 |
+
"source_metadata": {
|
| 7 |
+
"source_name": "lm-evaluation-harness",
|
| 8 |
+
"source_type": "evaluation_run",
|
| 9 |
+
"source_organization_name": "unknown",
|
| 10 |
+
"evaluator_relationship": "third_party"
|
| 11 |
+
},
|
| 12 |
+
"eval_library": {
|
| 13 |
+
"name": "lm_eval",
|
| 14 |
+
"version": "0.4.12.dev0"
|
| 15 |
+
},
|
| 16 |
+
"model_info": {
|
| 17 |
+
"name": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 18 |
+
"id": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 19 |
+
"developer": "inference-optimization",
|
| 20 |
+
"additional_details": {
|
| 21 |
+
"model_args": "{'model': 'inference-optimization/MiniMax-M2.5.w4a16', 'max_length': 196608, 'base_url': 'http://0.0.0.0:8000/v1/chat/completions', 'num_concurrent': 128, 'max_retries': 3, 'tokenized_requests': False, 'tokenizer_backend': None, 'timeout': 2400}",
|
| 22 |
+
"seed": "1234",
|
| 23 |
+
"num_seeds_merged": "3"
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"evaluation_results": [
|
| 27 |
+
{
|
| 28 |
+
"evaluation_name": "ifeval",
|
| 29 |
+
"source_data": {
|
| 30 |
+
"dataset_name": "ifeval",
|
| 31 |
+
"source_type": "hf_dataset",
|
| 32 |
+
"hf_repo": "google/IFEval",
|
| 33 |
+
"hf_split": "train"
|
| 34 |
+
},
|
| 35 |
+
"evaluation_timestamp": "1776888058",
|
| 36 |
+
"metric_config": {
|
| 37 |
+
"evaluation_description": "prompt_level_strict_acc",
|
| 38 |
+
"lower_is_better": false,
|
| 39 |
+
"score_type": "continuous",
|
| 40 |
+
"min_score": 0.0,
|
| 41 |
+
"max_score": 1.0
|
| 42 |
+
},
|
| 43 |
+
"score_details": {
|
| 44 |
+
"score": 0.8558225508317929,
|
| 45 |
+
"details": {
|
| 46 |
+
"seed_scores": "[0.8595194085027726, 0.8576709796672828, 0.8502772643253235]",
|
| 47 |
+
"seed_values": "[1234, 4158, 42]"
|
| 48 |
+
},
|
| 49 |
+
"uncertainty": {
|
| 50 |
+
"standard_error": {
|
| 51 |
+
"value": 0.0028235216851237458,
|
| 52 |
+
"method": "across_seeds"
|
| 53 |
+
},
|
| 54 |
+
"num_samples": 3
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"generation_config": {
|
| 58 |
+
"generation_args": {
|
| 59 |
+
"temperature": 1.0,
|
| 60 |
+
"top_p": 0.95,
|
| 61 |
+
"top_k": 40.0,
|
| 62 |
+
"max_tokens": 64000,
|
| 63 |
+
"max_attempts": 1
|
| 64 |
+
},
|
| 65 |
+
"additional_details": {
|
| 66 |
+
"until": "[]",
|
| 67 |
+
"do_sample": "true",
|
| 68 |
+
"min_p": "0.0",
|
| 69 |
+
"presence_penalty": "1.5",
|
| 70 |
+
"repetition_penalty": "1.0",
|
| 71 |
+
"seed": "1234",
|
| 72 |
+
"num_fewshot": "0"
|
| 73 |
+
}
|
| 74 |
+
}
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"evaluation_name": "ifeval",
|
| 78 |
+
"source_data": {
|
| 79 |
+
"dataset_name": "ifeval",
|
| 80 |
+
"source_type": "hf_dataset",
|
| 81 |
+
"hf_repo": "google/IFEval",
|
| 82 |
+
"hf_split": "train"
|
| 83 |
+
},
|
| 84 |
+
"evaluation_timestamp": "1776888058",
|
| 85 |
+
"metric_config": {
|
| 86 |
+
"evaluation_description": "inst_level_strict_acc",
|
| 87 |
+
"lower_is_better": false,
|
| 88 |
+
"score_type": "continuous",
|
| 89 |
+
"min_score": 0.0,
|
| 90 |
+
"max_score": 1.0
|
| 91 |
+
},
|
| 92 |
+
"score_details": {
|
| 93 |
+
"score": 0.904476418864908,
|
| 94 |
+
"details": {
|
| 95 |
+
"seed_scores": "[0.9088729016786571, 0.9040767386091128, 0.9004796163069544]",
|
| 96 |
+
"seed_values": "[1234, 4158, 42]"
|
| 97 |
+
},
|
| 98 |
+
"uncertainty": {
|
| 99 |
+
"standard_error": {
|
| 100 |
+
"value": 0.0024311600840520617,
|
| 101 |
+
"method": "across_seeds"
|
| 102 |
+
},
|
| 103 |
+
"num_samples": 3
|
| 104 |
+
}
|
| 105 |
+
},
|
| 106 |
+
"generation_config": {
|
| 107 |
+
"generation_args": {
|
| 108 |
+
"temperature": 1.0,
|
| 109 |
+
"top_p": 0.95,
|
| 110 |
+
"top_k": 40.0,
|
| 111 |
+
"max_tokens": 64000,
|
| 112 |
+
"max_attempts": 1
|
| 113 |
+
},
|
| 114 |
+
"additional_details": {
|
| 115 |
+
"until": "[]",
|
| 116 |
+
"do_sample": "true",
|
| 117 |
+
"min_p": "0.0",
|
| 118 |
+
"presence_penalty": "1.5",
|
| 119 |
+
"repetition_penalty": "1.0",
|
| 120 |
+
"seed": "1234",
|
| 121 |
+
"num_fewshot": "0"
|
| 122 |
+
}
|
| 123 |
+
}
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"evaluation_name": "ifeval",
|
| 127 |
+
"source_data": {
|
| 128 |
+
"dataset_name": "ifeval",
|
| 129 |
+
"source_type": "hf_dataset",
|
| 130 |
+
"hf_repo": "google/IFEval",
|
| 131 |
+
"hf_split": "train"
|
| 132 |
+
},
|
| 133 |
+
"evaluation_timestamp": "1776888058",
|
| 134 |
+
"metric_config": {
|
| 135 |
+
"evaluation_description": "prompt_level_loose_acc",
|
| 136 |
+
"lower_is_better": false,
|
| 137 |
+
"score_type": "continuous",
|
| 138 |
+
"min_score": 0.0,
|
| 139 |
+
"max_score": 1.0
|
| 140 |
+
},
|
| 141 |
+
"score_details": {
|
| 142 |
+
"score": 0.88909426987061,
|
| 143 |
+
"details": {
|
| 144 |
+
"seed_scores": "[0.8909426987060998, 0.88909426987061, 0.8872458410351202]",
|
| 145 |
+
"seed_values": "[1234, 4158, 42]"
|
| 146 |
+
},
|
| 147 |
+
"uncertainty": {
|
| 148 |
+
"standard_error": {
|
| 149 |
+
"value": 0.0010671908857479034,
|
| 150 |
+
"method": "across_seeds"
|
| 151 |
+
},
|
| 152 |
+
"num_samples": 3
|
| 153 |
+
}
|
| 154 |
+
},
|
| 155 |
+
"generation_config": {
|
| 156 |
+
"generation_args": {
|
| 157 |
+
"temperature": 1.0,
|
| 158 |
+
"top_p": 0.95,
|
| 159 |
+
"top_k": 40.0,
|
| 160 |
+
"max_tokens": 64000,
|
| 161 |
+
"max_attempts": 1
|
| 162 |
+
},
|
| 163 |
+
"additional_details": {
|
| 164 |
+
"until": "[]",
|
| 165 |
+
"do_sample": "true",
|
| 166 |
+
"min_p": "0.0",
|
| 167 |
+
"presence_penalty": "1.5",
|
| 168 |
+
"repetition_penalty": "1.0",
|
| 169 |
+
"seed": "1234",
|
| 170 |
+
"num_fewshot": "0"
|
| 171 |
+
}
|
| 172 |
+
}
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"evaluation_name": "ifeval",
|
| 176 |
+
"source_data": {
|
| 177 |
+
"dataset_name": "ifeval",
|
| 178 |
+
"source_type": "hf_dataset",
|
| 179 |
+
"hf_repo": "google/IFEval",
|
| 180 |
+
"hf_split": "train"
|
| 181 |
+
},
|
| 182 |
+
"evaluation_timestamp": "1776888058",
|
| 183 |
+
"metric_config": {
|
| 184 |
+
"evaluation_description": "inst_level_loose_acc",
|
| 185 |
+
"lower_is_better": false,
|
| 186 |
+
"score_type": "continuous",
|
| 187 |
+
"min_score": 0.0,
|
| 188 |
+
"max_score": 1.0
|
| 189 |
+
},
|
| 190 |
+
"score_details": {
|
| 191 |
+
"score": 0.9260591526778578,
|
| 192 |
+
"details": {
|
| 193 |
+
"seed_scores": "[0.9292565947242206, 0.9244604316546763, 0.9244604316546763]",
|
| 194 |
+
"seed_values": "[1234, 4158, 42]"
|
| 195 |
+
},
|
| 196 |
+
"uncertainty": {
|
| 197 |
+
"standard_error": {
|
| 198 |
+
"value": 0.001598721023181445,
|
| 199 |
+
"method": "across_seeds"
|
| 200 |
+
},
|
| 201 |
+
"num_samples": 3
|
| 202 |
+
}
|
| 203 |
+
},
|
| 204 |
+
"generation_config": {
|
| 205 |
+
"generation_args": {
|
| 206 |
+
"temperature": 1.0,
|
| 207 |
+
"top_p": 0.95,
|
| 208 |
+
"top_k": 40.0,
|
| 209 |
+
"max_tokens": 64000,
|
| 210 |
+
"max_attempts": 1
|
| 211 |
+
},
|
| 212 |
+
"additional_details": {
|
| 213 |
+
"until": "[]",
|
| 214 |
+
"do_sample": "true",
|
| 215 |
+
"min_p": "0.0",
|
| 216 |
+
"presence_penalty": "1.5",
|
| 217 |
+
"repetition_penalty": "1.0",
|
| 218 |
+
"seed": "1234",
|
| 219 |
+
"num_fewshot": "0"
|
| 220 |
+
}
|
| 221 |
+
}
|
| 222 |
+
}
|
| 223 |
+
]
|
| 224 |
+
}
|
every_eval_ever_results/math_500.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.2.2",
|
| 3 |
+
"evaluation_id": "math_500/inference-optimization/MiniMax-M2.5.w4a16/1777302681.44404",
|
| 4 |
+
"evaluation_timestamp": "5922700",
|
| 5 |
+
"retrieved_timestamp": "1777302681.44404",
|
| 6 |
+
"source_metadata": {
|
| 7 |
+
"source_name": "lighteval",
|
| 8 |
+
"source_type": "evaluation_run",
|
| 9 |
+
"source_organization_name": "unknown",
|
| 10 |
+
"evaluator_relationship": "third_party"
|
| 11 |
+
},
|
| 12 |
+
"eval_library": {
|
| 13 |
+
"name": "lighteval",
|
| 14 |
+
"version": "unknown"
|
| 15 |
+
},
|
| 16 |
+
"model_info": {
|
| 17 |
+
"name": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 18 |
+
"id": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 19 |
+
"developer": "inference-optimization",
|
| 20 |
+
"inference_engine": {
|
| 21 |
+
"name": "vllm"
|
| 22 |
+
},
|
| 23 |
+
"additional_details": {
|
| 24 |
+
"provider": "hosted_vllm",
|
| 25 |
+
"base_url": "http://0.0.0.0:8000/v1",
|
| 26 |
+
"concurrent_requests": "8",
|
| 27 |
+
"verbose": "False",
|
| 28 |
+
"api_max_retry": "8",
|
| 29 |
+
"api_retry_sleep": "1.0",
|
| 30 |
+
"api_retry_multiplier": "2.0",
|
| 31 |
+
"timeout": "2400.0",
|
| 32 |
+
"num_seeds_merged": "3"
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"evaluation_results": [
|
| 36 |
+
{
|
| 37 |
+
"evaluation_name": "math_500",
|
| 38 |
+
"source_data": {
|
| 39 |
+
"dataset_name": "math_500",
|
| 40 |
+
"source_type": "hf_dataset",
|
| 41 |
+
"hf_repo": "HuggingFaceH4/MATH-500",
|
| 42 |
+
"hf_split": "test"
|
| 43 |
+
},
|
| 44 |
+
"evaluation_timestamp": "5936819",
|
| 45 |
+
"metric_config": {
|
| 46 |
+
"evaluation_description": "pass@k:k=1&n=1",
|
| 47 |
+
"lower_is_better": false,
|
| 48 |
+
"score_type": "continuous",
|
| 49 |
+
"min_score": 0.0,
|
| 50 |
+
"max_score": 1.0
|
| 51 |
+
},
|
| 52 |
+
"score_details": {
|
| 53 |
+
"score": 0.876,
|
| 54 |
+
"details": {
|
| 55 |
+
"seed_scores": "[0.878, 0.87, 0.88]",
|
| 56 |
+
"seed_values": "[1234, 4158, 42]"
|
| 57 |
+
},
|
| 58 |
+
"uncertainty": {
|
| 59 |
+
"standard_error": {
|
| 60 |
+
"value": 0.0030550504633038962,
|
| 61 |
+
"method": "across_seeds"
|
| 62 |
+
},
|
| 63 |
+
"num_samples": 3
|
| 64 |
+
}
|
| 65 |
+
},
|
| 66 |
+
"generation_config": {
|
| 67 |
+
"generation_args": {
|
| 68 |
+
"temperature": 1.0,
|
| 69 |
+
"top_p": 0.95,
|
| 70 |
+
"top_k": 40.0,
|
| 71 |
+
"max_tokens": 64000,
|
| 72 |
+
"max_attempts": 1
|
| 73 |
+
},
|
| 74 |
+
"additional_details": {
|
| 75 |
+
"repetition_penalty": "1.0",
|
| 76 |
+
"presence_penalty": "1.5",
|
| 77 |
+
"seed": "1234",
|
| 78 |
+
"min_p": "0.0"
|
| 79 |
+
}
|
| 80 |
+
}
|
| 81 |
+
}
|
| 82 |
+
]
|
| 83 |
+
}
|
every_eval_ever_results/mmlu_pro_chat.json
ADDED
|
@@ -0,0 +1,743 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.2.2",
|
| 3 |
+
"evaluation_id": "mmlu_pro_chat/inference-optimization/MiniMax-M2.5.w4a16/1777302998.11639",
|
| 4 |
+
"evaluation_timestamp": "1777037460",
|
| 5 |
+
"retrieved_timestamp": "1777302998.11639",
|
| 6 |
+
"source_metadata": {
|
| 7 |
+
"source_name": "lm-evaluation-harness",
|
| 8 |
+
"source_type": "evaluation_run",
|
| 9 |
+
"source_organization_name": "unknown",
|
| 10 |
+
"evaluator_relationship": "third_party"
|
| 11 |
+
},
|
| 12 |
+
"eval_library": {
|
| 13 |
+
"name": "lm_eval",
|
| 14 |
+
"version": "0.4.12.dev0"
|
| 15 |
+
},
|
| 16 |
+
"model_info": {
|
| 17 |
+
"name": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 18 |
+
"id": "inference-optimization/MiniMax-M2.5.w4a16",
|
| 19 |
+
"developer": "inference-optimization",
|
| 20 |
+
"additional_details": {
|
| 21 |
+
"model_args": "{'model': 'inference-optimization/MiniMax-M2.5.w4a16', 'max_length': 196608, 'base_url': 'http://0.0.0.0:8000/v1/chat/completions', 'num_concurrent': 28, 'max_retries': 3, 'tokenized_requests': False, 'tokenizer_backend': None, 'timeout': 2400}",
|
| 22 |
+
"seed": "1234",
|
| 23 |
+
"num_seeds_merged": "3"
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"evaluation_results": [
|
| 27 |
+
{
|
| 28 |
+
"evaluation_name": "mmlu_pro_chat_biology/custom-extract",
|
| 29 |
+
"source_data": {
|
| 30 |
+
"dataset_name": "mmlu_pro_chat",
|
| 31 |
+
"source_type": "hf_dataset",
|
| 32 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 33 |
+
"hf_split": "test"
|
| 34 |
+
},
|
| 35 |
+
"evaluation_timestamp": "1777088443",
|
| 36 |
+
"metric_config": {
|
| 37 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 38 |
+
"lower_is_better": false,
|
| 39 |
+
"score_type": "continuous",
|
| 40 |
+
"min_score": 0.0,
|
| 41 |
+
"max_score": 1.0
|
| 42 |
+
},
|
| 43 |
+
"score_details": {
|
| 44 |
+
"score": 0.9079497907949791,
|
| 45 |
+
"details": {
|
| 46 |
+
"seed_scores": "[0.9093444909344491, 0.9065550906555091, 0.9079497907949791]",
|
| 47 |
+
"seed_values": "[1234, 4158, 42]"
|
| 48 |
+
},
|
| 49 |
+
"uncertainty": {
|
| 50 |
+
"standard_error": {
|
| 51 |
+
"value": 0.0008052305009618233,
|
| 52 |
+
"method": "across_seeds"
|
| 53 |
+
},
|
| 54 |
+
"num_samples": 3
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"generation_config": {
|
| 58 |
+
"generation_args": {
|
| 59 |
+
"temperature": 1.0,
|
| 60 |
+
"top_p": 0.95,
|
| 61 |
+
"top_k": 40.0,
|
| 62 |
+
"max_tokens": 64000,
|
| 63 |
+
"max_attempts": 1
|
| 64 |
+
},
|
| 65 |
+
"additional_details": {
|
| 66 |
+
"until": "[]",
|
| 67 |
+
"do_sample": "true",
|
| 68 |
+
"min_p": "0.0",
|
| 69 |
+
"presence_penalty": "1.5",
|
| 70 |
+
"repetition_penalty": "1.0",
|
| 71 |
+
"seed": "1234",
|
| 72 |
+
"num_fewshot": "0"
|
| 73 |
+
}
|
| 74 |
+
}
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"evaluation_name": "mmlu_pro_chat_business/custom-extract",
|
| 78 |
+
"source_data": {
|
| 79 |
+
"dataset_name": "mmlu_pro_chat",
|
| 80 |
+
"source_type": "hf_dataset",
|
| 81 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 82 |
+
"hf_split": "test"
|
| 83 |
+
},
|
| 84 |
+
"evaluation_timestamp": "1777088443",
|
| 85 |
+
"metric_config": {
|
| 86 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 87 |
+
"lower_is_better": false,
|
| 88 |
+
"score_type": "continuous",
|
| 89 |
+
"min_score": 0.0,
|
| 90 |
+
"max_score": 1.0
|
| 91 |
+
},
|
| 92 |
+
"score_details": {
|
| 93 |
+
"score": 0.8719898605830165,
|
| 94 |
+
"details": {
|
| 95 |
+
"seed_scores": "[0.8694550063371356, 0.8719898605830165, 0.8745247148288974]",
|
| 96 |
+
"seed_values": "[1234, 4158, 42]"
|
| 97 |
+
},
|
| 98 |
+
"uncertainty": {
|
| 99 |
+
"standard_error": {
|
| 100 |
+
"value": 0.0014634987812158046,
|
| 101 |
+
"method": "across_seeds"
|
| 102 |
+
},
|
| 103 |
+
"num_samples": 3
|
| 104 |
+
}
|
| 105 |
+
},
|
| 106 |
+
"generation_config": {
|
| 107 |
+
"generation_args": {
|
| 108 |
+
"temperature": 1.0,
|
| 109 |
+
"top_p": 0.95,
|
| 110 |
+
"top_k": 40.0,
|
| 111 |
+
"max_tokens": 64000,
|
| 112 |
+
"max_attempts": 1
|
| 113 |
+
},
|
| 114 |
+
"additional_details": {
|
| 115 |
+
"until": "[]",
|
| 116 |
+
"do_sample": "true",
|
| 117 |
+
"min_p": "0.0",
|
| 118 |
+
"presence_penalty": "1.5",
|
| 119 |
+
"repetition_penalty": "1.0",
|
| 120 |
+
"seed": "1234",
|
| 121 |
+
"num_fewshot": "0"
|
| 122 |
+
}
|
| 123 |
+
}
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"evaluation_name": "mmlu_pro_chat_chemistry/custom-extract",
|
| 127 |
+
"source_data": {
|
| 128 |
+
"dataset_name": "mmlu_pro_chat",
|
| 129 |
+
"source_type": "hf_dataset",
|
| 130 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 131 |
+
"hf_split": "test"
|
| 132 |
+
},
|
| 133 |
+
"evaluation_timestamp": "1777088443",
|
| 134 |
+
"metric_config": {
|
| 135 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 136 |
+
"lower_is_better": false,
|
| 137 |
+
"score_type": "continuous",
|
| 138 |
+
"min_score": 0.0,
|
| 139 |
+
"max_score": 1.0
|
| 140 |
+
},
|
| 141 |
+
"score_details": {
|
| 142 |
+
"score": 0.8695524146054181,
|
| 143 |
+
"details": {
|
| 144 |
+
"seed_scores": "[0.8692579505300353, 0.8719081272084805, 0.8674911660777385]",
|
| 145 |
+
"seed_values": "[1234, 4158, 42]"
|
| 146 |
+
},
|
| 147 |
+
"uncertainty": {
|
| 148 |
+
"standard_error": {
|
| 149 |
+
"value": 0.0012835391470967918,
|
| 150 |
+
"method": "across_seeds"
|
| 151 |
+
},
|
| 152 |
+
"num_samples": 3
|
| 153 |
+
}
|
| 154 |
+
},
|
| 155 |
+
"generation_config": {
|
| 156 |
+
"generation_args": {
|
| 157 |
+
"temperature": 1.0,
|
| 158 |
+
"top_p": 0.95,
|
| 159 |
+
"top_k": 40.0,
|
| 160 |
+
"max_tokens": 64000,
|
| 161 |
+
"max_attempts": 1
|
| 162 |
+
},
|
| 163 |
+
"additional_details": {
|
| 164 |
+
"until": "[]",
|
| 165 |
+
"do_sample": "true",
|
| 166 |
+
"min_p": "0.0",
|
| 167 |
+
"presence_penalty": "1.5",
|
| 168 |
+
"repetition_penalty": "1.0",
|
| 169 |
+
"seed": "1234",
|
| 170 |
+
"num_fewshot": "0"
|
| 171 |
+
}
|
| 172 |
+
}
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"evaluation_name": "mmlu_pro_chat_computer_science/custom-extract",
|
| 176 |
+
"source_data": {
|
| 177 |
+
"dataset_name": "mmlu_pro_chat",
|
| 178 |
+
"source_type": "hf_dataset",
|
| 179 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 180 |
+
"hf_split": "test"
|
| 181 |
+
},
|
| 182 |
+
"evaluation_timestamp": "1777088443",
|
| 183 |
+
"metric_config": {
|
| 184 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 185 |
+
"lower_is_better": false,
|
| 186 |
+
"score_type": "continuous",
|
| 187 |
+
"min_score": 0.0,
|
| 188 |
+
"max_score": 1.0
|
| 189 |
+
},
|
| 190 |
+
"score_details": {
|
| 191 |
+
"score": 0.8601626016260162,
|
| 192 |
+
"details": {
|
| 193 |
+
"seed_scores": "[0.8658536585365854, 0.8609756097560975, 0.8536585365853658]",
|
| 194 |
+
"seed_values": "[1234, 4158, 42]"
|
| 195 |
+
},
|
| 196 |
+
"uncertainty": {
|
| 197 |
+
"standard_error": {
|
| 198 |
+
"value": 0.003543820279301364,
|
| 199 |
+
"method": "across_seeds"
|
| 200 |
+
},
|
| 201 |
+
"num_samples": 3
|
| 202 |
+
}
|
| 203 |
+
},
|
| 204 |
+
"generation_config": {
|
| 205 |
+
"generation_args": {
|
| 206 |
+
"temperature": 1.0,
|
| 207 |
+
"top_p": 0.95,
|
| 208 |
+
"top_k": 40.0,
|
| 209 |
+
"max_tokens": 64000,
|
| 210 |
+
"max_attempts": 1
|
| 211 |
+
},
|
| 212 |
+
"additional_details": {
|
| 213 |
+
"until": "[]",
|
| 214 |
+
"do_sample": "true",
|
| 215 |
+
"min_p": "0.0",
|
| 216 |
+
"presence_penalty": "1.5",
|
| 217 |
+
"repetition_penalty": "1.0",
|
| 218 |
+
"seed": "1234",
|
| 219 |
+
"num_fewshot": "0"
|
| 220 |
+
}
|
| 221 |
+
}
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"evaluation_name": "mmlu_pro_chat_economics/custom-extract",
|
| 225 |
+
"source_data": {
|
| 226 |
+
"dataset_name": "mmlu_pro_chat",
|
| 227 |
+
"source_type": "hf_dataset",
|
| 228 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 229 |
+
"hf_split": "test"
|
| 230 |
+
},
|
| 231 |
+
"evaluation_timestamp": "1777088443",
|
| 232 |
+
"metric_config": {
|
| 233 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 234 |
+
"lower_is_better": false,
|
| 235 |
+
"score_type": "continuous",
|
| 236 |
+
"min_score": 0.0,
|
| 237 |
+
"max_score": 1.0
|
| 238 |
+
},
|
| 239 |
+
"score_details": {
|
| 240 |
+
"score": 0.8499210110584517,
|
| 241 |
+
"details": {
|
| 242 |
+
"seed_scores": "[0.8542654028436019, 0.8483412322274881, 0.8471563981042654]",
|
| 243 |
+
"seed_values": "[1234, 4158, 42]"
|
| 244 |
+
},
|
| 245 |
+
"uncertainty": {
|
| 246 |
+
"standard_error": {
|
| 247 |
+
"value": 0.0021989590690481933,
|
| 248 |
+
"method": "across_seeds"
|
| 249 |
+
},
|
| 250 |
+
"num_samples": 3
|
| 251 |
+
}
|
| 252 |
+
},
|
| 253 |
+
"generation_config": {
|
| 254 |
+
"generation_args": {
|
| 255 |
+
"temperature": 1.0,
|
| 256 |
+
"top_p": 0.95,
|
| 257 |
+
"top_k": 40.0,
|
| 258 |
+
"max_tokens": 64000,
|
| 259 |
+
"max_attempts": 1
|
| 260 |
+
},
|
| 261 |
+
"additional_details": {
|
| 262 |
+
"until": "[]",
|
| 263 |
+
"do_sample": "true",
|
| 264 |
+
"min_p": "0.0",
|
| 265 |
+
"presence_penalty": "1.5",
|
| 266 |
+
"repetition_penalty": "1.0",
|
| 267 |
+
"seed": "1234",
|
| 268 |
+
"num_fewshot": "0"
|
| 269 |
+
}
|
| 270 |
+
}
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"evaluation_name": "mmlu_pro_chat_engineering/custom-extract",
|
| 274 |
+
"source_data": {
|
| 275 |
+
"dataset_name": "mmlu_pro_chat",
|
| 276 |
+
"source_type": "hf_dataset",
|
| 277 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 278 |
+
"hf_split": "test"
|
| 279 |
+
},
|
| 280 |
+
"evaluation_timestamp": "1777088443",
|
| 281 |
+
"metric_config": {
|
| 282 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 283 |
+
"lower_is_better": false,
|
| 284 |
+
"score_type": "continuous",
|
| 285 |
+
"min_score": 0.0,
|
| 286 |
+
"max_score": 1.0
|
| 287 |
+
},
|
| 288 |
+
"score_details": {
|
| 289 |
+
"score": 0.7282421740626075,
|
| 290 |
+
"details": {
|
| 291 |
+
"seed_scores": "[0.7347781217750258, 0.7213622291021672, 0.7285861713106295]",
|
| 292 |
+
"seed_values": "[1234, 4158, 42]"
|
| 293 |
+
},
|
| 294 |
+
"uncertainty": {
|
| 295 |
+
"standard_error": {
|
| 296 |
+
"value": 0.003876652105120269,
|
| 297 |
+
"method": "across_seeds"
|
| 298 |
+
},
|
| 299 |
+
"num_samples": 3
|
| 300 |
+
}
|
| 301 |
+
},
|
| 302 |
+
"generation_config": {
|
| 303 |
+
"generation_args": {
|
| 304 |
+
"temperature": 1.0,
|
| 305 |
+
"top_p": 0.95,
|
| 306 |
+
"top_k": 40.0,
|
| 307 |
+
"max_tokens": 64000,
|
| 308 |
+
"max_attempts": 1
|
| 309 |
+
},
|
| 310 |
+
"additional_details": {
|
| 311 |
+
"until": "[]",
|
| 312 |
+
"do_sample": "true",
|
| 313 |
+
"min_p": "0.0",
|
| 314 |
+
"presence_penalty": "1.5",
|
| 315 |
+
"repetition_penalty": "1.0",
|
| 316 |
+
"seed": "1234",
|
| 317 |
+
"num_fewshot": "0"
|
| 318 |
+
}
|
| 319 |
+
}
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"evaluation_name": "mmlu_pro_chat_health/custom-extract",
|
| 323 |
+
"source_data": {
|
| 324 |
+
"dataset_name": "mmlu_pro_chat",
|
| 325 |
+
"source_type": "hf_dataset",
|
| 326 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 327 |
+
"hf_split": "test"
|
| 328 |
+
},
|
| 329 |
+
"evaluation_timestamp": "1777088443",
|
| 330 |
+
"metric_config": {
|
| 331 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 332 |
+
"lower_is_better": false,
|
| 333 |
+
"score_type": "continuous",
|
| 334 |
+
"min_score": 0.0,
|
| 335 |
+
"max_score": 1.0
|
| 336 |
+
},
|
| 337 |
+
"score_details": {
|
| 338 |
+
"score": 0.7921760391198044,
|
| 339 |
+
"details": {
|
| 340 |
+
"seed_scores": "[0.7946210268948656, 0.7970660146699267, 0.784841075794621]",
|
| 341 |
+
"seed_values": "[1234, 4158, 42]"
|
| 342 |
+
},
|
| 343 |
+
"uncertainty": {
|
| 344 |
+
"standard_error": {
|
| 345 |
+
"value": 0.0037347805174864275,
|
| 346 |
+
"method": "across_seeds"
|
| 347 |
+
},
|
| 348 |
+
"num_samples": 3
|
| 349 |
+
}
|
| 350 |
+
},
|
| 351 |
+
"generation_config": {
|
| 352 |
+
"generation_args": {
|
| 353 |
+
"temperature": 1.0,
|
| 354 |
+
"top_p": 0.95,
|
| 355 |
+
"top_k": 40.0,
|
| 356 |
+
"max_tokens": 64000,
|
| 357 |
+
"max_attempts": 1
|
| 358 |
+
},
|
| 359 |
+
"additional_details": {
|
| 360 |
+
"until": "[]",
|
| 361 |
+
"do_sample": "true",
|
| 362 |
+
"min_p": "0.0",
|
| 363 |
+
"presence_penalty": "1.5",
|
| 364 |
+
"repetition_penalty": "1.0",
|
| 365 |
+
"seed": "1234",
|
| 366 |
+
"num_fewshot": "0"
|
| 367 |
+
}
|
| 368 |
+
}
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"evaluation_name": "mmlu_pro_chat_history/custom-extract",
|
| 372 |
+
"source_data": {
|
| 373 |
+
"dataset_name": "mmlu_pro_chat",
|
| 374 |
+
"source_type": "hf_dataset",
|
| 375 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 376 |
+
"hf_split": "test"
|
| 377 |
+
},
|
| 378 |
+
"evaluation_timestamp": "1777088443",
|
| 379 |
+
"metric_config": {
|
| 380 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 381 |
+
"lower_is_better": false,
|
| 382 |
+
"score_type": "continuous",
|
| 383 |
+
"min_score": 0.0,
|
| 384 |
+
"max_score": 1.0
|
| 385 |
+
},
|
| 386 |
+
"score_details": {
|
| 387 |
+
"score": 0.6815398075240595,
|
| 388 |
+
"details": {
|
| 389 |
+
"seed_scores": "[0.6850393700787402, 0.6797900262467191, 0.6797900262467191]",
|
| 390 |
+
"seed_values": "[1234, 4158, 42]"
|
| 391 |
+
},
|
| 392 |
+
"uncertainty": {
|
| 393 |
+
"standard_error": {
|
| 394 |
+
"value": 0.0017497812773403416,
|
| 395 |
+
"method": "across_seeds"
|
| 396 |
+
},
|
| 397 |
+
"num_samples": 3
|
| 398 |
+
}
|
| 399 |
+
},
|
| 400 |
+
"generation_config": {
|
| 401 |
+
"generation_args": {
|
| 402 |
+
"temperature": 1.0,
|
| 403 |
+
"top_p": 0.95,
|
| 404 |
+
"top_k": 40.0,
|
| 405 |
+
"max_tokens": 64000,
|
| 406 |
+
"max_attempts": 1
|
| 407 |
+
},
|
| 408 |
+
"additional_details": {
|
| 409 |
+
"until": "[]",
|
| 410 |
+
"do_sample": "true",
|
| 411 |
+
"min_p": "0.0",
|
| 412 |
+
"presence_penalty": "1.5",
|
| 413 |
+
"repetition_penalty": "1.0",
|
| 414 |
+
"seed": "1234",
|
| 415 |
+
"num_fewshot": "0"
|
| 416 |
+
}
|
| 417 |
+
}
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"evaluation_name": "mmlu_pro_chat_law/custom-extract",
|
| 421 |
+
"source_data": {
|
| 422 |
+
"dataset_name": "mmlu_pro_chat",
|
| 423 |
+
"source_type": "hf_dataset",
|
| 424 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 425 |
+
"hf_split": "test"
|
| 426 |
+
},
|
| 427 |
+
"evaluation_timestamp": "1777088443",
|
| 428 |
+
"metric_config": {
|
| 429 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 430 |
+
"lower_is_better": false,
|
| 431 |
+
"score_type": "continuous",
|
| 432 |
+
"min_score": 0.0,
|
| 433 |
+
"max_score": 1.0
|
| 434 |
+
},
|
| 435 |
+
"score_details": {
|
| 436 |
+
"score": 0.5812897366030881,
|
| 437 |
+
"details": {
|
| 438 |
+
"seed_scores": "[0.5876475930971844, 0.5767484105358764, 0.5794732061762035]",
|
| 439 |
+
"seed_values": "[1234, 4158, 42]"
|
| 440 |
+
},
|
| 441 |
+
"uncertainty": {
|
| 442 |
+
"standard_error": {
|
| 443 |
+
"value": 0.0032747967987865603,
|
| 444 |
+
"method": "across_seeds"
|
| 445 |
+
},
|
| 446 |
+
"num_samples": 3
|
| 447 |
+
}
|
| 448 |
+
},
|
| 449 |
+
"generation_config": {
|
| 450 |
+
"generation_args": {
|
| 451 |
+
"temperature": 1.0,
|
| 452 |
+
"top_p": 0.95,
|
| 453 |
+
"top_k": 40.0,
|
| 454 |
+
"max_tokens": 64000,
|
| 455 |
+
"max_attempts": 1
|
| 456 |
+
},
|
| 457 |
+
"additional_details": {
|
| 458 |
+
"until": "[]",
|
| 459 |
+
"do_sample": "true",
|
| 460 |
+
"min_p": "0.0",
|
| 461 |
+
"presence_penalty": "1.5",
|
| 462 |
+
"repetition_penalty": "1.0",
|
| 463 |
+
"seed": "1234",
|
| 464 |
+
"num_fewshot": "0"
|
| 465 |
+
}
|
| 466 |
+
}
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"evaluation_name": "mmlu_pro_chat_math/custom-extract",
|
| 470 |
+
"source_data": {
|
| 471 |
+
"dataset_name": "mmlu_pro_chat",
|
| 472 |
+
"source_type": "hf_dataset",
|
| 473 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 474 |
+
"hf_split": "test"
|
| 475 |
+
},
|
| 476 |
+
"evaluation_timestamp": "1777088443",
|
| 477 |
+
"metric_config": {
|
| 478 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 479 |
+
"lower_is_better": false,
|
| 480 |
+
"score_type": "continuous",
|
| 481 |
+
"min_score": 0.0,
|
| 482 |
+
"max_score": 1.0
|
| 483 |
+
},
|
| 484 |
+
"score_details": {
|
| 485 |
+
"score": 0.9346163335800641,
|
| 486 |
+
"details": {
|
| 487 |
+
"seed_scores": "[0.9333826794966691, 0.9370836417468542, 0.9333826794966691]",
|
| 488 |
+
"seed_values": "[1234, 4158, 42]"
|
| 489 |
+
},
|
| 490 |
+
"uncertainty": {
|
| 491 |
+
"standard_error": {
|
| 492 |
+
"value": 0.0012336540833950418,
|
| 493 |
+
"method": "across_seeds"
|
| 494 |
+
},
|
| 495 |
+
"num_samples": 3
|
| 496 |
+
}
|
| 497 |
+
},
|
| 498 |
+
"generation_config": {
|
| 499 |
+
"generation_args": {
|
| 500 |
+
"temperature": 1.0,
|
| 501 |
+
"top_p": 0.95,
|
| 502 |
+
"top_k": 40.0,
|
| 503 |
+
"max_tokens": 64000,
|
| 504 |
+
"max_attempts": 1
|
| 505 |
+
},
|
| 506 |
+
"additional_details": {
|
| 507 |
+
"until": "[]",
|
| 508 |
+
"do_sample": "true",
|
| 509 |
+
"min_p": "0.0",
|
| 510 |
+
"presence_penalty": "1.5",
|
| 511 |
+
"repetition_penalty": "1.0",
|
| 512 |
+
"seed": "1234",
|
| 513 |
+
"num_fewshot": "0"
|
| 514 |
+
}
|
| 515 |
+
}
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"evaluation_name": "mmlu_pro_chat_other/custom-extract",
|
| 519 |
+
"source_data": {
|
| 520 |
+
"dataset_name": "mmlu_pro_chat",
|
| 521 |
+
"source_type": "hf_dataset",
|
| 522 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 523 |
+
"hf_split": "test"
|
| 524 |
+
},
|
| 525 |
+
"evaluation_timestamp": "1777088443",
|
| 526 |
+
"metric_config": {
|
| 527 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 528 |
+
"lower_is_better": false,
|
| 529 |
+
"score_type": "continuous",
|
| 530 |
+
"min_score": 0.0,
|
| 531 |
+
"max_score": 1.0
|
| 532 |
+
},
|
| 533 |
+
"score_details": {
|
| 534 |
+
"score": 0.7853535353535354,
|
| 535 |
+
"details": {
|
| 536 |
+
"seed_scores": "[0.7954545454545454, 0.775974025974026, 0.7846320346320347]",
|
| 537 |
+
"seed_values": "[1234, 4158, 42]"
|
| 538 |
+
},
|
| 539 |
+
"uncertainty": {
|
| 540 |
+
"standard_error": {
|
| 541 |
+
"value": 0.005635100776267413,
|
| 542 |
+
"method": "across_seeds"
|
| 543 |
+
},
|
| 544 |
+
"num_samples": 3
|
| 545 |
+
}
|
| 546 |
+
},
|
| 547 |
+
"generation_config": {
|
| 548 |
+
"generation_args": {
|
| 549 |
+
"temperature": 1.0,
|
| 550 |
+
"top_p": 0.95,
|
| 551 |
+
"top_k": 40.0,
|
| 552 |
+
"max_tokens": 64000,
|
| 553 |
+
"max_attempts": 1
|
| 554 |
+
},
|
| 555 |
+
"additional_details": {
|
| 556 |
+
"until": "[]",
|
| 557 |
+
"do_sample": "true",
|
| 558 |
+
"min_p": "0.0",
|
| 559 |
+
"presence_penalty": "1.5",
|
| 560 |
+
"repetition_penalty": "1.0",
|
| 561 |
+
"seed": "1234",
|
| 562 |
+
"num_fewshot": "0"
|
| 563 |
+
}
|
| 564 |
+
}
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"evaluation_name": "mmlu_pro_chat_philosophy/custom-extract",
|
| 568 |
+
"source_data": {
|
| 569 |
+
"dataset_name": "mmlu_pro_chat",
|
| 570 |
+
"source_type": "hf_dataset",
|
| 571 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 572 |
+
"hf_split": "test"
|
| 573 |
+
},
|
| 574 |
+
"evaluation_timestamp": "1777088443",
|
| 575 |
+
"metric_config": {
|
| 576 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 577 |
+
"lower_is_better": false,
|
| 578 |
+
"score_type": "continuous",
|
| 579 |
+
"min_score": 0.0,
|
| 580 |
+
"max_score": 1.0
|
| 581 |
+
},
|
| 582 |
+
"score_details": {
|
| 583 |
+
"score": 0.7247828991315964,
|
| 584 |
+
"details": {
|
| 585 |
+
"seed_scores": "[0.7234468937875751, 0.7274549098196392, 0.7234468937875751]",
|
| 586 |
+
"seed_values": "[1234, 4158, 42]"
|
| 587 |
+
},
|
| 588 |
+
"uncertainty": {
|
| 589 |
+
"standard_error": {
|
| 590 |
+
"value": 0.0013360053440213775,
|
| 591 |
+
"method": "across_seeds"
|
| 592 |
+
},
|
| 593 |
+
"num_samples": 3
|
| 594 |
+
}
|
| 595 |
+
},
|
| 596 |
+
"generation_config": {
|
| 597 |
+
"generation_args": {
|
| 598 |
+
"temperature": 1.0,
|
| 599 |
+
"top_p": 0.95,
|
| 600 |
+
"top_k": 40.0,
|
| 601 |
+
"max_tokens": 64000,
|
| 602 |
+
"max_attempts": 1
|
| 603 |
+
},
|
| 604 |
+
"additional_details": {
|
| 605 |
+
"until": "[]",
|
| 606 |
+
"do_sample": "true",
|
| 607 |
+
"min_p": "0.0",
|
| 608 |
+
"presence_penalty": "1.5",
|
| 609 |
+
"repetition_penalty": "1.0",
|
| 610 |
+
"seed": "1234",
|
| 611 |
+
"num_fewshot": "0"
|
| 612 |
+
}
|
| 613 |
+
}
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"evaluation_name": "mmlu_pro_chat_physics/custom-extract",
|
| 617 |
+
"source_data": {
|
| 618 |
+
"dataset_name": "mmlu_pro_chat",
|
| 619 |
+
"source_type": "hf_dataset",
|
| 620 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 621 |
+
"hf_split": "test"
|
| 622 |
+
},
|
| 623 |
+
"evaluation_timestamp": "1777088443",
|
| 624 |
+
"metric_config": {
|
| 625 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 626 |
+
"lower_is_better": false,
|
| 627 |
+
"score_type": "continuous",
|
| 628 |
+
"min_score": 0.0,
|
| 629 |
+
"max_score": 1.0
|
| 630 |
+
},
|
| 631 |
+
"score_details": {
|
| 632 |
+
"score": 0.8734924300744162,
|
| 633 |
+
"details": {
|
| 634 |
+
"seed_scores": "[0.8775981524249422, 0.8729792147806005, 0.8698999230177059]",
|
| 635 |
+
"seed_values": "[1234, 4158, 42]"
|
| 636 |
+
},
|
| 637 |
+
"uncertainty": {
|
| 638 |
+
"standard_error": {
|
| 639 |
+
"value": 0.002237053602022402,
|
| 640 |
+
"method": "across_seeds"
|
| 641 |
+
},
|
| 642 |
+
"num_samples": 3
|
| 643 |
+
}
|
| 644 |
+
},
|
| 645 |
+
"generation_config": {
|
| 646 |
+
"generation_args": {
|
| 647 |
+
"temperature": 1.0,
|
| 648 |
+
"top_p": 0.95,
|
| 649 |
+
"top_k": 40.0,
|
| 650 |
+
"max_tokens": 64000,
|
| 651 |
+
"max_attempts": 1
|
| 652 |
+
},
|
| 653 |
+
"additional_details": {
|
| 654 |
+
"until": "[]",
|
| 655 |
+
"do_sample": "true",
|
| 656 |
+
"min_p": "0.0",
|
| 657 |
+
"presence_penalty": "1.5",
|
| 658 |
+
"repetition_penalty": "1.0",
|
| 659 |
+
"seed": "1234",
|
| 660 |
+
"num_fewshot": "0"
|
| 661 |
+
}
|
| 662 |
+
}
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"evaluation_name": "mmlu_pro_chat_psychology/custom-extract",
|
| 666 |
+
"source_data": {
|
| 667 |
+
"dataset_name": "mmlu_pro_chat",
|
| 668 |
+
"source_type": "hf_dataset",
|
| 669 |
+
"hf_repo": "TIGER-Lab/MMLU-Pro",
|
| 670 |
+
"hf_split": "test"
|
| 671 |
+
},
|
| 672 |
+
"evaluation_timestamp": "1777088443",
|
| 673 |
+
"metric_config": {
|
| 674 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 675 |
+
"lower_is_better": false,
|
| 676 |
+
"score_type": "continuous",
|
| 677 |
+
"min_score": 0.0,
|
| 678 |
+
"max_score": 1.0
|
| 679 |
+
},
|
| 680 |
+
"score_details": {
|
| 681 |
+
"score": 0.8074352548036758,
|
| 682 |
+
"details": {
|
| 683 |
+
"seed_scores": "[0.7944862155388471, 0.8233082706766918, 0.8045112781954887]",
|
| 684 |
+
"seed_values": "[1234, 4158, 42]"
|
| 685 |
+
},
|
| 686 |
+
"uncertainty": {
|
| 687 |
+
"standard_error": {
|
| 688 |
+
"value": 0.008447681042671974,
|
| 689 |
+
"method": "across_seeds"
|
| 690 |
+
},
|
| 691 |
+
"num_samples": 3
|
| 692 |
+
}
|
| 693 |
+
},
|
| 694 |
+
"generation_config": {
|
| 695 |
+
"generation_args": {
|
| 696 |
+
"temperature": 1.0,
|
| 697 |
+
"top_p": 0.95,
|
| 698 |
+
"top_k": 40.0,
|
| 699 |
+
"max_tokens": 64000,
|
| 700 |
+
"max_attempts": 1
|
| 701 |
+
},
|
| 702 |
+
"additional_details": {
|
| 703 |
+
"until": "[]",
|
| 704 |
+
"do_sample": "true",
|
| 705 |
+
"min_p": "0.0",
|
| 706 |
+
"presence_penalty": "1.5",
|
| 707 |
+
"repetition_penalty": "1.0",
|
| 708 |
+
"seed": "1234",
|
| 709 |
+
"num_fewshot": "0"
|
| 710 |
+
}
|
| 711 |
+
}
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"evaluation_name": "mmlu_pro_chat/custom-extract",
|
| 715 |
+
"source_data": {
|
| 716 |
+
"dataset_name": "mmlu_pro_chat",
|
| 717 |
+
"source_type": "other"
|
| 718 |
+
},
|
| 719 |
+
"evaluation_timestamp": "1777088443",
|
| 720 |
+
"metric_config": {
|
| 721 |
+
"evaluation_description": "exact_match (filter: custom-extract)",
|
| 722 |
+
"lower_is_better": false,
|
| 723 |
+
"score_type": "continuous",
|
| 724 |
+
"min_score": 0.0,
|
| 725 |
+
"max_score": 1.0
|
| 726 |
+
},
|
| 727 |
+
"score_details": {
|
| 728 |
+
"score": 0.8124722960992908,
|
| 729 |
+
"details": {
|
| 730 |
+
"seed_scores": "[0.8144115691489362, 0.8125, 0.8105053191489362]",
|
| 731 |
+
"seed_values": "[1234, 4158, 42]"
|
| 732 |
+
},
|
| 733 |
+
"uncertainty": {
|
| 734 |
+
"standard_error": {
|
| 735 |
+
"value": 0.0011277223203151384,
|
| 736 |
+
"method": "across_seeds"
|
| 737 |
+
},
|
| 738 |
+
"num_samples": 3
|
| 739 |
+
}
|
| 740 |
+
}
|
| 741 |
+
}
|
| 742 |
+
]
|
| 743 |
+
}
|