mehrshadk's picture
Training in progress, epoch 5, checkpoint
0e8a5e6 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 50,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.5,
"grad_norm": 21.752552032470703,
"learning_rate": 2.75e-05,
"loss": 0.6935,
"step": 50
},
{
"epoch": 0.5,
"eval_electronics_AUC": 0.7358429858429858,
"eval_electronics_accuracy": 0.7,
"eval_electronics_f1": 0.6103896103896104,
"eval_electronics_loss": 0.6476753354072571,
"eval_electronics_macro_f1": 0.68324358568261,
"eval_electronics_micro_f1": 0.7,
"eval_electronics_precision": 0.5875,
"eval_electronics_recall": 0.6351351351351351,
"eval_electronics_runtime": 1.1589,
"eval_electronics_samples_per_second": 345.158,
"eval_electronics_specificity": 0.7380952380952381,
"eval_electronics_steps_per_second": 0.863,
"step": 50
},
{
"epoch": 1.0,
"grad_norm": 6.66761589050293,
"learning_rate": 2.5e-05,
"loss": 0.5986,
"step": 100
},
{
"epoch": 1.0,
"eval_electronics_AUC": 0.8359877734877735,
"eval_electronics_accuracy": 0.7575,
"eval_electronics_f1": 0.7220630372492837,
"eval_electronics_loss": 0.47413870692253113,
"eval_electronics_macro_f1": 0.753492715963888,
"eval_electronics_micro_f1": 0.7575,
"eval_electronics_precision": 0.6268656716417911,
"eval_electronics_recall": 0.8513513513513513,
"eval_electronics_runtime": 1.1471,
"eval_electronics_samples_per_second": 348.696,
"eval_electronics_specificity": 0.7023809523809523,
"eval_electronics_steps_per_second": 0.872,
"step": 100
},
{
"epoch": 1.5,
"grad_norm": 6.782297611236572,
"learning_rate": 2.25e-05,
"loss": 0.5047,
"step": 150
},
{
"epoch": 1.5,
"eval_electronics_AUC": 0.8423959673959673,
"eval_electronics_accuracy": 0.7675,
"eval_electronics_f1": 0.6593406593406593,
"eval_electronics_loss": 0.560465931892395,
"eval_electronics_macro_f1": 0.7414350355526826,
"eval_electronics_micro_f1": 0.7675,
"eval_electronics_precision": 0.72,
"eval_electronics_recall": 0.6081081081081081,
"eval_electronics_runtime": 1.1397,
"eval_electronics_samples_per_second": 350.962,
"eval_electronics_specificity": 0.8611111111111112,
"eval_electronics_steps_per_second": 0.877,
"step": 150
},
{
"epoch": 2.0,
"grad_norm": 9.472657203674316,
"learning_rate": 1.9999999999999998e-05,
"loss": 0.4905,
"step": 200
},
{
"epoch": 2.0,
"eval_electronics_AUC": 0.8667953667953668,
"eval_electronics_accuracy": 0.7775,
"eval_electronics_f1": 0.7405247813411079,
"eval_electronics_loss": 0.46520760655403137,
"eval_electronics_macro_f1": 0.7728882112394817,
"eval_electronics_micro_f1": 0.7775,
"eval_electronics_precision": 0.6512820512820513,
"eval_electronics_recall": 0.8581081081081081,
"eval_electronics_runtime": 1.1429,
"eval_electronics_samples_per_second": 349.973,
"eval_electronics_specificity": 0.7301587301587301,
"eval_electronics_steps_per_second": 0.875,
"step": 200
},
{
"epoch": 2.5,
"grad_norm": 2.6548750400543213,
"learning_rate": 1.7500000000000002e-05,
"loss": 0.3569,
"step": 250
},
{
"epoch": 2.5,
"eval_electronics_AUC": 0.884009009009009,
"eval_electronics_accuracy": 0.785,
"eval_electronics_f1": 0.6971830985915493,
"eval_electronics_loss": 0.4542394280433655,
"eval_electronics_macro_f1": 0.7652582159624413,
"eval_electronics_micro_f1": 0.785,
"eval_electronics_precision": 0.7279411764705882,
"eval_electronics_recall": 0.668918918918919,
"eval_electronics_runtime": 1.1506,
"eval_electronics_samples_per_second": 347.647,
"eval_electronics_specificity": 0.8531746031746031,
"eval_electronics_steps_per_second": 0.869,
"step": 250
