| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.2719526290893555, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5617, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7192982456140351, | |
| "eval_f1": 0.6543102914784331, | |
| "eval_loss": 0.511669397354126, | |
| "eval_precision": 0.6580196140461879, | |
| "eval_recall": 0.6513911620294599, | |
| "eval_runtime": 5.1602, | |
| "eval_samples_per_second": 77.323, | |
| "eval_steps_per_second": 9.69, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.8476994037628174, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5046, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7418546365914787, | |
| "eval_f1": 0.7111966887091448, | |
| "eval_loss": 0.49168047308921814, | |
| "eval_precision": 0.7042004048582996, | |
| "eval_recall": 0.7323604291689398, | |
| "eval_runtime": 5.0484, | |
| "eval_samples_per_second": 79.035, | |
| "eval_steps_per_second": 9.904, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.1617326736450195, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4798, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7593984962406015, | |
| "eval_f1": 0.7179215270413574, | |
| "eval_loss": 0.4465886950492859, | |
| "eval_precision": 0.7129198966408269, | |
| "eval_recall": 0.7247681396617567, | |
| "eval_runtime": 5.0834, | |
| "eval_samples_per_second": 78.491, | |
| "eval_steps_per_second": 9.836, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.705305337905884, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4374, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.7740798993394149, | |
| "eval_loss": 0.3993551731109619, | |
| "eval_precision": 0.7865881658357387, | |
| "eval_recall": 0.7648208765230042, | |
| "eval_runtime": 5.1044, | |
| "eval_samples_per_second": 78.167, | |
| "eval_steps_per_second": 9.795, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.9225200414657593, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4037, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7575187969924813, | |
| "eval_loss": 0.41500648856163025, | |
| "eval_precision": 0.748013422818792, | |
| "eval_recall": 0.7800054555373704, | |
| "eval_runtime": 5.0621, | |
| "eval_samples_per_second": 78.822, | |
| "eval_steps_per_second": 9.877, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.181605339050293, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3741, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8049369344976196, | |
| "eval_loss": 0.3736521899700165, | |
| "eval_precision": 0.8027777777777778, | |
| "eval_recall": 0.8072376795781051, | |
| "eval_runtime": 5.0549, | |
| "eval_samples_per_second": 78.934, | |
| "eval_steps_per_second": 9.891, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.7816860675811768, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3574, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8220551378446115, | |
| "eval_f1": 0.790906742443813, | |
| "eval_loss": 0.3775876462459564, | |
| "eval_precision": 0.7844931964944649, | |
| "eval_recall": 0.7990998363338788, | |
| "eval_runtime": 5.059, | |
| "eval_samples_per_second": 78.869, | |
| "eval_steps_per_second": 9.883, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.030299663543701, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3387, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8134839254478557, | |
| "eval_loss": 0.3653636872768402, | |
| "eval_precision": 0.8119747899159664, | |
| "eval_recall": 0.8150572831423895, | |
| "eval_runtime": 5.1112, | |
| "eval_samples_per_second": 78.064, | |
| "eval_steps_per_second": 9.782, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.920233726501465, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3293, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8067969337812972, | |
| "eval_loss": 0.3626542389392853, | |
| "eval_precision": 0.8021114369501466, | |
| "eval_recall": 0.8122385888343335, | |
| "eval_runtime": 5.0586, | |
| "eval_samples_per_second": 78.875, | |
| "eval_steps_per_second": 9.884, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.010580539703369, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3209, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8039756622954537, | |
| "eval_loss": 0.35534289479255676, | |
| "eval_precision": 0.8032299897460643, | |
| "eval_recall": 0.8047372249499909, | |
| "eval_runtime": 5.0627, | |
| "eval_samples_per_second": 78.811, | |
| "eval_steps_per_second": 9.876, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.613595485687256, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2967, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8051873113570456, | |
| "eval_loss": 0.3674112856388092, | |
| "eval_precision": 0.7989231125521075, | |
| "eval_recall": 0.8129659938170577, | |
| "eval_runtime": 5.0568, | |
| "eval_samples_per_second": 78.903, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 11.547273635864258, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2928, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.8026866442779643, | |
| "eval_loss": 0.37071213126182556, | |
| "eval_precision": 0.795995733394834, | |
| "eval_recall": 0.8111929441716675, | |
| "eval_runtime": 5.0714, | |
| "eval_samples_per_second": 78.677, | |
| "eval_steps_per_second": 9.859, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.8734816312789917, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2967, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8160386984618873, | |
| "eval_loss": 0.3514226973056793, | |
| "eval_precision": 0.8152632848784607, | |
| "eval_recall": 0.8168303327877796, | |
| "eval_runtime": 5.0516, | |
| "eval_samples_per_second": 78.984, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 11.493008613586426, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2934, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8118502107020158, | |
| "eval_loss": 0.3507131040096283, | |
| "eval_precision": 0.8083091673078061, | |
| "eval_recall": 0.8157846881251136, | |
| "eval_runtime": 5.0578, | |
| "eval_samples_per_second": 78.888, | |
| "eval_steps_per_second": 9.886, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.0511583089828491, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2811, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8042838456507522, | |
| "eval_loss": 0.35527709126472473, | |
| "eval_precision": 0.7990802919708029, | |
| "eval_recall": 0.8104655391889435, | |
| "eval_runtime": 5.0928, | |
| "eval_samples_per_second": 78.345, | |
| "eval_steps_per_second": 9.818, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 7.683447360992432, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2738, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8136136136136136, | |
| "eval_loss": 0.35545966029167175, | |
| "eval_precision": 0.8076923076923077, | |
| "eval_recall": 0.820785597381342, | |
| "eval_runtime": 5.1153, | |
| "eval_samples_per_second": 78.002, | |
| "eval_steps_per_second": 9.775, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 5.83898401260376, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2717, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8203781512605042, | |
| "eval_loss": 0.34679991006851196, | |
| "eval_precision": 0.8174088828111065, | |
| "eval_recall": 0.823604291689398, | |
| "eval_runtime": 5.1026, | |
| "eval_samples_per_second": 78.196, | |
| "eval_steps_per_second": 9.799, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.768757343292236, | |
| "learning_rate": 5e-06, | |
| "loss": 0.278, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8127416435111035, | |
| "eval_loss": 0.3509637117385864, | |
| "eval_precision": 0.8079618768328446, | |
| "eval_recall": 0.8182851427532278, | |
| "eval_runtime": 5.0773, | |
| "eval_samples_per_second": 78.585, | |
| "eval_steps_per_second": 9.848, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.281126022338867, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2701, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8178232198860786, | |
| "eval_loss": 0.34708452224731445, | |
| "eval_precision": 0.8142125821151684, | |
| "eval_recall": 0.8218312420440079, | |
| "eval_runtime": 5.0544, | |
| "eval_samples_per_second": 78.942, | |
| "eval_steps_per_second": 9.892, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.043084621429443, | |
| "learning_rate": 0.0, | |
| "loss": 0.2722, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8152777777777778, | |
| "eval_loss": 0.34833839535713196, | |
| "eval_precision": 0.8110639802050195, | |
| "eval_recall": 0.8200581923986179, | |
| "eval_runtime": 5.0742, | |
| "eval_samples_per_second": 78.633, | |
| "eval_steps_per_second": 9.854, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7597037114448000.0, | |
| "train_loss": 0.34670459247026286, | |
| "train_runtime": 1953.5719, | |
| "train_samples_per_second": 37.245, | |
| "train_steps_per_second": 1.249 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7597037114448000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |