| { | |
| "best_global_step": 6136, | |
| "best_metric": 0.7974135875701904, | |
| "best_model_checkpoint": "bert_base_km_10_v1_mnli/checkpoint-6136", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 13806, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.0922600030899048, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.9841, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5852266938359654, | |
| "eval_loss": 0.8973079919815063, | |
| "eval_runtime": 6.2434, | |
| "eval_samples_per_second": 1572.062, | |
| "eval_steps_per_second": 6.247, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.525976300239563, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.8612, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6234335201222618, | |
| "eval_loss": 0.8426495790481567, | |
| "eval_runtime": 6.1557, | |
| "eval_samples_per_second": 1594.47, | |
| "eval_steps_per_second": 6.336, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.8445866107940674, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.7666, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6523688232297504, | |
| "eval_loss": 0.7988670468330383, | |
| "eval_runtime": 6.2027, | |
| "eval_samples_per_second": 1582.374, | |
| "eval_steps_per_second": 6.288, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.149568557739258, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6813, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6523688232297504, | |
| "eval_loss": 0.7974135875701904, | |
| "eval_runtime": 6.155, | |
| "eval_samples_per_second": 1594.635, | |
| "eval_steps_per_second": 6.336, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.793645143508911, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5958, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6611309220580743, | |
| "eval_loss": 0.8223918676376343, | |
| "eval_runtime": 6.177, | |
| "eval_samples_per_second": 1588.968, | |
| "eval_steps_per_second": 6.314, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.275442123413086, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5054, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6592969943963322, | |
| "eval_loss": 0.8842700123786926, | |
| "eval_runtime": 6.1769, | |
| "eval_samples_per_second": 1588.979, | |
| "eval_steps_per_second": 6.314, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.047013998031616, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.4197, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6620478858889455, | |
| "eval_loss": 0.9591888189315796, | |
| "eval_runtime": 6.1732, | |
| "eval_samples_per_second": 1589.924, | |
| "eval_steps_per_second": 6.318, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.33532977104187, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.3427, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6568517575140091, | |
| "eval_loss": 1.1082422733306885, | |
| "eval_runtime": 6.1745, | |
| "eval_samples_per_second": 1589.599, | |
| "eval_steps_per_second": 6.316, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.84657621383667, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.279, | |
| "step": 13806 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6505348955680081, | |
| "eval_loss": 1.2692910432815552, | |
| "eval_runtime": 6.1573, | |
| "eval_samples_per_second": 1594.048, | |
| "eval_steps_per_second": 6.334, | |
| "step": 13806 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 13806, | |
| "total_flos": 4.6496324415879475e+17, | |
| "train_loss": 0.6039889038740233, | |
| "train_runtime": 5732.2536, | |
| "train_samples_per_second": 3425.372, | |
| "train_steps_per_second": 13.38 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 76700, | |
| "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": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.6496324415879475e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |