| { | |
| "best_metric": 0.7803163444639719, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-21/checkpoint-1712", | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 2782, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.750798225402832, | |
| "learning_rate": 0.00029949232941444245, | |
| "loss": 0.5927, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7147041593438781, | |
| "eval_f1": 0.4236686390532544, | |
| "eval_loss": 0.5422016978263855, | |
| "eval_mcc": 0.293664298583451, | |
| "eval_precision": 0.6485507246376812, | |
| "eval_recall": 0.3145869947275923, | |
| "eval_runtime": 3.1455, | |
| "eval_samples_per_second": 542.682, | |
| "eval_steps_per_second": 17.167, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.565903902053833, | |
| "learning_rate": 0.00027453463529657224, | |
| "loss": 0.5298, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7357937902753369, | |
| "eval_f1": 0.5124324324324324, | |
| "eval_loss": 0.5132952928543091, | |
| "eval_mcc": 0.36196008516504746, | |
| "eval_precision": 0.6657303370786517, | |
| "eval_recall": 0.4165202108963093, | |
| "eval_runtime": 3.2065, | |
| "eval_samples_per_second": 532.363, | |
| "eval_steps_per_second": 16.841, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.1464009284973145, | |
| "learning_rate": 0.00024957694117870203, | |
| "loss": 0.5051, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7340363210310487, | |
| "eval_f1": 0.41191709844559593, | |
| "eval_loss": 0.5232194066047668, | |
| "eval_mcc": 0.3506417522239517, | |
| "eval_precision": 0.7832512315270936, | |
| "eval_recall": 0.27943760984182775, | |
| "eval_runtime": 3.1624, | |
| "eval_samples_per_second": 539.782, | |
| "eval_steps_per_second": 17.076, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.173802375793457, | |
| "learning_rate": 0.00022461924706083182, | |
| "loss": 0.4904, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7434094903339191, | |
| "eval_f1": 0.46715328467153294, | |
| "eval_loss": 0.49749329686164856, | |
| "eval_mcc": 0.37656942971572427, | |
| "eval_precision": 0.758893280632411, | |
| "eval_recall": 0.3374340949033392, | |
| "eval_runtime": 3.4589, | |
| "eval_samples_per_second": 493.508, | |
| "eval_steps_per_second": 15.612, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.341954708099365, | |
| "learning_rate": 0.00019966155294296164, | |
| "loss": 0.4762, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7381370826010545, | |
| "eval_f1": 0.6499608457321848, | |
| "eval_loss": 0.5032187104225159, | |
| "eval_mcc": 0.4515022458491435, | |
| "eval_precision": 0.5861581920903954, | |
| "eval_recall": 0.7293497363796133, | |
| "eval_runtime": 3.1483, | |
| "eval_samples_per_second": 542.198, | |
| "eval_steps_per_second": 17.152, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.139582633972168, | |
| "learning_rate": 0.00017470385882509143, | |
| "loss": 0.4708, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7662565905096661, | |
| "eval_f1": 0.6288372093023256, | |
| "eval_loss": 0.48133140802383423, | |
| "eval_mcc": 0.46078884011912263, | |
| "eval_precision": 0.6679841897233202, | |
| "eval_recall": 0.5940246045694201, | |
| "eval_runtime": 3.3146, | |
| "eval_samples_per_second": 514.988, | |
| "eval_steps_per_second": 16.291, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.1577534675598145, | |
| "learning_rate": 0.00014974616470722122, | |
| "loss": 0.4562, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7738722905682484, | |
| "eval_f1": 0.5936842105263158, | |
| "eval_loss": 0.48448288440704346, | |
| "eval_mcc": 0.4625979514063081, | |
| "eval_precision": 0.7401574803149606, | |
| "eval_recall": 0.4956063268892794, | |
| "eval_runtime": 3.1532, | |
| "eval_samples_per_second": 541.359, | |
| "eval_steps_per_second": 17.126, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.001567840576172, | |
| "learning_rate": 0.00012478847058935101, | |
| "loss": 0.4448, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7803163444639719, | |
| "eval_f1": 0.6397694524495678, | |
| "eval_loss": 0.46665722131729126, | |
| "eval_mcc": 0.48808014304461034, | |
| "eval_precision": 0.7055084745762712, | |
| "eval_recall": 0.5852372583479789, | |
| "eval_runtime": 3.3884, | |
| "eval_samples_per_second": 503.777, | |
| "eval_steps_per_second": 15.937, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 9.252790451049805, | |
| "learning_rate": 9.983077647148082e-05, | |
| "loss": 0.4375, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7785588752196837, | |
| "eval_f1": 0.6204819277108434, | |
| "eval_loss": 0.46744513511657715, | |
| "eval_mcc": 0.4782291935803437, | |
| "eval_precision": 0.7236533957845434, | |
| "eval_recall": 0.5430579964850615, | |
| "eval_runtime": 3.1528, | |
| "eval_samples_per_second": 541.426, | |
| "eval_steps_per_second": 17.128, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.555416107177734, | |
| "learning_rate": 7.487308235361061e-05, | |
| "loss": 0.4281, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7709431751611013, | |
| "eval_f1": 0.6530612244897959, | |
| "eval_loss": 0.4783839285373688, | |
| "eval_mcc": 0.4821703835032232, | |
| "eval_precision": 0.6594982078853047, | |
| "eval_recall": 0.6467486818980668, | |
| "eval_runtime": 3.3012, | |
| "eval_samples_per_second": 517.082, | |
| "eval_steps_per_second": 16.358, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.022094488143921, | |
| "learning_rate": 4.991538823574041e-05, | |
| "loss": 0.4205, | |
| "step": 2354 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7803163444639719, | |
| "eval_f1": 0.6362754607177498, | |
| "eval_loss": 0.4678700566291809, | |
| "eval_mcc": 0.48668710736858267, | |
| "eval_precision": 0.70995670995671, | |
| "eval_recall": 0.5764499121265377, | |
| "eval_runtime": 3.1644, | |
| "eval_samples_per_second": 539.433, | |
| "eval_steps_per_second": 17.065, | |
| "step": 2354 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.7709765434265137, | |
| "learning_rate": 2.4957694117870205e-05, | |
| "loss": 0.4138, | |
| "step": 2568 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7797305213825425, | |
| "eval_f1": 0.6581818181818181, | |
| "eval_loss": 0.4624987542629242, | |
| "eval_mcc": 0.4966232685018691, | |
| "eval_precision": 0.6817325800376648, | |
| "eval_recall": 0.6362038664323374, | |
| "eval_runtime": 3.2014, | |
| "eval_samples_per_second": 533.205, | |
| "eval_steps_per_second": 16.868, | |
| "step": 2568 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 9.869558334350586, | |
| "learning_rate": 0.0, | |
| "loss": 0.4067, | |
| "step": 2782 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7768014059753954, | |
| "eval_f1": 0.6613333333333332, | |
| "eval_loss": 0.470124214887619, | |
| "eval_mcc": 0.4949917624648792, | |
| "eval_precision": 0.6690647482014388, | |
| "eval_recall": 0.6537785588752196, | |
| "eval_runtime": 3.1694, | |
| "eval_samples_per_second": 538.585, | |
| "eval_steps_per_second": 17.038, | |
| "step": 2782 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2782, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 13, | |
| "save_steps": 500, | |
| "total_flos": 6822083637720.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9862491765682508, | |
| "learning_rate": 0.00032445002353231265, | |
| "num_train_epochs": 13, | |
| "temperature": 19 | |
| } | |
| } | |