| { | |
| "best_metric": 0.18682542443275452, | |
| "best_model_checkpoint": "ontonotes_roberta-large-v2/checkpoint-360460", | |
| "epoch": 14.0, | |
| "eval_steps": 500, | |
| "global_step": 504644, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.8596975803375244, | |
| "learning_rate": 1.960019974477057e-05, | |
| "loss": 0.3672, | |
| "step": 36046 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.23480436205863953, | |
| "eval_runtime": 162.404, | |
| "eval_samples_per_second": 164.497, | |
| "eval_steps_per_second": 20.566, | |
| "step": 36046 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.4112672805786133, | |
| "learning_rate": 1.9200332907950954e-05, | |
| "loss": 0.2279, | |
| "step": 72092 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.21355144679546356, | |
| "eval_runtime": 158.0799, | |
| "eval_samples_per_second": 168.997, | |
| "eval_steps_per_second": 21.129, | |
| "step": 72092 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.2005605697631836, | |
| "learning_rate": 1.880049936192643e-05, | |
| "loss": 0.2002, | |
| "step": 108138 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.19775843620300293, | |
| "eval_runtime": 159.9354, | |
| "eval_samples_per_second": 167.036, | |
| "eval_steps_per_second": 20.883, | |
| "step": 108138 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.314330130815506, | |
| "learning_rate": 1.840064362203851e-05, | |
| "loss": 0.1831, | |
| "step": 144184 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.19093795120716095, | |
| "eval_runtime": 163.8453, | |
| "eval_samples_per_second": 163.05, | |
| "eval_steps_per_second": 20.385, | |
| "step": 144184 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.5714001655578613, | |
| "learning_rate": 1.8000776785218888e-05, | |
| "loss": 0.1709, | |
| "step": 180230 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.19262374937534332, | |
| "eval_runtime": 157.711, | |
| "eval_samples_per_second": 169.392, | |
| "eval_steps_per_second": 21.178, | |
| "step": 180230 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 11.868833541870117, | |
| "learning_rate": 1.7600921045330967e-05, | |
| "loss": 0.1611, | |
| "step": 216276 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.19287440180778503, | |
| "eval_runtime": 158.8625, | |
| "eval_samples_per_second": 168.164, | |
| "eval_steps_per_second": 21.024, | |
| "step": 216276 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.1779515743255615, | |
| "learning_rate": 1.7201076402374746e-05, | |
| "loss": 0.1529, | |
| "step": 252322 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.1881980001926422, | |
| "eval_runtime": 158.7846, | |
| "eval_samples_per_second": 168.247, | |
| "eval_steps_per_second": 21.035, | |
| "step": 252322 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.401430606842041, | |
| "learning_rate": 1.680118737169173e-05, | |
| "loss": 0.1453, | |
| "step": 288368 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.19533155858516693, | |
| "eval_runtime": 159.1977, | |
| "eval_samples_per_second": 167.81, | |
| "eval_steps_per_second": 20.98, | |
| "step": 288368 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.25393956899642944, | |
| "learning_rate": 1.6401342728735506e-05, | |
| "loss": 0.1392, | |
| "step": 324414 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.19019867479801178, | |
| "eval_runtime": 159.8332, | |
| "eval_samples_per_second": 167.143, | |
| "eval_steps_per_second": 20.897, | |
| "step": 324414 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.6049373149871826, | |
| "learning_rate": 1.6001498085779284e-05, | |
| "loss": 0.1328, | |
| "step": 360460 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.18682542443275452, | |
| "eval_runtime": 163.9057, | |
| "eval_samples_per_second": 162.99, | |
| "eval_steps_per_second": 20.378, | |
| "step": 360460 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.0430878400802612, | |
| "learning_rate": 1.5601631248959664e-05, | |
| "loss": 0.1273, | |
| "step": 396506 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.1927616000175476, | |
| "eval_runtime": 162.6725, | |
| "eval_samples_per_second": 164.226, | |
| "eval_steps_per_second": 20.532, | |
| "step": 396506 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.0110228061676025, | |
| "learning_rate": 1.5201797702935139e-05, | |
| "loss": 0.1225, | |
| "step": 432552 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.20162169635295868, | |
| "eval_runtime": 158.6594, | |
| "eval_samples_per_second": 168.38, | |
| "eval_steps_per_second": 21.051, | |
| "step": 432552 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.405910462141037, | |
| "learning_rate": 1.4801930866115519e-05, | |
| "loss": 0.1172, | |
| "step": 468598 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.20160554349422455, | |
| "eval_runtime": 158.4325, | |
| "eval_samples_per_second": 168.621, | |
| "eval_steps_per_second": 21.082, | |
| "step": 468598 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 9.577973365783691, | |
| "learning_rate": 1.44020751262276e-05, | |
| "loss": 0.1126, | |
| "step": 504644 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.20640429854393005, | |
| "eval_runtime": 158.7595, | |
| "eval_samples_per_second": 168.273, | |
| "eval_steps_per_second": 21.038, | |
| "step": 504644 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1802300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.8514301904299226e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |