| { | |
| "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": 5.93964147567749, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5623, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6301313943104988, | |
| "eval_loss": 0.5216777324676514, | |
| "eval_precision": 0.6603762281332375, | |
| "eval_recall": 0.6217039461720313, | |
| "eval_runtime": 5.0877, | |
| "eval_samples_per_second": 78.424, | |
| "eval_steps_per_second": 9.828, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.8485560417175293, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5061, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7568922305764411, | |
| "eval_f1": 0.7089058868656783, | |
| "eval_loss": 0.48978015780448914, | |
| "eval_precision": 0.7074372759856631, | |
| "eval_recall": 0.7104928168757956, | |
| "eval_runtime": 5.0629, | |
| "eval_samples_per_second": 78.809, | |
| "eval_steps_per_second": 9.876, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 5.005080699920654, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4443, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8120300751879699, | |
| "eval_f1": 0.7678883071553229, | |
| "eval_loss": 0.40850991010665894, | |
| "eval_precision": 0.7750572737686139, | |
| "eval_recall": 0.7620021822149482, | |
| "eval_runtime": 5.144, | |
| "eval_samples_per_second": 77.566, | |
| "eval_steps_per_second": 9.72, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.8029582500457764, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3805, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7751786979200206, | |
| "eval_loss": 0.3672122061252594, | |
| "eval_precision": 0.7980263157894737, | |
| "eval_recall": 0.7608656119294417, | |
| "eval_runtime": 5.0838, | |
| "eval_samples_per_second": 78.485, | |
| "eval_steps_per_second": 9.835, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.751908540725708, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3488, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8229427559286084, | |
| "eval_loss": 0.35348454117774963, | |
| "eval_precision": 0.8206541218637993, | |
| "eval_recall": 0.8253773413347881, | |
| "eval_runtime": 5.0536, | |
| "eval_samples_per_second": 78.953, | |
| "eval_steps_per_second": 9.894, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.316796779632568, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3156, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8254579780661698, | |
| "eval_loss": 0.33368828892707825, | |
| "eval_precision": 0.8299216027874565, | |
| "eval_recall": 0.8214220767412257, | |
| "eval_runtime": 5.0533, | |
| "eval_samples_per_second": 78.959, | |
| "eval_steps_per_second": 9.895, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 18.979951858520508, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3055, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8297847585805701, | |
| "eval_loss": 0.3217175304889679, | |
| "eval_precision": 0.8385357006491028, | |
| "eval_recall": 0.8224677214038916, | |
| "eval_runtime": 5.0744, | |
| "eval_samples_per_second": 78.63, | |
| "eval_steps_per_second": 9.853, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.143575191497803, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2995, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8271551457392166, | |
| "eval_loss": 0.31452828645706177, | |
| "eval_precision": 0.8347358430876305, | |
| "eval_recall": 0.8206946717585015, | |
| "eval_runtime": 5.0883, | |
| "eval_samples_per_second": 78.414, | |
| "eval_steps_per_second": 9.826, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.147844314575195, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2825, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8393634395533442, | |
| "eval_loss": 0.3090471923351288, | |
| "eval_precision": 0.8402278542707444, | |
| "eval_recall": 0.8385160938352427, | |
| "eval_runtime": 5.0782, | |
| "eval_samples_per_second": 78.572, | |
| "eval_steps_per_second": 9.846, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.50480842590332, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.272, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8461962888779714, | |
| "eval_loss": 0.29923897981643677, | |
| "eval_precision": 0.8453465227094517, | |
| "eval_recall": 0.8470631023822512, | |
| "eval_runtime": 5.0745, | |
| "eval_samples_per_second": 78.629, | |
| "eval_steps_per_second": 9.853, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.085771083831787, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2626, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8439846096096095, | |
| "eval_loss": 0.30075788497924805, | |
| "eval_precision": 0.8568027210884354, | |
| "eval_recall": 0.8338334242589562, | |
| "eval_runtime": 5.0607, | |
| "eval_samples_per_second": 78.843, | |
| "eval_steps_per_second": 9.88, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.957924842834473, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2641, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8488361520276414, | |
| "eval_loss": 0.2949255406856537, | |
| "eval_precision": 0.8488361520276414, | |
| "eval_recall": 0.8488361520276414, | |
| "eval_runtime": 5.0742, | |
| "eval_samples_per_second": 78.634, | |
| "eval_steps_per_second": 9.854, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 3.114020347595215, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.257, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8475258334958082, | |
| "eval_loss": 0.28850188851356506, | |
| "eval_precision": 0.8591828192414193, | |
| "eval_recall": 0.8381069285324605, | |
| "eval_runtime": 5.0549, | |
| "eval_samples_per_second": 78.934, | |
| "eval_steps_per_second": 9.891, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.347434043884277, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2473, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8568221901555235, | |
| "eval_loss": 0.2826312482357025, | |
| "eval_precision": 0.8596491228070176, | |
| "eval_recall": 0.8541553009638116, | |
| "eval_runtime": 5.0506, | |
| "eval_samples_per_second": 79.001, | |
| "eval_steps_per_second": 9.9, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.6674721240997314, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2456, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.2825632393360138, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 5.071, | |
| "eval_samples_per_second": 78.683, | |
| "eval_steps_per_second": 9.86, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 5.605799674987793, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2477, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8602260265626904, | |
| "eval_loss": 0.2795054018497467, | |
| "eval_precision": 0.8620943049601959, | |
| "eval_recall": 0.8584288052373159, | |
| "eval_runtime": 5.063, | |
| "eval_samples_per_second": 78.807, | |
| "eval_steps_per_second": 9.876, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.4996304512023926, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2426, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8526315789473684, | |
| "eval_loss": 0.2793760895729065, | |
| "eval_precision": 0.8585304054054055, | |
| "eval_recall": 0.8473813420621932, | |
| "eval_runtime": 5.0574, | |
| "eval_samples_per_second": 78.894, | |
| "eval_steps_per_second": 9.886, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.353811264038086, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2359, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8629148629148629, | |
| "eval_loss": 0.27959930896759033, | |
| "eval_precision": 0.8657894736842104, | |
| "eval_recall": 0.860201854882706, | |
| "eval_runtime": 5.0719, | |
| "eval_samples_per_second": 78.668, | |
| "eval_steps_per_second": 9.858, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 6.069816589355469, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2417, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8587719298245614, | |
| "eval_loss": 0.27870801091194153, | |
| "eval_precision": 0.864771021021021, | |
| "eval_recall": 0.8534278959810875, | |
| "eval_runtime": 5.0523, | |
| "eval_samples_per_second": 78.973, | |
| "eval_steps_per_second": 9.896, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.868233680725098, | |
| "learning_rate": 0.0, | |
| "loss": 0.2319, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8587719298245614, | |
| "eval_loss": 0.2786270081996918, | |
| "eval_precision": 0.864771021021021, | |
| "eval_recall": 0.8534278959810875, | |
| "eval_runtime": 5.0608, | |
| "eval_samples_per_second": 78.841, | |
| "eval_steps_per_second": 9.88, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.30967485005738304, | |
| "train_runtime": 1934.4038, | |
| "train_samples_per_second": 37.614, | |
| "train_steps_per_second": 1.261 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7609911792720000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |