| { | |
| "best_metric": 0.662101149559021, | |
| "best_model_checkpoint": "distilbert_add_GLUE_Experiment_qnli_384/checkpoint-820", | |
| "epoch": 7.0, | |
| "global_step": 2870, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6894, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5932637744828848, | |
| "eval_loss": 0.6660351753234863, | |
| "eval_runtime": 3.2658, | |
| "eval_samples_per_second": 1672.787, | |
| "eval_steps_per_second": 6.736, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6593, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.600219659527732, | |
| "eval_loss": 0.662101149559021, | |
| "eval_runtime": 3.2317, | |
| "eval_samples_per_second": 1690.46, | |
| "eval_steps_per_second": 6.808, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6441, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6004027091341754, | |
| "eval_loss": 0.6633503437042236, | |
| "eval_runtime": 3.2521, | |
| "eval_samples_per_second": 1679.844, | |
| "eval_steps_per_second": 6.765, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6338, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5941790225151016, | |
| "eval_loss": 0.6694063544273376, | |
| "eval_runtime": 3.2458, | |
| "eval_samples_per_second": 1683.098, | |
| "eval_steps_per_second": 6.778, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6238, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5919824272377814, | |
| "eval_loss": 0.6732103228569031, | |
| "eval_runtime": 3.3256, | |
| "eval_samples_per_second": 1642.691, | |
| "eval_steps_per_second": 6.615, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6125, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5969247666117518, | |
| "eval_loss": 0.6865329742431641, | |
| "eval_runtime": 3.2519, | |
| "eval_samples_per_second": 1679.947, | |
| "eval_steps_per_second": 6.765, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6026, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5799011532125206, | |
| "eval_loss": 0.7079607248306274, | |
| "eval_runtime": 3.2508, | |
| "eval_samples_per_second": 1680.489, | |
| "eval_steps_per_second": 6.767, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 2870, | |
| "total_flos": 1.917231202880717e+16, | |
| "train_loss": 0.6379064101375354, | |
| "train_runtime": 940.7499, | |
| "train_samples_per_second": 5566.995, | |
| "train_steps_per_second": 21.791 | |
| } | |
| ], | |
| "max_steps": 20500, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.917231202880717e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |