| { |
| "best_metric": 0.8256880733944955, |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-21/checkpoint-2635", |
| "epoch": 6.0, |
| "eval_steps": 500, |
| "global_step": 3162, |
| "is_hyper_param_search": true, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "learning_rate": 0.0007813674039938339, |
| "loss": 1.4569, |
| "step": 527 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8211009174311926, |
| "eval_loss": 1.8369688987731934, |
| "eval_runtime": 0.957, |
| "eval_samples_per_second": 911.136, |
| "eval_steps_per_second": 7.314, |
| "step": 527 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.0006251532075201045, |
| "loss": 0.6203, |
| "step": 1054 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.819954128440367, |
| "eval_loss": 2.187230110168457, |
| "eval_runtime": 0.833, |
| "eval_samples_per_second": 1046.865, |
| "eval_steps_per_second": 8.404, |
| "step": 1054 |
| }, |
| { |
| "epoch": 3.0, |
| "learning_rate": 0.0004689390110463753, |
| "loss": 0.4011, |
| "step": 1581 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8222477064220184, |
| "eval_loss": 2.2777745723724365, |
| "eval_runtime": 0.8303, |
| "eval_samples_per_second": 1050.221, |
| "eval_steps_per_second": 8.431, |
| "step": 1581 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.000312724814572646, |
| "loss": 0.2824, |
| "step": 2108 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8176605504587156, |
| "eval_loss": 2.049936294555664, |
| "eval_runtime": 0.9199, |
| "eval_samples_per_second": 947.944, |
| "eval_steps_per_second": 7.61, |
| "step": 2108 |
| }, |
| { |
| "epoch": 5.0, |
| "learning_rate": 0.00015651061809891667, |
| "loss": 0.2061, |
| "step": 2635 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8256880733944955, |
| "eval_loss": 2.054544687271118, |
| "eval_runtime": 0.8429, |
| "eval_samples_per_second": 1034.572, |
| "eval_steps_per_second": 8.305, |
| "step": 2635 |
| }, |
| { |
| "epoch": 6.0, |
| "learning_rate": 2.9642162518734216e-07, |
| "loss": 0.1638, |
| "step": 3162 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8222477064220184, |
| "eval_loss": 2.0563035011291504, |
| "eval_runtime": 0.8349, |
| "eval_samples_per_second": 1044.492, |
| "eval_steps_per_second": 8.385, |
| "step": 3162 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 3162, |
| "num_train_epochs": 6, |
| "save_steps": 500, |
| "total_flos": 48509902416420.0, |
| "trial_name": null, |
| "trial_params": { |
| "alpha": 0.12545237557795993, |
| "learning_rate": 0.0009372851788423759, |
| "num_train_epochs": 6, |
| "temperature": 30 |
| } |
| } |
|
|
