| { | |
| "best_metric": 0.8073394495412844, | |
| "best_model_checkpoint": "tiny-bert-sst2/run-4/checkpoint-500", | |
| "epoch": 0.9487666034155597, | |
| "eval_steps": 100, | |
| "global_step": 500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18975332068311196, | |
| "grad_norm": 22.14986228942871, | |
| "learning_rate": 0.00020336877465804552, | |
| "loss": 3.1045, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18975332068311196, | |
| "eval_accuracy": 0.7878440366972477, | |
| "eval_loss": 2.06337571144104, | |
| "eval_runtime": 2.5167, | |
| "eval_samples_per_second": 346.481, | |
| "eval_steps_per_second": 2.781, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3795066413662239, | |
| "grad_norm": 14.039098739624023, | |
| "learning_rate": 0.0001915586281128006, | |
| "loss": 1.8121, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3795066413662239, | |
| "eval_accuracy": 0.7935779816513762, | |
| "eval_loss": 1.8058583736419678, | |
| "eval_runtime": 2.4835, | |
| "eval_samples_per_second": 351.124, | |
| "eval_steps_per_second": 2.819, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5692599620493358, | |
| "grad_norm": 12.429423332214355, | |
| "learning_rate": 0.0001715981466068485, | |
| "loss": 1.4044, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5692599620493358, | |
| "eval_accuracy": 0.801605504587156, | |
| "eval_loss": 1.6292372941970825, | |
| "eval_runtime": 2.4416, | |
| "eval_samples_per_second": 357.144, | |
| "eval_steps_per_second": 2.867, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7590132827324478, | |
| "grad_norm": 13.686047554016113, | |
| "learning_rate": 0.00014532636714010877, | |
| "loss": 1.196, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7590132827324478, | |
| "eval_accuracy": 0.819954128440367, | |
| "eval_loss": 1.5613027811050415, | |
| "eval_runtime": 2.4718, | |
| "eval_samples_per_second": 352.778, | |
| "eval_steps_per_second": 2.832, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9487666034155597, | |
| "grad_norm": 16.022205352783203, | |
| "learning_rate": 0.00011516381120544472, | |
| "loss": 1.1018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9487666034155597, | |
| "eval_accuracy": 0.8073394495412844, | |
| "eval_loss": 1.5502208471298218, | |
| "eval_runtime": 2.4506, | |
| "eval_samples_per_second": 355.838, | |
| "eval_steps_per_second": 2.856, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1054, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 7672788856320.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.28231841434241045, | |
| "learning_rate": 0.0002061935616423689, | |
| "num_train_epochs": 2, | |
| "temperature": 15, | |
| "warmup_ratio": 0.020976865960675092 | |
| } | |
| } | |