| { | |
| "best_metric": 0.8648728562980484, | |
| "best_model_checkpoint": "./models/entities/test_v4/checkpoint-3500", | |
| "epoch": 9.979267449896337, | |
| "eval_steps": 500, | |
| "global_step": 3610, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.38217000691085, | |
| "grad_norm": 0.9774827361106873, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5764, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.38217000691085, | |
| "eval_accuracy": 0.9593812772012053, | |
| "eval_f1": 0.7594292979234355, | |
| "eval_loss": 0.1411387324333191, | |
| "eval_precision": 0.708673872923807, | |
| "eval_recall": 0.8180158247108947, | |
| "eval_runtime": 199.3049, | |
| "eval_samples_per_second": 24.886, | |
| "eval_steps_per_second": 24.886, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.7643400138217, | |
| "grad_norm": 0.7694200873374939, | |
| "learning_rate": 0.00016784565916398716, | |
| "loss": 0.09, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.7643400138217, | |
| "eval_accuracy": 0.9687447689989956, | |
| "eval_f1": 0.8259538274605103, | |
| "eval_loss": 0.11496158689260483, | |
| "eval_precision": 0.792852477372399, | |
| "eval_recall": 0.8619395414891459, | |
| "eval_runtime": 48.3856, | |
| "eval_samples_per_second": 102.51, | |
| "eval_steps_per_second": 102.51, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.14651002073255, | |
| "grad_norm": 0.625320315361023, | |
| "learning_rate": 0.0001356913183279743, | |
| "loss": 0.0463, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.14651002073255, | |
| "eval_accuracy": 0.9703716103113492, | |
| "eval_f1": 0.8380395025603511, | |
| "eval_loss": 0.13039253652095795, | |
| "eval_precision": 0.8069878839109609, | |
| "eval_recall": 0.8715763846622033, | |
| "eval_runtime": 46.9044, | |
| "eval_samples_per_second": 105.747, | |
| "eval_steps_per_second": 105.747, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.5286800276434, | |
| "grad_norm": 0.4629702866077423, | |
| "learning_rate": 0.00010353697749196143, | |
| "loss": 0.023, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.5286800276434, | |
| "eval_accuracy": 0.9716950535654503, | |
| "eval_f1": 0.8417248620403379, | |
| "eval_loss": 0.1349513679742813, | |
| "eval_precision": 0.8115641774178359, | |
| "eval_recall": 0.8742138364779874, | |
| "eval_runtime": 47.4889, | |
| "eval_samples_per_second": 104.446, | |
| "eval_steps_per_second": 104.446, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.91085003455425, | |
| "grad_norm": 0.1776706427335739, | |
| "learning_rate": 7.138263665594856e-05, | |
| "loss": 0.0147, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.91085003455425, | |
| "eval_accuracy": 0.9736305239370606, | |
| "eval_f1": 0.8591340450771056, | |
| "eval_loss": 0.14083334803581238, | |
| "eval_precision": 0.8377674956622325, | |
| "eval_recall": 0.8816189896530736, | |
| "eval_runtime": 47.4474, | |
| "eval_samples_per_second": 104.537, | |
| "eval_steps_per_second": 104.537, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.2930200414651, | |
| "grad_norm": 0.7263190746307373, | |
| "learning_rate": 3.92282958199357e-05, | |
| "loss": 0.0078, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.2930200414651, | |
| "eval_accuracy": 0.9740594660194175, | |
| "eval_f1": 0.8621834580504485, | |
| "eval_loss": 0.14879639446735382, | |
| "eval_precision": 0.8428446855924814, | |
| "eval_recall": 0.8824305132886995, | |
| "eval_runtime": 49.4794, | |
| "eval_samples_per_second": 100.244, | |
| "eval_steps_per_second": 100.244, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.675190048375951, | |
| "grad_norm": 0.21800673007965088, | |
| "learning_rate": 7.07395498392283e-06, | |
| "loss": 0.0045, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.675190048375951, | |
| "eval_accuracy": 0.9743680950786743, | |
| "eval_f1": 0.8648728562980484, | |
| "eval_loss": 0.16495844721794128, | |
| "eval_precision": 0.8410005750431282, | |
| "eval_recall": 0.8901399878271454, | |
| "eval_runtime": 46.0436, | |
| "eval_samples_per_second": 107.724, | |
| "eval_steps_per_second": 107.724, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.979267449896337, | |
| "step": 3610, | |
| "total_flos": 6112468258365600.0, | |
| "train_loss": 0.10573644003214268, | |
| "train_runtime": 1226.1602, | |
| "train_samples_per_second": 188.784, | |
| "train_steps_per_second": 2.944 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3610, | |
| "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": 6112468258365600.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |