| { | |
| "best_metric": 0.6869744658470154, | |
| "best_model_checkpoint": "hBERTv2_new_pretrain_qnli/checkpoint-1638", | |
| "epoch": 7.0, | |
| "global_step": 5733, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 0.6921, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.505033864177192, | |
| "eval_loss": 0.6903212070465088, | |
| "eval_runtime": 9.3321, | |
| "eval_samples_per_second": 585.401, | |
| "eval_steps_per_second": 4.608, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.6906, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5502471169686985, | |
| "eval_loss": 0.6869744658470154, | |
| "eval_runtime": 9.3282, | |
| "eval_samples_per_second": 585.645, | |
| "eval_steps_per_second": 4.61, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.6889, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5231557752150833, | |
| "eval_loss": 0.692358672618866, | |
| "eval_runtime": 9.3343, | |
| "eval_samples_per_second": 585.259, | |
| "eval_steps_per_second": 4.607, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.6932, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5260845689181768, | |
| "eval_loss": 0.6916021704673767, | |
| "eval_runtime": 9.3343, | |
| "eval_samples_per_second": 585.262, | |
| "eval_steps_per_second": 4.607, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6927, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5313930075050338, | |
| "eval_loss": 0.692005455493927, | |
| "eval_runtime": 9.3373, | |
| "eval_samples_per_second": 585.071, | |
| "eval_steps_per_second": 4.605, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.6918, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5297455610470437, | |
| "eval_loss": 0.689666211605072, | |
| "eval_runtime": 9.3306, | |
| "eval_samples_per_second": 585.493, | |
| "eval_steps_per_second": 4.608, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.6927, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5086948563060589, | |
| "eval_loss": 0.692981481552124, | |
| "eval_runtime": 9.3271, | |
| "eval_samples_per_second": 585.713, | |
| "eval_steps_per_second": 4.61, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 5733, | |
| "total_flos": 1.0729602688011469e+17, | |
| "train_loss": 0.69172189108015, | |
| "train_runtime": 3811.1135, | |
| "train_samples_per_second": 1374.178, | |
| "train_steps_per_second": 10.745 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.0729602688011469e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |