| { | |
| "best_metric": 0.7912844036697247, | |
| "best_model_checkpoint": "tiny-bert-sst2/run-2/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": 1.0882924795150757, | |
| "learning_rate": 4.232847990599429e-05, | |
| "loss": 1.5987, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18975332068311196, | |
| "eval_accuracy": 0.6811926605504587, | |
| "eval_loss": 1.443600058555603, | |
| "eval_runtime": 2.5162, | |
| "eval_samples_per_second": 346.555, | |
| "eval_steps_per_second": 2.782, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3795066413662239, | |
| "grad_norm": 2.405395984649658, | |
| "learning_rate": 4.145818368618804e-05, | |
| "loss": 1.3874, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3795066413662239, | |
| "eval_accuracy": 0.7545871559633027, | |
| "eval_loss": 1.168924331665039, | |
| "eval_runtime": 2.4154, | |
| "eval_samples_per_second": 361.016, | |
| "eval_steps_per_second": 2.898, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5692599620493358, | |
| "grad_norm": 3.982295274734497, | |
| "learning_rate": 3.9736321558974306e-05, | |
| "loss": 1.1384, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5692599620493358, | |
| "eval_accuracy": 0.7855504587155964, | |
| "eval_loss": 1.0294885635375977, | |
| "eval_runtime": 2.4987, | |
| "eval_samples_per_second": 348.984, | |
| "eval_steps_per_second": 2.801, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7590132827324478, | |
| "grad_norm": 6.110784530639648, | |
| "learning_rate": 3.723532957026996e-05, | |
| "loss": 1.0003, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7590132827324478, | |
| "eval_accuracy": 0.7912844036697247, | |
| "eval_loss": 0.9760427474975586, | |
| "eval_runtime": 2.4446, | |
| "eval_samples_per_second": 356.7, | |
| "eval_steps_per_second": 2.863, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9487666034155597, | |
| "grad_norm": 3.9430387020111084, | |
| "learning_rate": 3.4060420540351555e-05, | |
| "loss": 0.9476, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9487666034155597, | |
| "eval_accuracy": 0.7912844036697247, | |
| "eval_loss": 0.9497756361961365, | |
| "eval_runtime": 2.4069, | |
| "eval_samples_per_second": 362.295, | |
| "eval_steps_per_second": 2.908, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1581, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 0.7521996188553136, | |
| "learning_rate": 4.243157651358427e-05, | |
| "num_train_epochs": 3, | |
| "temperature": 6, | |
| "warmup_ratio": 0.03249388147757522 | |
| } | |
| } | |