{ "best_metric": 0.4975282549858093, "best_model_checkpoint": "/content/drive/MyDrive/Azari/Models/POS_TAGGER_v1/checkpoint-45", "epoch": 15.0, "global_step": 45, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 1.9e-05, "loss": 2.0841, "step": 3 }, { "epoch": 1.0, "eval_accuracy": 0.6952554744525548, "eval_f1": 0.28720626631853785, "eval_loss": 1.180311918258667, "eval_precision": 0.45454545454545453, "eval_recall": 0.2099236641221374, "eval_runtime": 0.0976, "eval_samples_per_second": 204.909, "eval_steps_per_second": 10.245, "step": 3 }, { "epoch": 2.0, "learning_rate": 1.8e-05, "loss": 1.1453, "step": 6 }, { "epoch": 2.0, "eval_accuracy": 0.7627737226277372, "eval_f1": 0.4230769230769231, "eval_loss": 0.930380642414093, "eval_precision": 0.48058252427184467, "eval_recall": 0.37786259541984735, "eval_runtime": 0.0924, "eval_samples_per_second": 216.409, "eval_steps_per_second": 10.82, "step": 6 }, { "epoch": 3.0, "learning_rate": 1.7e-05, "loss": 0.8983, "step": 9 }, { "epoch": 3.0, "eval_accuracy": 0.8047445255474452, "eval_f1": 0.532258064516129, "eval_loss": 0.7620761394500732, "eval_precision": 0.5641025641025641, "eval_recall": 0.5038167938931297, "eval_runtime": 0.1238, "eval_samples_per_second": 161.595, "eval_steps_per_second": 8.08, "step": 9 }, { "epoch": 4.0, "learning_rate": 1.6000000000000003e-05, "loss": 0.7406, "step": 12 }, { "epoch": 4.0, "eval_accuracy": 0.8375912408759124, "eval_f1": 0.6052104208416834, "eval_loss": 0.6653763651847839, "eval_precision": 0.6371308016877637, "eval_recall": 0.5763358778625954, "eval_runtime": 0.131, "eval_samples_per_second": 152.687, "eval_steps_per_second": 7.634, "step": 12 }, { "epoch": 5.0, "learning_rate": 1.5000000000000002e-05, "loss": 0.6451, "step": 15 }, { "epoch": 5.0, "eval_accuracy": 0.8375912408759124, "eval_f1": 0.6127167630057803, "eval_loss": 0.6271026730537415, "eval_precision": 0.6186770428015564, "eval_recall": 0.6068702290076335, "eval_runtime": 0.117, "eval_samples_per_second": 170.987, "eval_steps_per_second": 8.549, "step": 15 }, { "epoch": 6.0, "learning_rate": 1.4e-05, "loss": 0.5668, "step": 18 }, { "epoch": 6.0, "eval_accuracy": 0.8375912408759124, "eval_f1": 0.6135922330097087, "eval_loss": 0.589028537273407, "eval_precision": 0.6245059288537549, "eval_recall": 0.6030534351145038, "eval_runtime": 0.273, "eval_samples_per_second": 73.251, "eval_steps_per_second": 3.663, "step": 18 }, { "epoch": 7.0, "learning_rate": 1.3000000000000001e-05, "loss": 0.5085, "step": 21 }, { "epoch": 7.0, "eval_accuracy": 0.8521897810218978, "eval_f1": 0.6459143968871595, "eval_loss": 0.5673397779464722, "eval_precision": 0.6587301587301587, "eval_recall": 0.6335877862595419, "eval_runtime": 0.1023, "eval_samples_per_second": 195.551, "eval_steps_per_second": 9.778, "step": 21 }, { "epoch": 8.0, "learning_rate": 1.2e-05, "loss": 0.4746, "step": 24 }, { "epoch": 8.0, "eval_accuracy": 0.8540145985401459, "eval_f1": 0.6472868217054263, "eval_loss": 0.5449931621551514, "eval_precision": 0.65748031496063, "eval_recall": 0.6374045801526718, "eval_runtime": 0.0875, "eval_samples_per_second": 228.5, "eval_steps_per_second": 11.425, "step": 24 }, { "epoch": 9.0, "learning_rate": 1.1000000000000001e-05, "loss": 0.4312, "step": 27 }, { "epoch": 9.0, "eval_accuracy": 0.8448905109489051, "eval_f1": 0.6374045801526718, "eval_loss": 0.5437765121459961, "eval_precision": 0.6374045801526718, "eval_recall": 0.6374045801526718, "eval_runtime": 0.1112, "eval_samples_per_second": 179.849, "eval_steps_per_second": 8.992, "step": 27 }, { "epoch": 10.0, "learning_rate": 1e-05, "loss": 0.3862, "step": 30 }, { "epoch": 10.0, "eval_accuracy": 0.8540145985401459, "eval_f1": 0.6511627906976745, "eval_loss": 0.5256561040878296, "eval_precision": 0.6614173228346457, "eval_recall": 0.6412213740458015, "eval_runtime": 0.0917, "eval_samples_per_second": 218.041, "eval_steps_per_second": 10.902, "step": 30 }, { "epoch": 11.0, "learning_rate": 9e-06, "loss": 0.3544, "step": 33 }, { "epoch": 11.0, "eval_accuracy": 0.8613138686131386, "eval_f1": 0.665377176015474, "eval_loss": 0.5157161951065063, "eval_precision": 0.6745098039215687, "eval_recall": 0.6564885496183206, "eval_runtime": 0.1237, "eval_samples_per_second": 161.719, "eval_steps_per_second": 8.086, "step": 33 }, { "epoch": 12.0, "learning_rate": 8.000000000000001e-06, "loss": 0.3395, "step": 36 }, { "epoch": 12.0, "eval_accuracy": 0.8686131386861314, "eval_f1": 0.6821705426356589, "eval_loss": 0.5101653337478638, "eval_precision": 0.6929133858267716, "eval_recall": 0.6717557251908397, "eval_runtime": 0.0907, "eval_samples_per_second": 220.625, "eval_steps_per_second": 11.031, "step": 36 }, { "epoch": 13.0, "learning_rate": 7e-06, "loss": 0.3123, "step": 39 }, { "epoch": 13.0, "eval_accuracy": 0.8686131386861314, "eval_f1": 0.6807692307692308, "eval_loss": 0.505219578742981, "eval_precision": 0.686046511627907, "eval_recall": 0.6755725190839694, "eval_runtime": 0.0947, "eval_samples_per_second": 211.236, "eval_steps_per_second": 10.562, "step": 39 }, { "epoch": 14.0, "learning_rate": 6e-06, "loss": 0.2963, "step": 42 }, { "epoch": 14.0, "eval_accuracy": 0.8686131386861314, "eval_f1": 0.6870229007633588, "eval_loss": 0.502780556678772, "eval_precision": 0.6870229007633588, "eval_recall": 0.6870229007633588, "eval_runtime": 0.093, "eval_samples_per_second": 215.022, "eval_steps_per_second": 10.751, "step": 42 }, { "epoch": 15.0, "learning_rate": 5e-06, "loss": 0.2763, "step": 45 }, { "epoch": 15.0, "eval_accuracy": 0.8686131386861314, "eval_f1": 0.6794625719769674, "eval_loss": 0.4975282549858093, "eval_precision": 0.6833976833976834, "eval_recall": 0.6755725190839694, "eval_runtime": 0.1181, "eval_samples_per_second": 169.318, "eval_steps_per_second": 8.466, "step": 45 } ], "max_steps": 60, "num_train_epochs": 20, "total_flos": 78396114124800.0, "trial_name": null, "trial_params": null }