| { | |
| "best_metric": 0.7626262626262627, | |
| "best_model_checkpoint": "tinybert-toxigen-bothpretrained\\run-5\\checkpoint-220", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 220, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5.445560435088227e-05, | |
| "loss": 0.265, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6898989898989899, | |
| "eval_loss": 0.25976529717445374, | |
| "eval_runtime": 3.3203, | |
| "eval_samples_per_second": 596.328, | |
| "eval_steps_per_second": 4.819, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.3564483480705814e-05, | |
| "loss": 0.2555, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7368686868686869, | |
| "eval_loss": 0.2521827816963196, | |
| "eval_runtime": 3.3412, | |
| "eval_samples_per_second": 592.604, | |
| "eval_steps_per_second": 4.789, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.267336261052936e-05, | |
| "loss": 0.2496, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.73989898989899, | |
| "eval_loss": 0.25000467896461487, | |
| "eval_runtime": 3.3777, | |
| "eval_samples_per_second": 586.19, | |
| "eval_steps_per_second": 4.737, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.1782241740352907e-05, | |
| "loss": 0.247, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7626262626262627, | |
| "eval_loss": 0.24974669516086578, | |
| "eval_runtime": 3.3894, | |
| "eval_samples_per_second": 584.175, | |
| "eval_steps_per_second": 4.721, | |
| "step": 220 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 330, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 3368572315560.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.38098176211619805, | |
| "learning_rate": 6.534672522105872e-05, | |
| "num_train_epochs": 6, | |
| "temperature": 20 | |
| } | |
| } | |