| { | |
| "best_metric": 0.9241842610364683, | |
| "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-3760", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 4700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9985068813295248, | |
| "eval_f1": 0.9050170482221139, | |
| "eval_loss": 0.004383066203445196, | |
| "eval_precision": 0.9107843137254902, | |
| "eval_recall": 0.8993223620522749, | |
| "eval_runtime": 17.5557, | |
| "eval_samples_per_second": 387.225, | |
| "eval_steps_per_second": 48.417, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 0.40936729311943054, | |
| "learning_rate": 4.468085106382979e-05, | |
| "loss": 0.0122, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9983575694624772, | |
| "eval_f1": 0.8947626040137053, | |
| "eval_loss": 0.005130653269588947, | |
| "eval_precision": 0.904950495049505, | |
| "eval_recall": 0.8848015488867377, | |
| "eval_runtime": 17.5062, | |
| "eval_samples_per_second": 388.319, | |
| "eval_steps_per_second": 48.554, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 0.011164379306137562, | |
| "learning_rate": 3.936170212765958e-05, | |
| "loss": 0.0032, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.998526356790444, | |
| "eval_f1": 0.9067837969741337, | |
| "eval_loss": 0.00494628818705678, | |
| "eval_precision": 0.9143700787401575, | |
| "eval_recall": 0.8993223620522749, | |
| "eval_runtime": 17.4826, | |
| "eval_samples_per_second": 388.844, | |
| "eval_steps_per_second": 48.62, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "grad_norm": 0.031265705823898315, | |
| "learning_rate": 3.4042553191489365e-05, | |
| "loss": 0.0017, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9986367177356531, | |
| "eval_f1": 0.9194956353055286, | |
| "eval_loss": 0.005987724754959345, | |
| "eval_precision": 0.9212827988338192, | |
| "eval_recall": 0.9177153920619555, | |
| "eval_runtime": 17.7527, | |
| "eval_samples_per_second": 382.928, | |
| "eval_steps_per_second": 47.88, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "grad_norm": 0.4819318950176239, | |
| "learning_rate": 2.8723404255319154e-05, | |
| "loss": 0.0011, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9985068813295248, | |
| "eval_f1": 0.9063893016344725, | |
| "eval_loss": 0.007134584244340658, | |
| "eval_precision": 0.9279918864097363, | |
| "eval_recall": 0.8857696030977735, | |
| "eval_runtime": 17.5767, | |
| "eval_samples_per_second": 386.763, | |
| "eval_steps_per_second": 48.36, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "grad_norm": 0.01076541654765606, | |
| "learning_rate": 2.340425531914894e-05, | |
| "loss": 0.0007, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9986302259153467, | |
| "eval_f1": 0.9160671462829737, | |
| "eval_loss": 0.006035692058503628, | |
| "eval_precision": 0.9077946768060836, | |
| "eval_recall": 0.9244917715392061, | |
| "eval_runtime": 17.4857, | |
| "eval_samples_per_second": 388.774, | |
| "eval_steps_per_second": 48.611, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 6.382978723404255, | |
| "grad_norm": 0.00033298603375442326, | |
| "learning_rate": 1.8085106382978724e-05, | |
| "loss": 0.0005, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9987535705011685, | |
| "eval_f1": 0.9233009708737864, | |
| "eval_loss": 0.005892382934689522, | |
| "eval_precision": 0.9259980525803311, | |
| "eval_recall": 0.920619554695063, | |
| "eval_runtime": 17.4716, | |
| "eval_samples_per_second": 389.089, | |
| "eval_steps_per_second": 48.65, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 7.446808510638298, | |
| "grad_norm": 0.017745647579431534, | |
| "learning_rate": 1.2765957446808511e-05, | |
| "loss": 0.0004, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9987276032199429, | |
| "eval_f1": 0.9241842610364683, | |
| "eval_loss": 0.006376999896019697, | |
| "eval_precision": 0.9162702188392008, | |
| "eval_recall": 0.9322362052274927, | |
| "eval_runtime": 17.4782, | |
| "eval_samples_per_second": 388.941, | |
| "eval_steps_per_second": 48.632, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 8.51063829787234, | |
| "grad_norm": 0.016069656237959862, | |
| "learning_rate": 7.446808510638298e-06, | |
| "loss": 0.0002, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9986367177356531, | |
| "eval_f1": 0.9230028873917228, | |
| "eval_loss": 0.006694914773106575, | |
| "eval_precision": 0.9177033492822967, | |
| "eval_recall": 0.9283639883833494, | |
| "eval_runtime": 17.7234, | |
| "eval_samples_per_second": 383.562, | |
| "eval_steps_per_second": 47.959, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 9.574468085106384, | |
| "grad_norm": 0.007091396953910589, | |
| "learning_rate": 2.1276595744680853e-06, | |
| "loss": 0.0001, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9987340950402492, | |
| "eval_f1": 0.9227076332213154, | |
| "eval_loss": 0.006876921746879816, | |
| "eval_precision": 0.9152380952380952, | |
| "eval_recall": 0.9303000968054211, | |
| "eval_runtime": 17.518, | |
| "eval_samples_per_second": 388.059, | |
| "eval_steps_per_second": 48.522, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 4700, | |
| "total_flos": 1.7002643858525088e+16, | |
| "train_loss": 0.0021406798980178987, | |
| "train_runtime": 2600.6679, | |
| "train_samples_per_second": 115.601, | |
| "train_steps_per_second": 1.807 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7002643858525088e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |