| { | |
| "best_metric": 0.6087076297937304, | |
| "best_model_checkpoint": "output/pretraining/vihealthbert-w_unsup-SynPD/lr3e-5_wr0.1_wd0.0/checkpoint-10000", | |
| "epoch": 1.7232465965879717, | |
| "eval_steps": 2000, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00017232465965879716, | |
| "grad_norm": 35.608882904052734, | |
| "learning_rate": 5.169739789763915e-09, | |
| "loss": 10.7382, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34464931931759435, | |
| "grad_norm": 4.884454250335693, | |
| "learning_rate": 1.033947957952783e-05, | |
| "loss": 5.8543, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34464931931759435, | |
| "eval_accuracy": 0.3950195480446223, | |
| "eval_loss": 3.896709680557251, | |
| "eval_runtime": 24.6709, | |
| "eval_samples_per_second": 396.175, | |
| "eval_steps_per_second": 24.766, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6892986386351887, | |
| "grad_norm": 5.278345108032227, | |
| "learning_rate": 2.067895915905566e-05, | |
| "loss": 3.4544, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6892986386351887, | |
| "eval_accuracy": 0.530626322082222, | |
| "eval_loss": 2.811899423599243, | |
| "eval_runtime": 24.6443, | |
| "eval_samples_per_second": 396.602, | |
| "eval_steps_per_second": 24.793, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.033947957952783, | |
| "grad_norm": 4.731122016906738, | |
| "learning_rate": 2.988684014015739e-05, | |
| "loss": 2.8312, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.033947957952783, | |
| "eval_accuracy": 0.5771132321637026, | |
| "eval_loss": 2.4039628505706787, | |
| "eval_runtime": 24.6266, | |
| "eval_samples_per_second": 396.889, | |
| "eval_steps_per_second": 24.811, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.3785972772703774, | |
| "grad_norm": 12.213573455810547, | |
| "learning_rate": 2.873800907576541e-05, | |
| "loss": 2.5914, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.3785972772703774, | |
| "eval_accuracy": 0.5349752225383133, | |
| "eval_loss": 2.6481692790985107, | |
| "eval_runtime": 24.6033, | |
| "eval_samples_per_second": 397.264, | |
| "eval_steps_per_second": 24.834, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.7232465965879717, | |
| "grad_norm": 5.172228813171387, | |
| "learning_rate": 2.758917801137343e-05, | |
| "loss": 2.5649, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.7232465965879717, | |
| "eval_accuracy": 0.6087076297937304, | |
| "eval_loss": 2.13348388671875, | |
| "eval_runtime": 24.6211, | |
| "eval_samples_per_second": 396.976, | |
| "eval_steps_per_second": 24.816, | |
| "step": 10000 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 58030, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 2000, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |