MLLM_leaderboard / eval-results /JosephusCheung /Pwen-VL-Chat-20_30 /results_2023-11-04T15-47-38.506264.json
Wwwduojin's picture
Upload 1529 files
a312f2f
{
"config_general": {
"lighteval_sha": "167773f1d5d1647c60dadc31c9e731ab7dbcbbad",
"num_few_shot_default": 0,
"num_fewshot_seeds": 1,
"override_batch_size": 1,
"max_samples": null,
"job_id": "",
"model_name": "JosephusCheung/Pwen-VL-Chat-20_30",
"model_sha": "64a9b89fb18140fc1af1f11471dc9fe34ebc7446",
"model_dtype": "torch.bfloat16",
"model_size": "14.51 GB"
},
"results": {
"harness|drop|3": {
"em": 0.3234060402684564,
"em_stderr": 0.004790466119380845,
"f1": 0.3795564177852361,
"f1_stderr": 0.004705234681743664
},
"harness|gsm8k|5": {
"acc": 0.1910538286580743,
"acc_stderr": 0.010828791191755175
},
"harness|winogrande|5": {
"acc": 0.6835043409629045,
"acc_stderr": 0.013071868328051477
},
"all": {
"em": 0.3234060402684564,
"em_stderr": 0.004790466119380845,
"f1": 0.3795564177852361,
"f1_stderr": 0.004705234681743664,
"acc": 0.4372790848104894,
"acc_stderr": 0.011950329759903327
}
},
"versions": {
"all": 0,
"harness|drop|3": 1,
"harness|gsm8k|5": 0,
"harness|winogrande|5": 0
},
"config_tasks": {
"harness|drop": "LM Harness task",
"harness|gsm8k": "LM Harness task",
"harness|winogrande": "LM Harness task"
},
"summary_tasks": {
"harness|drop|3": {
"hashes": {
"hash_examples": "1d27416e8324e9a3",
"hash_full_prompts": "a5513ff9a741b385",
"hash_input_tokens": "e4d9d658ccb42fc3",
"hash_cont_tokens": "3acffef0fcdaa041"
},
"truncated": 0,
"non_truncated": 9536,
"padded": 0,
"non_padded": 9536,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|gsm8k|5": {
"hashes": {
"hash_examples": "4c0843a5d99bcfdc",
"hash_full_prompts": "41d55e83abc0e02d",
"hash_input_tokens": "2282d6efefcc7579",
"hash_cont_tokens": "dcabb1d9463db6f1"
},
"truncated": 0,
"non_truncated": 1319,
"padded": 0,
"non_padded": 1319,
"effective_few_shots": 5.0,
"num_truncated_few_shots": 0
},
"harness|winogrande|5": {
"hashes": {
"hash_examples": "aada0a176fd81218",
"hash_full_prompts": "c8655cbd12de8409",
"hash_input_tokens": "288ed7294cb59f7d",
"hash_cont_tokens": "f4a307afe0c47a4a"
},
"truncated": 0,
"non_truncated": 1267,
"padded": 2429,
"non_padded": 105,
"effective_few_shots": 5.0,
"num_truncated_few_shots": 0
}
},
"summary_general": {
"hashes": {
"hash_examples": "9b4d8993161e637d",
"hash_full_prompts": "08215e527b7e60a5",
"hash_input_tokens": "329083a90a12723b",
"hash_cont_tokens": "1578bed2931bae1e"
},
"truncated": 0,
"non_truncated": 12122,
"padded": 2429,
"non_padded": 10960,
"num_truncated_few_shots": 0,
"total_evaluation_time_secondes": 0
}
}