| { | |
| "best_metric": 0.34269535541534424, | |
| "best_model_checkpoint": "robertafinetuninglargedatacsv/checkpoint-32", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 32, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 3.306401252746582, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.7102, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 2.7323529720306396, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.6948, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 3.1906545162200928, | |
| "learning_rate": 3.75e-06, | |
| "loss": 0.6916, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.5763970613479614, | |
| "learning_rate": 5e-06, | |
| "loss": 0.692, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 1.4885941743850708, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.6793, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 2.1479969024658203, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.683, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 1.085968017578125, | |
| "learning_rate": 8.75e-06, | |
| "loss": 0.6944, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.4903621673583984, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6797, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 1.4184839725494385, | |
| "learning_rate": 1.125e-05, | |
| "loss": 0.6806, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 2.2620794773101807, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.7063, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 3.0892398357391357, | |
| "learning_rate": 1.3750000000000002e-05, | |
| "loss": 0.7136, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.0407421588897705, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.6863, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 2.0362789630889893, | |
| "learning_rate": 1.6250000000000002e-05, | |
| "loss": 0.6805, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 2.197580099105835, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.6669, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 2.920393228530884, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.727, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.498661994934082, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7028, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5, | |
| "eval_auc": 0.796875, | |
| "eval_f1": 0.6666666666666666, | |
| "eval_loss": 0.6834808588027954, | |
| "eval_precision": 0.5, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 5.1228, | |
| "eval_samples_per_second": 6.247, | |
| "eval_steps_per_second": 0.39, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 3.0374414920806885, | |
| "learning_rate": 2.125e-05, | |
| "loss": 0.6991, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 3.0186045169830322, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.6261, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1875, | |
| "grad_norm": 2.047957181930542, | |
| "learning_rate": 2.375e-05, | |
| "loss": 0.6778, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 3.6687753200531006, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6227, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3125, | |
| "grad_norm": 1.6625615358352661, | |
| "learning_rate": 2.625e-05, | |
| "loss": 0.6775, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 2.2672011852264404, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.6381, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 3.273030996322632, | |
| "learning_rate": 2.8749999999999997e-05, | |
| "loss": 0.5787, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.291239261627197, | |
| "learning_rate": 3e-05, | |
| "loss": 0.71, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 3.6905951499938965, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.5826, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 4.929114818572998, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.5892, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6875, | |
| "grad_norm": 5.820431232452393, | |
| "learning_rate": 3.375000000000001e-05, | |
| "loss": 0.5683, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 10.009285926818848, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4001, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 13.668211936950684, | |
| "learning_rate": 3.625e-05, | |
| "loss": 0.6159, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 11.277912139892578, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.7048, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9375, | |
| "grad_norm": 7.137956142425537, | |
| "learning_rate": 3.875e-05, | |
| "loss": 0.4689, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.847255229949951, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3861, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9375, | |
| "eval_auc": 0.9765625, | |
| "eval_f1": 0.9375, | |
| "eval_loss": 0.34269535541534424, | |
| "eval_precision": 0.9375, | |
| "eval_recall": 0.9375, | |
| "eval_runtime": 5.178, | |
| "eval_samples_per_second": 6.18, | |
| "eval_steps_per_second": 0.386, | |
| "step": 32 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 16839107543040.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |