| { | |
| "best_metric": 0.6984848484848485, | |
| "best_model_checkpoint": "tinybert-toxigen-bothpretrained\\run-0\\checkpoint-165", | |
| "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": 2.1639231036686423e-05, | |
| "loss": 0.3688, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6378787878787879, | |
| "eval_loss": 0.3670942187309265, | |
| "eval_runtime": 3.2946, | |
| "eval_samples_per_second": 600.98, | |
| "eval_steps_per_second": 4.856, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.6229423277514814e-05, | |
| "loss": 0.3633, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6782828282828283, | |
| "eval_loss": 0.3586779534816742, | |
| "eval_runtime": 3.2782, | |
| "eval_samples_per_second": 603.989, | |
| "eval_steps_per_second": 4.881, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.0819615518343211e-05, | |
| "loss": 0.3544, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6984848484848485, | |
| "eval_loss": 0.3513254225254059, | |
| "eval_runtime": 3.3521, | |
| "eval_samples_per_second": 590.677, | |
| "eval_steps_per_second": 4.773, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.409807759171606e-06, | |
| "loss": 0.3484, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.696969696969697, | |
| "eval_loss": 0.3481346070766449, | |
| "eval_runtime": 3.354, | |
| "eval_samples_per_second": 590.334, | |
| "eval_steps_per_second": 4.77, | |
| "step": 220 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 275, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 3368572315560.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.5339855376932713, | |
| "learning_rate": 2.7049038795858027e-05, | |
| "num_train_epochs": 5, | |
| "temperature": 24 | |
| } | |
| } | |