DunnBC22's picture
End of training
88d61ac
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 17.856,
"global_step": 558,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 8.928571428571428e-07,
"loss": 2.9786,
"step": 1
},
{
"epoch": 0.26,
"learning_rate": 7.142857142857143e-06,
"loss": 3.0172,
"step": 8
},
{
"epoch": 0.51,
"learning_rate": 1.4285714285714285e-05,
"loss": 2.9016,
"step": 16
},
{
"epoch": 0.77,
"learning_rate": 2.1428571428571428e-05,
"loss": 2.7266,
"step": 24
},
{
"epoch": 0.99,
"eval_Macro F1": 0.18274435256625565,
"eval_Macro Precision": 0.22464899203701116,
"eval_Macro Recall": 0.21014375581515982,
"eval_Micro F1": 0.208,
"eval_Micro Precision": 0.208,
"eval_Micro Recall": 0.208,
"eval_Weighted F1": 0.18112534226726865,
"eval_Weighted Precision": 0.21432408746391318,
"eval_Weighted Recall": 0.208,
"eval_accuracy": 0.208,
"eval_loss": 2.473806381225586,
"eval_runtime": 340.1389,
"eval_samples_per_second": 2.94,
"eval_steps_per_second": 0.094,
"step": 31
},
{
"epoch": 1.02,
"learning_rate": 2.857142857142857e-05,
"loss": 2.5835,
"step": 32
},
{
"epoch": 1.28,
"learning_rate": 3.571428571428572e-05,
"loss": 2.4181,
"step": 40
},
{
"epoch": 1.54,
"learning_rate": 4.2857142857142856e-05,
"loss": 2.2732,
"step": 48
},
{
"epoch": 1.79,
"learning_rate": 5e-05,
"loss": 2.171,
"step": 56
},
{
"epoch": 1.98,
"eval_Macro F1": 0.39250727765144267,
"eval_Macro Precision": 0.4446397468859132,
"eval_Macro Recall": 0.42427064875651993,
"eval_Micro F1": 0.42299999999999993,
"eval_Micro Precision": 0.423,
"eval_Micro Recall": 0.423,
"eval_Weighted F1": 0.39356324128346765,
"eval_Weighted Precision": 0.45034474867370766,
"eval_Weighted Recall": 0.423,
"eval_accuracy": 0.423,
"eval_loss": 1.8510247468948364,
"eval_runtime": 23.3293,
"eval_samples_per_second": 42.864,
"eval_steps_per_second": 1.372,
"step": 62
},
{
"epoch": 2.05,
"learning_rate": 4.9203187250996016e-05,
"loss": 2.0057,
"step": 64
},
{
"epoch": 2.3,
"learning_rate": 4.840637450199204e-05,
"loss": 1.8883,
"step": 72
},
{
"epoch": 2.56,
"learning_rate": 4.760956175298805e-05,
"loss": 1.7538,
"step": 80
},
{
"epoch": 2.82,
"learning_rate": 4.6812749003984064e-05,
"loss": 1.6525,
"step": 88
},
{
"epoch": 2.98,
"eval_Macro F1": 0.5854998981328875,
"eval_Macro Precision": 0.6283041062692261,
"eval_Macro Recall": 0.6124154907646651,
"eval_Micro F1": 0.61,
"eval_Micro Precision": 0.61,
"eval_Micro Recall": 0.61,
"eval_Weighted F1": 0.5883778569595041,
"eval_Weighted Precision": 0.6376755054054996,
"eval_Weighted Recall": 0.61,
"eval_accuracy": 0.61,
"eval_loss": 1.2633185386657715,
"eval_runtime": 23.4866,
"eval_samples_per_second": 42.577,
"eval_steps_per_second": 1.362,
"step": 93
},
{
"epoch": 3.07,
"learning_rate": 4.601593625498008e-05,
"loss": 1.4741,
"step": 96
},
{
"epoch": 3.33,
"learning_rate": 4.52191235059761e-05,
"loss": 1.4995,
"step": 104
},
{
"epoch": 3.58,
"learning_rate": 4.442231075697211e-05,
"loss": 1.3476,
"step": 112
},
{
"epoch": 3.84,
"learning_rate": 4.362549800796813e-05,
"loss": 1.346,
"step": 120
},
{
"epoch": 4.0,
"eval_Macro F1": 0.6991540885965197,
"eval_Macro Precision": 0.7033970309397422,
"eval_Macro Recall": 0.705813786350264,
"eval_Micro F1": 0.706,
"eval_Micro Precision": 0.706,
"eval_Micro Recall": 0.706,
"eval_Weighted F1": 0.7023066954415087,
"eval_Weighted Precision": 0.7095418750091215,
"eval_Weighted Recall": 0.706,
"eval_accuracy": 0.706,
"eval_loss": 1.025865077972412,
"eval_runtime": 25.2258,
"eval_samples_per_second": 39.642,
"eval_steps_per_second": 1.269,
"step": 125
},
{
"epoch": 4.1,
"learning_rate": 4.2828685258964146e-05,
"loss": 1.361,
"step": 128
},
{
"epoch": 4.35,
"learning_rate": 4.203187250996016e-05,
"loss": 1.2724,
"step": 136
},
{
"epoch": 4.61,
"learning_rate": 4.123505976095618e-05,
"loss": 1.2634,
"step": 144
},
{
"epoch": 4.86,
"learning_rate": 4.043824701195219e-05,
"loss": 1.253,
"step": 152
},
{
"epoch": 4.99,
"eval_Macro F1": 0.7239315752163233,
"eval_Macro Precision": 0.7261038012066161,
"eval_Macro Recall": 0.7290773391581917,
"eval_Micro F1": 0.729,
"eval_Micro Precision": 0.729,
"eval_Micro Recall": 0.729,
"eval_Weighted F1": 0.7276898829930567,
"eval_Weighted Precision": 0.7340103739336716,
"eval_Weighted Recall": 0.729,
"eval_accuracy": 0.729,
"eval_loss": 0.9180329442024231,
"eval_runtime": 24.244,
"eval_samples_per_second": 41.247,
"eval_steps_per_second": 1.32,
"step": 156
},
{
"epoch": 5.12,
"learning_rate": 3.964143426294821e-05,
"loss": 1.1127,
"step": 160
},
{
"epoch": 5.38,
"learning_rate": 3.884462151394422e-05,
"loss": 1.2072,
"step": 168
},
{
"epoch": 5.63,
"learning_rate": 3.804780876494024e-05,
"loss": 1.1051,
"step": 176
},
{
"epoch": 5.89,
"learning_rate": 3.7250996015936256e-05,
"loss": 1.0975,
"step": 184
},
{
"epoch": 5.98,
"eval_Macro F1": 0.7437283226344332,
"eval_Macro Precision": 0.7525965488125351,
"eval_Macro Recall": 0.7471731443855207,
"eval_Micro F1": 0.747,
"eval_Micro Precision": 0.747,
"eval_Micro Recall": 0.747,
"eval_Weighted F1": 0.7479974803294792,
"eval_Weighted Precision": 0.7608663358741853,
"eval_Weighted Recall": 0.747,
"eval_accuracy": 0.747,
"eval_loss": 0.8858795166015625,
"eval_runtime": 24.5565,
"eval_samples_per_second": 40.722,
"eval_steps_per_second": 1.303,
"step": 187
},
{
"epoch": 6.14,
"learning_rate": 3.6454183266932277e-05,
"loss": 1.0922,
"step": 192
},
{
"epoch": 6.4,
"learning_rate": 3.565737051792829e-05,
"loss": 1.0797,
"step": 200
},
{
"epoch": 6.66,
"learning_rate": 3.4860557768924304e-05,
"loss": 1.0066,
"step": 208
},
{
"epoch": 6.91,
"learning_rate": 3.406374501992032e-05,
"loss": 1.1122,
"step": 216
},
{
"epoch": 6.98,
"eval_Macro F1": 0.7577762006111012,
"eval_Macro Precision": 0.7726630864911376,
"eval_Macro Recall": 0.7594437988791525,
"eval_Micro F1": 0.76,
"eval_Micro Precision": 0.76,
"eval_Micro Recall": 0.76,
"eval_Weighted F1": 0.7606134776349917,
"eval_Weighted Precision": 0.7772442524935343,
"eval_Weighted Recall": 0.76,
"eval_accuracy": 0.76,
"eval_loss": 0.8269779086112976,
"eval_runtime": 24.3399,
"eval_samples_per_second": 41.085,
"eval_steps_per_second": 1.315,
"step": 218
},
{
"epoch": 7.17,
"learning_rate": 3.326693227091633e-05,
"loss": 0.988,
"step": 224
},
{
"epoch": 7.42,
"learning_rate": 3.247011952191235e-05,
"loss": 1.0301,
"step": 232
},
{
"epoch": 7.68,
"learning_rate": 3.1673306772908366e-05,
"loss": 0.9724,
"step": 240
},
{
"epoch": 7.94,
"learning_rate": 3.0876494023904386e-05,
"loss": 1.0365,
"step": 248
},
{
"epoch": 8.0,
"eval_Macro F1": 0.7730173451661722,
"eval_Macro Precision": 0.7919871344293006,
"eval_Macro Recall": 0.7734992733449892,
"eval_Micro F1": 0.775,
"eval_Micro Precision": 0.775,
"eval_Micro Recall": 0.775,
"eval_Weighted F1": 0.7759067917731501,
"eval_Weighted Precision": 0.7957122736768526,
"eval_Weighted Recall": 0.775,
"eval_accuracy": 0.775,
"eval_loss": 0.7806075215339661,
"eval_runtime": 23.5653,
"eval_samples_per_second": 42.435,
"eval_steps_per_second": 1.358,
"step": 250
},
{
"epoch": 8.19,
"learning_rate": 3.00796812749004e-05,
"loss": 0.9482,
"step": 256
},
{
"epoch": 8.45,
"learning_rate": 2.9282868525896417e-05,
"loss": 0.9364,
"step": 264
},
{
"epoch": 8.7,
"learning_rate": 2.8486055776892434e-05,
"loss": 0.9417,
"step": 272
},
{
"epoch": 8.96,
"learning_rate": 2.7689243027888445e-05,
"loss": 1.004,
"step": 280
},
{
"epoch": 8.99,
"eval_Macro F1": 0.7956663948657638,
"eval_Macro Precision": 0.8150981558104353,
"eval_Macro Recall": 0.7955573765342054,
"eval_Micro F1": 0.796,
"eval_Micro Precision": 0.796,
"eval_Micro Recall": 0.796,
"eval_Weighted F1": 0.7977477337842563,
"eval_Weighted Precision": 0.8193174046095921,
"eval_Weighted Recall": 0.796,
"eval_accuracy": 0.796,
"eval_loss": 0.7471871972084045,
"eval_runtime": 24.2779,
"eval_samples_per_second": 41.19,
"eval_steps_per_second": 1.318,
"step": 281
},
{
"epoch": 9.22,
"learning_rate": 2.6892430278884462e-05,
"loss": 0.8964,
"step": 288
},
{
"epoch": 9.47,
"learning_rate": 2.609561752988048e-05,
"loss": 0.9098,
"step": 296
},
{
"epoch": 9.73,
"learning_rate": 2.5298804780876496e-05,
"loss": 0.9576,
"step": 304
},
{
"epoch": 9.98,
"learning_rate": 2.4501992031872513e-05,
"loss": 0.9278,
"step": 312
},
{
"epoch": 9.98,
"eval_Macro F1": 0.7956928280629123,
"eval_Macro Precision": 0.8114507473411248,
"eval_Macro Recall": 0.7953090431389529,
"eval_Micro F1": 0.795,
"eval_Micro Precision": 0.795,
"eval_Micro Recall": 0.795,
"eval_Weighted F1": 0.7974177696936208,
"eval_Weighted Precision": 0.8157219938612603,
"eval_Weighted Recall": 0.795,
"eval_accuracy": 0.795,
"eval_loss": 0.7296148538589478,
"eval_runtime": 24.012,
"eval_samples_per_second": 41.646,
"eval_steps_per_second": 1.333,
"step": 312
},
{
"epoch": 10.24,
"learning_rate": 2.3705179282868527e-05,
"loss": 0.8623,
"step": 320
},
{
"epoch": 10.5,
"learning_rate": 2.290836653386454e-05,
"loss": 0.9,
"step": 328
},
{
"epoch": 10.75,
"learning_rate": 2.2111553784860558e-05,
"loss": 0.8767,
"step": 336
},
{
"epoch": 10.98,
"eval_Macro F1": 0.8077597575155505,
"eval_Macro Precision": 0.8135983545887402,
"eval_Macro Recall": 0.8090640618814349,
"eval_Micro F1": 0.809,
"eval_Micro Precision": 0.809,
"eval_Micro Recall": 0.809,
"eval_Weighted F1": 0.8101031910520137,
"eval_Weighted Precision": 0.8181958157506283,
"eval_Weighted Recall": 0.809,
"eval_accuracy": 0.809,
"eval_loss": 0.7256603240966797,
"eval_runtime": 23.8159,
"eval_samples_per_second": 41.989,
"eval_steps_per_second": 1.344,
"step": 343
},
{
"epoch": 11.01,
"learning_rate": 2.1314741035856575e-05,
"loss": 0.9239,
"step": 344
},
{
"epoch": 11.26,
"learning_rate": 2.0517928286852592e-05,
"loss": 0.8971,
"step": 352
},
{
"epoch": 11.52,
"learning_rate": 1.9721115537848606e-05,
"loss": 0.9387,
"step": 360
},
{
"epoch": 11.78,
"learning_rate": 1.8924302788844623e-05,
"loss": 0.8656,
"step": 368
},
{
"epoch": 12.0,
"eval_Macro F1": 0.8106317572025369,
"eval_Macro Precision": 0.8164411111931356,
"eval_Macro Recall": 0.812159053433045,
"eval_Micro F1": 0.8140000000000001,
"eval_Micro Precision": 0.814,
"eval_Micro Recall": 0.814,
"eval_Weighted F1": 0.813675273019787,
"eval_Weighted Precision": 0.8206734023674667,
"eval_Weighted Recall": 0.814,
"eval_accuracy": 0.814,
"eval_loss": 0.687544584274292,
"eval_runtime": 24.187,
"eval_samples_per_second": 41.344,
"eval_steps_per_second": 1.323,
"step": 375
},
{
"epoch": 12.03,
"learning_rate": 1.812749003984064e-05,
"loss": 0.8568,
"step": 376
},
{
"epoch": 12.29,
"learning_rate": 1.7330677290836657e-05,
"loss": 0.9105,
"step": 384
},
{
"epoch": 12.54,
"learning_rate": 1.653386454183267e-05,
"loss": 0.7687,
"step": 392
},
{
"epoch": 12.8,
"learning_rate": 1.5737051792828685e-05,
"loss": 0.7905,
"step": 400
},
{
"epoch": 12.99,
"eval_Macro F1": 0.807144557531445,
"eval_Macro Precision": 0.8145226206650988,
"eval_Macro Recall": 0.8067641953560594,
"eval_Micro F1": 0.808,
"eval_Micro Precision": 0.808,
"eval_Micro Recall": 0.808,
"eval_Weighted F1": 0.8093388838073848,
"eval_Weighted Precision": 0.8181875581570741,
"eval_Weighted Recall": 0.808,
"eval_accuracy": 0.808,
"eval_loss": 0.7060463428497314,
"eval_runtime": 23.4383,
"eval_samples_per_second": 42.665,
"eval_steps_per_second": 1.365,
"step": 406
},
{
"epoch": 13.06,
"learning_rate": 1.4940239043824702e-05,
"loss": 0.8453,
"step": 408
},
{
"epoch": 13.31,
"learning_rate": 1.4143426294820719e-05,
"loss": 0.7926,
"step": 416
},
{
"epoch": 13.57,
"learning_rate": 1.3346613545816733e-05,
"loss": 0.8721,
"step": 424
},
{
"epoch": 13.82,
"learning_rate": 1.254980079681275e-05,
"loss": 0.8804,
"step": 432
},
{
"epoch": 13.98,
"eval_Macro F1": 0.8182926573190057,
"eval_Macro Precision": 0.8215050680984892,
"eval_Macro Recall": 0.8183483781309105,
"eval_Micro F1": 0.82,
"eval_Micro Precision": 0.82,
"eval_Micro Recall": 0.82,
"eval_Weighted F1": 0.8213528638428045,
"eval_Weighted Precision": 0.8260224129157536,
"eval_Weighted Recall": 0.82,
"eval_accuracy": 0.82,
"eval_loss": 0.6848881244659424,
"eval_runtime": 24.2686,
"eval_samples_per_second": 41.206,
"eval_steps_per_second": 1.319,
"step": 437
},
{
"epoch": 14.08,
"learning_rate": 1.1752988047808767e-05,
"loss": 0.8041,
"step": 440
},
{
"epoch": 14.34,
"learning_rate": 1.095617529880478e-05,
"loss": 0.8291,
"step": 448
},
{
"epoch": 14.59,
"learning_rate": 1.0159362549800798e-05,
"loss": 0.8109,
"step": 456
},
{
"epoch": 14.85,
"learning_rate": 9.362549800796813e-06,
"loss": 0.8265,
"step": 464
},
{
"epoch": 14.98,
"eval_Macro F1": 0.814299562240979,
"eval_Macro Precision": 0.8205669101031257,
"eval_Macro Recall": 0.8141933591904861,
"eval_Micro F1": 0.816,
"eval_Micro Precision": 0.816,
"eval_Micro Recall": 0.816,
"eval_Weighted F1": 0.8171172793618807,
"eval_Weighted Precision": 0.8242012281620977,
"eval_Weighted Recall": 0.816,
"eval_accuracy": 0.816,
"eval_loss": 0.6820688247680664,
"eval_runtime": 24.1434,
"eval_samples_per_second": 41.419,
"eval_steps_per_second": 1.325,
"step": 468
},
{
"epoch": 15.1,
"learning_rate": 8.565737051792829e-06,
"loss": 0.7845,
"step": 472
},
{
"epoch": 15.36,
"learning_rate": 7.768924302788846e-06,
"loss": 0.8545,
"step": 480
},
{
"epoch": 15.62,
"learning_rate": 6.97211155378486e-06,
"loss": 0.7685,
"step": 488
},
{
"epoch": 15.87,
"learning_rate": 6.175298804780877e-06,
"loss": 0.7929,
"step": 496
},
{
"epoch": 16.0,
"eval_Macro F1": 0.8151613063095204,
"eval_Macro Precision": 0.818634892770315,
"eval_Macro Recall": 0.8167488060127654,
"eval_Micro F1": 0.818,
"eval_Micro Precision": 0.818,
"eval_Micro Recall": 0.818,
"eval_Weighted F1": 0.8184268497331145,
"eval_Weighted Precision": 0.8239745058078853,
"eval_Weighted Recall": 0.818,
"eval_accuracy": 0.818,
"eval_loss": 0.6877326369285583,
"eval_runtime": 23.4065,
"eval_samples_per_second": 42.723,
"eval_steps_per_second": 1.367,
"step": 500
},
{
"epoch": 16.13,
"learning_rate": 5.378486055776893e-06,
"loss": 0.8338,
"step": 504
},
{
"epoch": 16.38,
"learning_rate": 4.581673306772908e-06,
"loss": 0.8154,
"step": 512
},
{
"epoch": 16.64,
"learning_rate": 3.7848605577689246e-06,
"loss": 0.8054,
"step": 520
},
{
"epoch": 16.9,
"learning_rate": 2.9880478087649404e-06,
"loss": 0.7993,
"step": 528
},
{
"epoch": 16.99,
"eval_Macro F1": 0.8233893661888394,
"eval_Macro Precision": 0.8281793690152945,
"eval_Macro Recall": 0.822655931109436,
"eval_Micro F1": 0.825,
"eval_Micro Precision": 0.825,
"eval_Micro Recall": 0.825,
"eval_Weighted F1": 0.8258766455781327,
"eval_Weighted Precision": 0.8305915821251276,
"eval_Weighted Recall": 0.825,
"eval_accuracy": 0.825,
"eval_loss": 0.6717957854270935,
"eval_runtime": 24.415,
"eval_samples_per_second": 40.958,
"eval_steps_per_second": 1.311,
"step": 531
},
{
"epoch": 17.15,
"learning_rate": 2.1912350597609563e-06,
"loss": 0.7572,
"step": 536
},
{
"epoch": 17.41,
"learning_rate": 1.3944223107569721e-06,
"loss": 0.8022,
"step": 544
},
{
"epoch": 17.66,
"learning_rate": 5.976095617529881e-07,
"loss": 0.7954,
"step": 552
},
{
"epoch": 17.86,
"eval_Macro F1": 0.8242217019056312,
"eval_Macro Precision": 0.8293094252297472,
"eval_Macro Recall": 0.8236754312981294,
"eval_Micro F1": 0.826,
"eval_Micro Precision": 0.826,
"eval_Micro Recall": 0.826,
"eval_Weighted F1": 0.827162008102242,
"eval_Weighted Precision": 0.8326500761444383,
"eval_Weighted Recall": 0.826,
"eval_accuracy": 0.826,
"eval_loss": 0.6715443134307861,
"eval_runtime": 24.1692,
"eval_samples_per_second": 41.375,
"eval_steps_per_second": 1.324,
"step": 558
},
{
"epoch": 17.86,
"step": 558,
"total_flos": 5.536126404560683e+18,
"train_loss": 1.1784635154149865,
"train_runtime": 5483.2083,
"train_samples_per_second": 13.131,
"train_steps_per_second": 0.102
}
],
"max_steps": 558,
"num_train_epochs": 18,
"total_flos": 5.536126404560683e+18,
"trial_name": null,
"trial_params": null
}