| { | |
| "best_metric": 0.7855504587155964, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-2635", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2635, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.747544288635254, | |
| "learning_rate": 8.421921303617797e-06, | |
| "loss": 1.1824, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7259174311926605, | |
| "eval_loss": 1.0221459865570068, | |
| "eval_runtime": 2.6361, | |
| "eval_samples_per_second": 330.793, | |
| "eval_steps_per_second": 2.655, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 10.284089088439941, | |
| "learning_rate": 6.737537042894237e-06, | |
| "loss": 0.974, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7775229357798165, | |
| "eval_loss": 0.8609867095947266, | |
| "eval_runtime": 2.6102, | |
| "eval_samples_per_second": 334.074, | |
| "eval_steps_per_second": 2.682, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 14.767024993896484, | |
| "learning_rate": 5.053152782170678e-06, | |
| "loss": 0.8451, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7809633027522935, | |
| "eval_loss": 0.8180754780769348, | |
| "eval_runtime": 2.6115, | |
| "eval_samples_per_second": 333.903, | |
| "eval_steps_per_second": 2.68, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.393446922302246, | |
| "learning_rate": 3.3687685214471185e-06, | |
| "loss": 0.7901, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7844036697247706, | |
| "eval_loss": 0.8021524548530579, | |
| "eval_runtime": 2.5985, | |
| "eval_samples_per_second": 335.573, | |
| "eval_steps_per_second": 2.694, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 8.2117280960083, | |
| "learning_rate": 1.6843842607235593e-06, | |
| "loss": 0.7574, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7855504587155964, | |
| "eval_loss": 0.7915852665901184, | |
| "eval_runtime": 2.6137, | |
| "eval_samples_per_second": 333.62, | |
| "eval_steps_per_second": 2.678, | |
| "step": 2635 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3162, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 40447180739880.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.8183970243354497, | |
| "learning_rate": 1.0106305564341356e-05, | |
| "num_train_epochs": 6, | |
| "temperature": 4 | |
| } | |
| } | |