},
{
"epoch": 3.0,
"grad_norm": 9.189770698547363,
"learning_rate": 1.5e-05,
"loss": 0.341,
"step": 300
},
{
"epoch": 3.0,
"eval_electronics_AUC": 0.8937419562419562,
"eval_electronics_accuracy": 0.81,
"eval_electronics_f1": 0.75,
"eval_electronics_loss": 0.4145946800708771,
"eval_electronics_macro_f1": 0.7983870967741935,
"eval_electronics_micro_f1": 0.81,
"eval_electronics_precision": 0.7307692307692307,
"eval_electronics_recall": 0.7702702702702703,
"eval_electronics_runtime": 1.138,
"eval_electronics_samples_per_second": 351.479,
"eval_electronics_specificity": 0.8333333333333334,
"eval_electronics_steps_per_second": 0.879,
"step": 300
},
{
"epoch": 3.5,
"grad_norm": 11.805768966674805,
"learning_rate": 1.25e-05,
"loss": 0.2177,
"step": 350
},
{
"epoch": 3.5,
"eval_electronics_AUC": 0.8913824538824539,
"eval_electronics_accuracy": 0.825,
"eval_electronics_f1": 0.7619047619047619,
"eval_electronics_loss": 0.6068415641784668,
"eval_electronics_macro_f1": 0.8117824204780726,
"eval_electronics_micro_f1": 0.825,
"eval_electronics_precision": 0.7671232876712328,
"eval_electronics_recall": 0.7567567567567568,
"eval_electronics_runtime": 1.1896,
"eval_electronics_samples_per_second": 336.239,
"eval_electronics_specificity": 0.8650793650793651,
"eval_electronics_steps_per_second": 0.841,
"step": 350
},
{
"epoch": 4.0,
"grad_norm": 4.138590335845947,
"learning_rate": 9.999999999999999e-06,
"loss": 0.213,
"step": 400
},
{
"epoch": 4.0,
"eval_electronics_AUC": 0.8949217074217075,
"eval_electronics_accuracy": 0.8175,
"eval_electronics_f1": 0.762214983713355,
"eval_electronics_loss": 0.46484577655792236,
"eval_electronics_macro_f1": 0.807070980700491,
"eval_electronics_micro_f1": 0.8175,
"eval_electronics_precision": 0.7358490566037735,
"eval_electronics_recall": 0.7905405405405406,
"eval_electronics_runtime": 1.1622,
"eval_electronics_samples_per_second": 344.168,
"eval_electronics_specificity": 0.8333333333333334,
"eval_electronics_steps_per_second": 0.86,
"step": 400
},
{
"epoch": 4.5,
"grad_norm": 0.15363281965255737,
"learning_rate": 7.5e-06,
"loss": 0.1091,
"step": 450
},
{
"epoch": 4.5,
"eval_electronics_AUC": 0.8612451737451737,
"eval_electronics_accuracy": 0.825,
"eval_electronics_f1": 0.7602739726027398,
"eval_electronics_loss": 0.8552714586257935,
"eval_electronics_macro_f1": 0.8112393485060942,
"eval_electronics_micro_f1": 0.825,
"eval_electronics_precision": 0.7708333333333334,
"eval_electronics_recall": 0.75,
"eval_electronics_runtime": 1.1603,
"eval_electronics_samples_per_second": 344.745,
"eval_electronics_specificity": 0.8690476190476191,
"eval_electronics_steps_per_second": 0.862,
"step": 450
},
{
"epoch": 5.0,
"grad_norm": 0.43008989095687866,
"learning_rate": 4.9999999999999996e-06,
"loss": 0.1342,
"step": 500
},
{
"epoch": 5.0,
"eval_electronics_AUC": 0.8647576147576147,
"eval_electronics_accuracy": 0.8125,
"eval_electronics_f1": 0.7368421052631579,
"eval_electronics_loss": 0.813143789768219,
"eval_electronics_macro_f1": 0.7956055186509964,
"eval_electronics_micro_f1": 0.8125,
"eval_electronics_precision": 0.7664233576642335,
"eval_electronics_recall": 0.7094594594594594,
"eval_electronics_runtime": 1.1599,
"eval_electronics_samples_per_second": 344.859,
"eval_electronics_specificity": 0.873015873015873,
"eval_electronics_steps_per_second": 0.862,
"step": 500
}
],
"logging_steps": 50,
"max_steps": 600,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 679093866240000.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}