| { | |
| "best_metric": 0.8761425959780622, | |
| "best_model_checkpoint": "berttiny-hate_speech18-bothpretrained\\run-1\\checkpoint-7660", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 9575, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.623415476779745e-05, | |
| "loss": 0.4826, | |
| "step": 1915 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8574040219378428, | |
| "eval_loss": 0.484986811876297, | |
| "eval_runtime": 10.9318, | |
| "eval_samples_per_second": 200.15, | |
| "eval_steps_per_second": 1.647, | |
| "step": 1915 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.9675616075848086e-05, | |
| "loss": 0.4636, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8638025594149908, | |
| "eval_loss": 0.4823402464389801, | |
| "eval_runtime": 10.7788, | |
| "eval_samples_per_second": 202.991, | |
| "eval_steps_per_second": 1.67, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.3117077383898726e-05, | |
| "loss": 0.4549, | |
| "step": 5745 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8743144424131627, | |
| "eval_loss": 0.4747394323348999, | |
| "eval_runtime": 10.8039, | |
| "eval_samples_per_second": 202.519, | |
| "eval_steps_per_second": 1.666, | |
| "step": 5745 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6.558538691949363e-06, | |
| "loss": 0.4488, | |
| "step": 7660 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8761425959780622, | |
| "eval_loss": 0.47820988297462463, | |
| "eval_runtime": 10.9143, | |
| "eval_samples_per_second": 200.471, | |
| "eval_steps_per_second": 1.649, | |
| "step": 7660 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.4448, | |
| "step": 9575 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8747714808043876, | |
| "eval_loss": 0.4775071442127228, | |
| "eval_runtime": 10.6763, | |
| "eval_samples_per_second": 204.939, | |
| "eval_steps_per_second": 1.686, | |
| "step": 9575 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9575, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 3867947686656.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 3.279269345974681e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 20 | |
| } | |
| } | |