| { | |
| "best_metric": 0.6678495407104492, | |
| "best_model_checkpoint": "hBERTv1_new_pretrain_48_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": 0.6818, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5815485996705108, | |
| "eval_loss": 0.6781542897224426, | |
| "eval_runtime": 9.6934, | |
| "eval_samples_per_second": 563.577, | |
| "eval_steps_per_second": 4.436, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.6686, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5837451949478308, | |
| "eval_loss": 0.6678495407104492, | |
| "eval_runtime": 9.7146, | |
| "eval_samples_per_second": 562.347, | |
| "eval_steps_per_second": 4.426, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.6472, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5846604429800476, | |
| "eval_loss": 0.6737925410270691, | |
| "eval_runtime": 9.726, | |
| "eval_samples_per_second": 561.693, | |
| "eval_steps_per_second": 4.421, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.6311, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5802672524254073, | |
| "eval_loss": 0.6779181361198425, | |
| "eval_runtime": 9.6997, | |
| "eval_samples_per_second": 563.212, | |
| "eval_steps_per_second": 4.433, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6142, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5850265421929343, | |
| "eval_loss": 0.6802344918251038, | |
| "eval_runtime": 9.6984, | |
| "eval_samples_per_second": 563.289, | |
| "eval_steps_per_second": 4.434, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.5969, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5861248398315944, | |
| "eval_loss": 0.7075582146644592, | |
| "eval_runtime": 9.6998, | |
| "eval_samples_per_second": 563.206, | |
| "eval_steps_per_second": 4.433, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.5814, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5793520043931906, | |
| "eval_loss": 0.7671993374824524, | |
| "eval_runtime": 9.7166, | |
| "eval_samples_per_second": 562.234, | |
| "eval_steps_per_second": 4.425, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 5733, | |
| "total_flos": 1.085847006787666e+17, | |
| "train_loss": 0.6315902105784221, | |
| "train_runtime": 3878.247, | |
| "train_samples_per_second": 1350.391, | |
| "train_steps_per_second": 10.559 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.085847006787666e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |