MLLM_leaderboard / eval-results /JosephusCheung /Pwen-7B-Chat-20_30 /results_2023-10-26T02-42-36.258115.json
Wwwduojin's picture
Upload 1529 files
a312f2f
{
"config_general": {
"model_name": "JosephusCheung/Pwen-7B-Chat-20_30",
"model_sha": "da8f573dfae6cc36e89b8bc40a036feb411978d9",
"model_size": "14.51 GB",
"model_dtype": "torch.bfloat16",
"lighteval_sha": "0f318ecf002208468154899217b3ba7c6ae09374",
"num_few_shot_default": 0,
"num_fewshot_seeds": 1,
"override_batch_size": 1,
"max_samples": null,
"job_id": ""
},
"results": {
"harness|drop|3": {
"em": 0.2954068791946309,
"em_stderr": 0.004672175556184236,
"f1": 0.3814209312080561,
"f1_stderr": 0.004573085663083055
},
"harness|gsm8k|5": {
"acc": 0.20621683093252463,
"acc_stderr": 0.011144364089781436
},
"harness|winogrande|5": {
"acc": 0.6842936069455406,
"acc_stderr": 0.01306309474300081
},
"all": {
"em": 0.2954068791946309,
"em_stderr": 0.004672175556184236,
"f1": 0.3814209312080561,
"f1_stderr": 0.004573085663083055,
"acc": 0.44525521893903264,
"acc_stderr": 0.012103729416391124
}
},
"versions": {
"harness|drop|3": 1,
"harness|gsm8k|5": 0,
"harness|winogrande|5": 0,
"all": 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": "3e5fda34e012b148"
},
"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": "3ca497ba221bb53d"
},
"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": 2534,
"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": "817b8222844b59c1"
},
"total_evaluation_time_secondes": "7674.348122358322",
"truncated": 0,
"non-truncated": 13389,
"padded": 2429,
"non-padded": 10960,
"num_truncated_few_shots": 0
}
}