{"task_name": "arc_easy", "task_hash": "ed6704ae05bb260463787386ca9d78ee", "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": "arc_easy", "task_core": "arc_easy", "limit": 1000, "split": "test", "num_shots": 5, "fewshot_seed": 1234, "primary_metric": "acc_per_char", "random_subsample_seed": 1234, "context_kwargs": {"description": null}, "generation_kwargs": null, "metric_kwargs": {"uncond_docid_offset": 1000000}, "native_id_field": "id", "fewshot_source": "OLMES:ARC-Easy", "dataset_path": "ai2_arc", "dataset_name": "ARC-Easy", "use_chat_format": null, "version": 0, "revision": null, "metadata": {"description": "ARC-Easy (RC) using OLMES-v0.1", "regimes": ["OLMES-v0.1"], "alias": "arc_easy:rc::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/core_9mcqa", "num_recorded_inputs": 3, "save_raw_requests": true, "recompute_metrics": false, "check_datalake": false, "push_datalake": false, "no_datalake": false}, "processing_time": 44.29335308074951, "current_date": "2024-11-19 21:10:37 UTC", "num_instances": 1000, "beaker_info": {}, "metrics": {"acc_raw": 0.694, "acc_per_token": 0.693, "acc_per_char": 0.714, "correct_loss_raw": 9.330107917129993, "incorrect_loss_raw": 14.000815374672404, "correct_loss_per_token": 2.2515681965954655, "incorrect_loss_per_token": 3.9243185699953256, "correct_loss_per_char": 0.41603878481733886, "incorrect_loss_per_char": 0.7056315904639505, "acc_uncond": 0.646, "correct_loss_uncond": -13.524798529326915, "incorrect_loss_uncond": -10.111074344178036, "primary_score": 0.714}, "task_idx": 1} |