MLLM_leaderboard / eval-results /chavinlo /gpt4-x-alpaca /results_2023-09-22T20-56-09.987040.json
Wwwduojin's picture
Upload 1529 files
a312f2f
{
"config_general": {
"model_name": "chavinlo/gpt4-x-alpaca",
"model_sha": "6a571f458cab9a23d14324ec63e0abd1744c8353",
"model_size": "24.28 GB",
"model_dtype": "torch.float16",
"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": 0.15478187919463088,
"em_stderr": 0.003704111989193061,
"f1": 0.24988045302013467,
"f1_stderr": 0.00385619985047934
},
"harness|gsm8k|5": {
"acc": 0.028051554207733132,
"acc_stderr": 0.004548229533836362
},
"harness|winogrande|5": {
"acc": 0.7016574585635359,
"acc_stderr": 0.012858885010030421
},
"all": {
"em": 0.15478187919463088,
"em_stderr": 0.003704111989193061,
"f1": 0.24988045302013467,
"f1_stderr": 0.00385619985047934,
"acc": 0.3648545063856345,
"acc_stderr": 0.008703557271933391
}
},
"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": "f70227603c1b1bfe",
"hash_cont_tokens": "030717cde54f2dff"
},
"truncated": 1263,
"non-truncated": 8273,
"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": "e3d5b3003c52b880",
"hash_cont_tokens": "5d7d3555e4c882cf"
},
"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": "5be2b0947cee07a9",
"hash_cont_tokens": "f08975ad6f2d5864"
},
"truncated": 0,
"non-truncated": 2534,
"padded": 2432,
"non-padded": 102,
"effective_few_shots": 5.0,
"num_truncated_few_shots": 0
}
},
"summary_general": {
"hashes": {
"hash_examples": "9b4d8993161e637d",
"hash_full_prompts": "08215e527b7e60a5",
"hash_input_tokens": "ce9af2df9f2847fa",
"hash_cont_tokens": "be421a34d9b78c47"
},
"total_evaluation_time_secondes": "14067.204483509064",
"truncated": 1263,
"non-truncated": 12126,
"padded": 2432,
"non-padded": 10957,
"num_truncated_few_shots": 0
}
}