| { | |
| "best_metric": 0.7055869698524475, | |
| "best_model_checkpoint": "bert_base_lda_100_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": 1.7309693098068237, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.9588, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6248599083036169, | |
| "eval_loss": 0.8420299887657166, | |
| "eval_runtime": 6.3265, | |
| "eval_samples_per_second": 1551.4, | |
| "eval_steps_per_second": 6.165, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.024912118911743, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7857, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.680794701986755, | |
| "eval_loss": 0.7450552582740784, | |
| "eval_runtime": 6.312, | |
| "eval_samples_per_second": 1554.976, | |
| "eval_steps_per_second": 6.179, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.8295753002166748, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6825, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6976057055527254, | |
| "eval_loss": 0.7161675691604614, | |
| "eval_runtime": 6.3639, | |
| "eval_samples_per_second": 1542.29, | |
| "eval_steps_per_second": 6.128, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.0847725868225098, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5973, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7112582781456953, | |
| "eval_loss": 0.7055869698524475, | |
| "eval_runtime": 6.3116, | |
| "eval_samples_per_second": 1555.078, | |
| "eval_steps_per_second": 6.179, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.3416082859039307, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5208, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7144167091186959, | |
| "eval_loss": 0.7459524869918823, | |
| "eval_runtime": 6.2457, | |
| "eval_samples_per_second": 1571.471, | |
| "eval_steps_per_second": 6.244, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.3869707584381104, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.4464, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7077941925624045, | |
| "eval_loss": 0.790677011013031, | |
| "eval_runtime": 6.3035, | |
| "eval_samples_per_second": 1557.08, | |
| "eval_steps_per_second": 6.187, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.6693198680877686, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.3775, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7171676006113092, | |
| "eval_loss": 0.8361655473709106, | |
| "eval_runtime": 6.3703, | |
| "eval_samples_per_second": 1540.741, | |
| "eval_steps_per_second": 6.122, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.255511999130249, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.316, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7101375445746306, | |
| "eval_loss": 0.9462762475013733, | |
| "eval_runtime": 6.4, | |
| "eval_samples_per_second": 1533.587, | |
| "eval_steps_per_second": 6.094, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.7821061611175537, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.2617, | |
| "step": 13806 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7110545084055018, | |
| "eval_loss": 1.009425401687622, | |
| "eval_runtime": 6.5216, | |
| "eval_samples_per_second": 1505.005, | |
| "eval_steps_per_second": 5.98, | |
| "step": 13806 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 13806, | |
| "total_flos": 4.6496324415879475e+17, | |
| "train_loss": 0.5496212416207741, | |
| "train_runtime": 5787.733, | |
| "train_samples_per_second": 3392.537, | |
| "train_steps_per_second": 13.252 | |
| } | |
| ], | |
| "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 | |
| } | |