| { | |
| "model_checkpoint": "microsoft/deberta-v3-large", | |
| "final_metrics": { | |
| "validation": { | |
| "eval_loss": 0.0013762748567387462, | |
| "eval_accuracy": 0.9998051441932969, | |
| "eval_precision": 0.9996682149966821, | |
| "eval_recall": 1.0, | |
| "eval_f1": 0.9998340799734527, | |
| "eval_runtime": 34.6469, | |
| "eval_samples_per_second": 148.123, | |
| "eval_steps_per_second": 1.183, | |
| "epoch": 0.9968847352024922 | |
| }, | |
| "test": { | |
| "eval_loss": 0.0015322713879868388, | |
| "eval_accuracy": 0.9994154325798909, | |
| "eval_precision": 0.9993362097577165, | |
| "eval_recall": 0.999667994687915, | |
| "eval_f1": 0.9995020746887967, | |
| "eval_runtime": 34.9714, | |
| "eval_samples_per_second": 146.748, | |
| "eval_steps_per_second": 1.172, | |
| "epoch": 0.9968847352024922 | |
| } | |
| }, | |
| "training_config": { | |
| "max_steps": 640, | |
| "batch_size": 64, | |
| "gradient_accumulation_steps": 2, | |
| "learning_rate": 3e-05, | |
| "max_length": 512, | |
| "gradient_checkpointing": "non-reentrant" | |
| }, | |
| "dataset_stats": { | |
| "total_examples": 51319, | |
| "train_size": 41055, | |
| "val_size": 5132, | |
| "test_size": 5132, | |
| "fake_samples": 30123, | |
| "real_samples": 21196 | |
| }, | |
| "runtime_hours": 0.4279349425766203 | |
| } |