| { | |
| "best_metric": 0.8747714808043876, | |
| "best_model_checkpoint": "berttiny-hate_speech18-bothpretrained\\run-27\\checkpoint-2874", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 2874, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.987812746965635e-05, | |
| "loss": 0.4822, | |
| "step": 958 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8628884826325411, | |
| "eval_loss": 0.47384002804756165, | |
| "eval_runtime": 10.814, | |
| "eval_samples_per_second": 202.33, | |
| "eval_steps_per_second": 1.665, | |
| "step": 958 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.4939063734828174e-05, | |
| "loss": 0.4605, | |
| "step": 1916 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.873400365630713, | |
| "eval_loss": 0.471261590719223, | |
| "eval_runtime": 10.7994, | |
| "eval_samples_per_second": 202.604, | |
| "eval_steps_per_second": 1.667, | |
| "step": 1916 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.4536, | |
| "step": 2874 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8747714808043876, | |
| "eval_loss": 0.4703117609024048, | |
| "eval_runtime": 10.6751, | |
| "eval_samples_per_second": 204.963, | |
| "eval_steps_per_second": 1.686, | |
| "step": 2874 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2874, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2950888285536.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 4.4817191204484525e-05, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 8, | |
| "seed": 32 | |
| } | |
| } | |