| { | |
| "best_metric": 0.6935691237449646, | |
| "best_model_checkpoint": "add_BERT_no_pretrain_sst2/checkpoint-527", | |
| "epoch": 6.0, | |
| "global_step": 3162, | |
| "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": 0.6983, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6935691237449646, | |
| "eval_runtime": 1.6343, | |
| "eval_samples_per_second": 533.567, | |
| "eval_steps_per_second": 4.283, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.6895, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7089212536811829, | |
| "eval_runtime": 1.6328, | |
| "eval_samples_per_second": 534.054, | |
| "eval_steps_per_second": 4.287, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.6881, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6993499994277954, | |
| "eval_runtime": 1.6457, | |
| "eval_samples_per_second": 529.872, | |
| "eval_steps_per_second": 4.254, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.6875, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6993725299835205, | |
| "eval_runtime": 1.6446, | |
| "eval_samples_per_second": 530.225, | |
| "eval_steps_per_second": 4.256, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6874, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.694109320640564, | |
| "eval_runtime": 1.638, | |
| "eval_samples_per_second": 532.346, | |
| "eval_steps_per_second": 4.273, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.687, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7002017498016357, | |
| "eval_runtime": 1.6315, | |
| "eval_samples_per_second": 534.473, | |
| "eval_steps_per_second": 4.29, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 3162, | |
| "total_flos": 5.771431023476736e+16, | |
| "train_loss": 0.6896291946324271, | |
| "train_runtime": 2408.1345, | |
| "train_samples_per_second": 1398.365, | |
| "train_steps_per_second": 10.942 | |
| } | |
| ], | |
| "max_steps": 26350, | |
| "num_train_epochs": 50, | |
| "total_flos": 5.771431023476736e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |