{"task_name": "mmlu_business_ethics:mc", "task_hash": "7de417726ca2cc155dd1475a38afc381", "model_hash": "03418cf8091a9882619950ffb07429a5", "model_config": {"model": "hf_checkpoints/dclm-pool-1b-1x-h-fasttext-open_lm_1b_swiglutorch-warm5000-lr0p003-wd0p033-cd3e-05-bs256-mult1-seed124-tokens28795904000", "revision": null, "trust_remote_code": null, "max_length": 2048, "model_path": null, "model_type": "hf"}, "task_config": {"task_name": "mmlu_business_ethics:mc", "task_core": "mmlu_business_ethics", "limit": null, "split": "test", "num_shots": 5, "fewshot_seed": 1234, "primary_metric": "acc_raw", "random_subsample_seed": 1234, "context_kwargs": {"no_label_prefix_space": false}, "generation_kwargs": null, "metric_kwargs": null, "native_id_field": "index", "fewshot_source": null, "dataset_path": "cais/mmlu", "dataset_name": "business_ethics", "use_chat_format": null, "version": 1, "revision": null, "metadata": {"regimes": ["OLMES-v0.1"], "alias": "mmlu_business_ethics:mc::olmes"}}, "compute_config": {"batch_size": "16", "max_batch_size": 32, "output_dir": "hf_checkpoints/dclm-pool-1b-1x-h-fasttext-open_lm_1b_swiglutorch-warm5000-lr0p003-wd0p033-cd3e-05-bs256-mult1-seed124-tokens28795904000/olmes_fp32/mmlu", "num_recorded_inputs": 3, "save_raw_requests": true, "recompute_metrics": false, "check_datalake": false, "push_datalake": false, "no_datalake": false}, "processing_time": 4.340962648391724, "current_date": "2024-11-19 21:10:27 UTC", "num_instances": 100, "beaker_info": {}, "metrics": {"acc_raw": 0.2, "acc_per_token": 0.2, "acc_per_char": 0.2, "correct_loss_raw": 1.4059372037649154, "incorrect_loss_raw": 1.4068191585938135, "correct_loss_per_token": 1.4059372037649154, "incorrect_loss_per_token": 1.4068191585938135, "correct_loss_per_char": 0.7029686018824577, "incorrect_loss_per_char": 0.7034095792969067, "primary_score": 0.2}, "task_idx": 3} |