| { | |
| "best_metric": 0.875968992248062, | |
| "best_model_checkpoint": "final_models/glue_models/structroberta_s1_final//finetune/sst2/checkpoint-2400", | |
| "epoch": 10.0, | |
| "global_step": 4220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.8444882035255432, | |
| "eval_f1": 0.8435643564356434, | |
| "eval_loss": 0.367070734500885, | |
| "eval_mcc": 0.6889720757983125, | |
| "eval_runtime": 1.0028, | |
| "eval_samples_per_second": 506.605, | |
| "eval_steps_per_second": 63.824, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.407582938388626e-05, | |
| "loss": 0.3432, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.8503937125205994, | |
| "eval_f1": 0.8576779026217228, | |
| "eval_loss": 0.4026205837726593, | |
| "eval_mcc": 0.7052069065942033, | |
| "eval_runtime": 0.9977, | |
| "eval_samples_per_second": 509.178, | |
| "eval_steps_per_second": 64.148, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.815165876777251e-05, | |
| "loss": 0.1697, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.8503937125205994, | |
| "eval_f1": 0.8429752066115702, | |
| "eval_loss": 0.4639718234539032, | |
| "eval_mcc": 0.7033257298160759, | |
| "eval_runtime": 1.002, | |
| "eval_samples_per_second": 506.968, | |
| "eval_steps_per_second": 63.87, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 3.222748815165877e-05, | |
| "loss": 0.1015, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.8562992215156555, | |
| "eval_f1": 0.856581532416503, | |
| "eval_loss": 0.5244479179382324, | |
| "eval_mcc": 0.7126570409670451, | |
| "eval_runtime": 0.9988, | |
| "eval_samples_per_second": 508.605, | |
| "eval_steps_per_second": 64.076, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.6303317535545023e-05, | |
| "loss": 0.0634, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_accuracy": 0.8582677245140076, | |
| "eval_f1": 0.8646616541353384, | |
| "eval_loss": 0.5205051302909851, | |
| "eval_mcc": 0.7204265556951228, | |
| "eval_runtime": 0.998, | |
| "eval_samples_per_second": 509.036, | |
| "eval_steps_per_second": 64.13, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_accuracy": 0.874015748500824, | |
| "eval_f1": 0.875968992248062, | |
| "eval_loss": 0.5028718709945679, | |
| "eval_mcc": 0.7486469063845578, | |
| "eval_runtime": 0.9979, | |
| "eval_samples_per_second": 509.05, | |
| "eval_steps_per_second": 64.132, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 2.037914691943128e-05, | |
| "loss": 0.0376, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "eval_accuracy": 0.8582677245140076, | |
| "eval_f1": 0.8615384615384615, | |
| "eval_loss": 0.6611855626106262, | |
| "eval_mcc": 0.7176830919534406, | |
| "eval_runtime": 0.9984, | |
| "eval_samples_per_second": 508.813, | |
| "eval_steps_per_second": 64.102, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.4454976303317535e-05, | |
| "loss": 0.0248, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "eval_accuracy": 0.8523622155189514, | |
| "eval_f1": 0.8565965583173996, | |
| "eval_loss": 0.770823061466217, | |
| "eval_mcc": 0.7063807287425677, | |
| "eval_runtime": 0.998, | |
| "eval_samples_per_second": 509.003, | |
| "eval_steps_per_second": 64.126, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 8.530805687203793e-06, | |
| "loss": 0.0126, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "eval_accuracy": 0.8385826945304871, | |
| "eval_f1": 0.8446969696969698, | |
| "eval_loss": 0.8427420854568481, | |
| "eval_mcc": 0.679823561390335, | |
| "eval_runtime": 0.9998, | |
| "eval_samples_per_second": 508.116, | |
| "eval_steps_per_second": 64.015, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.6066350710900472e-06, | |
| "loss": 0.0073, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "eval_accuracy": 0.8484252095222473, | |
| "eval_f1": 0.8516377649325626, | |
| "eval_loss": 0.8851189613342285, | |
| "eval_mcc": 0.6978210010131721, | |
| "eval_runtime": 0.9988, | |
| "eval_samples_per_second": 508.59, | |
| "eval_steps_per_second": 64.074, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 4220, | |
| "total_flos": 4.216516889591808e+16, | |
| "train_loss": 0.09028764292244662, | |
| "train_runtime": 2361.2466, | |
| "train_samples_per_second": 213.989, | |
| "train_steps_per_second": 1.787 | |
| } | |
| ], | |
| "max_steps": 4220, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.216516889591808e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |