matthieulel's picture
End of training
92e294f verified
{
"best_metric": 0.85456595264938,
"best_model_checkpoint": "convnextv2-pico-1k-224-finetuned-galaxy10-decals/checkpoint-1687",
"epoch": 29.76,
"eval_steps": 500,
"global_step": 1860,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.16,
"learning_rate": 2.688172043010753e-06,
"loss": 2.3623,
"step": 10
},
{
"epoch": 0.32,
"learning_rate": 5.376344086021506e-06,
"loss": 1.9999,
"step": 20
},
{
"epoch": 0.48,
"learning_rate": 8.064516129032258e-06,
"loss": 1.7176,
"step": 30
},
{
"epoch": 0.64,
"learning_rate": 1.0752688172043012e-05,
"loss": 1.426,
"step": 40
},
{
"epoch": 0.8,
"learning_rate": 1.3440860215053763e-05,
"loss": 1.2838,
"step": 50
},
{
"epoch": 0.96,
"learning_rate": 1.6129032258064517e-05,
"loss": 1.0752,
"step": 60
},
{
"epoch": 0.99,
"eval_accuracy": 0.6696730552423901,
"eval_f1": 0.6581224923833207,
"eval_loss": 0.9584333300590515,
"eval_precision": 0.681955075159791,
"eval_recall": 0.6696730552423901,
"eval_runtime": 11.3585,
"eval_samples_per_second": 156.182,
"eval_steps_per_second": 2.465,
"step": 62
},
{
"epoch": 1.12,
"learning_rate": 1.881720430107527e-05,
"loss": 1.0243,
"step": 70
},
{
"epoch": 1.28,
"learning_rate": 2.1505376344086024e-05,
"loss": 0.9554,
"step": 80
},
{
"epoch": 1.44,
"learning_rate": 2.4193548387096777e-05,
"loss": 0.8633,
"step": 90
},
{
"epoch": 1.6,
"learning_rate": 2.6881720430107527e-05,
"loss": 0.8495,
"step": 100
},
{
"epoch": 1.76,
"learning_rate": 2.9569892473118284e-05,
"loss": 0.7749,
"step": 110
},
{
"epoch": 1.92,
"learning_rate": 3.2258064516129034e-05,
"loss": 0.814,
"step": 120
},
{
"epoch": 2.0,
"eval_accuracy": 0.7728297632468997,
"eval_f1": 0.7695111043149447,
"eval_loss": 0.6715956926345825,
"eval_precision": 0.778035548923254,
"eval_recall": 0.7728297632468997,
"eval_runtime": 11.2325,
"eval_samples_per_second": 157.935,
"eval_steps_per_second": 2.493,
"step": 125
},
{
"epoch": 2.08,
"learning_rate": 3.494623655913979e-05,
"loss": 0.7576,
"step": 130
},
{
"epoch": 2.24,
"learning_rate": 3.763440860215054e-05,
"loss": 0.767,
"step": 140
},
{
"epoch": 2.4,
"learning_rate": 4.032258064516129e-05,
"loss": 0.729,
"step": 150
},
{
"epoch": 2.56,
"learning_rate": 4.301075268817205e-05,
"loss": 0.7455,
"step": 160
},
{
"epoch": 2.72,
"learning_rate": 4.56989247311828e-05,
"loss": 0.7787,
"step": 170
},
{
"epoch": 2.88,
"learning_rate": 4.8387096774193554e-05,
"loss": 0.7289,
"step": 180
},
{
"epoch": 2.99,
"eval_accuracy": 0.7959413754227734,
"eval_f1": 0.7942666017808426,
"eval_loss": 0.6071233153343201,
"eval_precision": 0.8092808628958313,
"eval_recall": 0.7959413754227734,
"eval_runtime": 11.2854,
"eval_samples_per_second": 157.195,
"eval_steps_per_second": 2.481,
"step": 187
},
{
"epoch": 3.04,
"learning_rate": 4.98805256869773e-05,
"loss": 0.7705,
"step": 190
},
{
"epoch": 3.2,
"learning_rate": 4.9581839904420555e-05,
"loss": 0.6834,
"step": 200
},
{
"epoch": 3.36,
"learning_rate": 4.92831541218638e-05,
"loss": 0.7082,
"step": 210
},
{
"epoch": 3.52,
"learning_rate": 4.898446833930705e-05,
"loss": 0.7319,
"step": 220
},
{
"epoch": 3.68,
"learning_rate": 4.86857825567503e-05,
"loss": 0.6649,
"step": 230
},
{
"epoch": 3.84,
"learning_rate": 4.8387096774193554e-05,
"loss": 0.6879,
"step": 240
},
{
"epoch": 4.0,
"learning_rate": 4.80884109916368e-05,
"loss": 0.6444,
"step": 250
},
{
"epoch": 4.0,
"eval_accuracy": 0.8055242390078917,
"eval_f1": 0.8019293657837152,
"eval_loss": 0.5873242020606995,
"eval_precision": 0.8102990272941125,
"eval_recall": 0.8055242390078917,
"eval_runtime": 11.3178,
"eval_samples_per_second": 156.745,
"eval_steps_per_second": 2.474,
"step": 250
},
{
"epoch": 4.16,
"learning_rate": 4.778972520908005e-05,
"loss": 0.7234,
"step": 260
},
{
"epoch": 4.32,
"learning_rate": 4.74910394265233e-05,
"loss": 0.6762,
"step": 270
},
{
"epoch": 4.48,
"learning_rate": 4.7192353643966546e-05,
"loss": 0.6355,
"step": 280
},
{
"epoch": 4.64,
"learning_rate": 4.6893667861409805e-05,
"loss": 0.6065,
"step": 290
},
{
"epoch": 4.8,
"learning_rate": 4.659498207885305e-05,
"loss": 0.6422,
"step": 300
},
{
"epoch": 4.96,
"learning_rate": 4.62962962962963e-05,
"loss": 0.5855,
"step": 310
},
{
"epoch": 4.99,
"eval_accuracy": 0.8105975197294251,
"eval_f1": 0.8105509002140089,
"eval_loss": 0.5889072418212891,
"eval_precision": 0.8225901061121342,
"eval_recall": 0.8105975197294251,
"eval_runtime": 11.0401,
"eval_samples_per_second": 160.687,
"eval_steps_per_second": 2.536,
"step": 312
},
{
"epoch": 5.12,
"learning_rate": 4.5997610513739546e-05,
"loss": 0.6029,
"step": 320
},
{
"epoch": 5.28,
"learning_rate": 4.56989247311828e-05,
"loss": 0.6041,
"step": 330
},
{
"epoch": 5.44,
"learning_rate": 4.540023894862604e-05,
"loss": 0.5844,
"step": 340
},
{
"epoch": 5.6,
"learning_rate": 4.51015531660693e-05,
"loss": 0.5983,
"step": 350
},
{
"epoch": 5.76,
"learning_rate": 4.4802867383512545e-05,
"loss": 0.5578,
"step": 360
},
{
"epoch": 5.92,
"learning_rate": 4.4504181600955796e-05,
"loss": 0.5778,
"step": 370
},
{
"epoch": 6.0,
"eval_accuracy": 0.8280721533258174,
"eval_f1": 0.8274047538243907,
"eval_loss": 0.5039396286010742,
"eval_precision": 0.8321267696941235,
"eval_recall": 0.8280721533258174,
"eval_runtime": 11.2617,
"eval_samples_per_second": 157.526,
"eval_steps_per_second": 2.486,
"step": 375
},
{
"epoch": 6.08,
"learning_rate": 4.420549581839905e-05,
"loss": 0.5485,
"step": 380
},
{
"epoch": 6.24,
"learning_rate": 4.390681003584229e-05,
"loss": 0.5595,
"step": 390
},
{
"epoch": 6.4,
"learning_rate": 4.360812425328555e-05,
"loss": 0.5289,
"step": 400
},
{
"epoch": 6.56,
"learning_rate": 4.3309438470728796e-05,
"loss": 0.5614,
"step": 410
},
{
"epoch": 6.72,
"learning_rate": 4.301075268817205e-05,
"loss": 0.5539,
"step": 420
},
{
"epoch": 6.88,
"learning_rate": 4.271206690561529e-05,
"loss": 0.5575,
"step": 430
},
{
"epoch": 6.99,
"eval_accuracy": 0.8139797068771139,
"eval_f1": 0.8148378668703895,
"eval_loss": 0.516225278377533,
"eval_precision": 0.8235359050533735,
"eval_recall": 0.8139797068771139,
"eval_runtime": 11.2754,
"eval_samples_per_second": 157.334,
"eval_steps_per_second": 2.483,
"step": 437
},
{
"epoch": 7.04,
"learning_rate": 4.241338112305854e-05,
"loss": 0.5851,
"step": 440
},
{
"epoch": 7.2,
"learning_rate": 4.2114695340501795e-05,
"loss": 0.5349,
"step": 450
},
{
"epoch": 7.36,
"learning_rate": 4.1816009557945046e-05,
"loss": 0.4965,
"step": 460
},
{
"epoch": 7.52,
"learning_rate": 4.15173237753883e-05,
"loss": 0.532,
"step": 470
},
{
"epoch": 7.68,
"learning_rate": 4.121863799283154e-05,
"loss": 0.5033,
"step": 480
},
{
"epoch": 7.84,
"learning_rate": 4.0919952210274794e-05,
"loss": 0.512,
"step": 490
},
{
"epoch": 8.0,
"learning_rate": 4.062126642771804e-05,
"loss": 0.5011,
"step": 500
},
{
"epoch": 8.0,
"eval_accuracy": 0.8207440811724915,
"eval_f1": 0.820522196295754,
"eval_loss": 0.5369052886962891,
"eval_precision": 0.8234412549383173,
"eval_recall": 0.8207440811724915,
"eval_runtime": 11.324,
"eval_samples_per_second": 156.659,
"eval_steps_per_second": 2.473,
"step": 500
},
{
"epoch": 8.16,
"learning_rate": 4.032258064516129e-05,
"loss": 0.4667,
"step": 510
},
{
"epoch": 8.32,
"learning_rate": 4.002389486260454e-05,
"loss": 0.4628,
"step": 520
},
{
"epoch": 8.48,
"learning_rate": 3.972520908004779e-05,
"loss": 0.5108,
"step": 530
},
{
"epoch": 8.64,
"learning_rate": 3.9426523297491045e-05,
"loss": 0.4899,
"step": 540
},
{
"epoch": 8.8,
"learning_rate": 3.912783751493429e-05,
"loss": 0.5208,
"step": 550
},
{
"epoch": 8.96,
"learning_rate": 3.882915173237754e-05,
"loss": 0.4968,
"step": 560
},
{
"epoch": 8.99,
"eval_accuracy": 0.8291995490417137,
"eval_f1": 0.8270441342966817,
"eval_loss": 0.5151572823524475,
"eval_precision": 0.828170653083022,
"eval_recall": 0.8291995490417137,
"eval_runtime": 11.2905,
"eval_samples_per_second": 157.123,
"eval_steps_per_second": 2.48,
"step": 562
},
{
"epoch": 9.12,
"learning_rate": 3.8530465949820786e-05,
"loss": 0.476,
"step": 570
},
{
"epoch": 9.28,
"learning_rate": 3.8231780167264044e-05,
"loss": 0.4657,
"step": 580
},
{
"epoch": 9.44,
"learning_rate": 3.793309438470729e-05,
"loss": 0.4585,
"step": 590
},
{
"epoch": 9.6,
"learning_rate": 3.763440860215054e-05,
"loss": 0.467,
"step": 600
},
{
"epoch": 9.76,
"learning_rate": 3.7335722819593785e-05,
"loss": 0.4452,
"step": 610
},
{
"epoch": 9.92,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.4593,
"step": 620
},
{
"epoch": 10.0,
"eval_accuracy": 0.8382187147688839,
"eval_f1": 0.8367497522940034,
"eval_loss": 0.48535043001174927,
"eval_precision": 0.8408349921437133,
"eval_recall": 0.8382187147688839,
"eval_runtime": 11.1396,
"eval_samples_per_second": 159.252,
"eval_steps_per_second": 2.514,
"step": 625
},
{
"epoch": 10.08,
"learning_rate": 3.673835125448029e-05,
"loss": 0.4543,
"step": 630
},
{
"epoch": 10.24,
"learning_rate": 3.643966547192354e-05,
"loss": 0.4221,
"step": 640
},
{
"epoch": 10.4,
"learning_rate": 3.614097968936679e-05,
"loss": 0.4882,
"step": 650
},
{
"epoch": 10.56,
"learning_rate": 3.5842293906810036e-05,
"loss": 0.4611,
"step": 660
},
{
"epoch": 10.72,
"learning_rate": 3.554360812425329e-05,
"loss": 0.4417,
"step": 670
},
{
"epoch": 10.88,
"learning_rate": 3.524492234169653e-05,
"loss": 0.4442,
"step": 680
},
{
"epoch": 10.99,
"eval_accuracy": 0.8416009019165727,
"eval_f1": 0.8411034299357704,
"eval_loss": 0.49225476384162903,
"eval_precision": 0.8423398049495254,
"eval_recall": 0.8416009019165727,
"eval_runtime": 11.3299,
"eval_samples_per_second": 156.577,
"eval_steps_per_second": 2.471,
"step": 687
},
{
"epoch": 11.04,
"learning_rate": 3.494623655913979e-05,
"loss": 0.4156,
"step": 690
},
{
"epoch": 11.2,
"learning_rate": 3.4647550776583035e-05,
"loss": 0.406,
"step": 700
},
{
"epoch": 11.36,
"learning_rate": 3.4348864994026287e-05,
"loss": 0.3917,
"step": 710
},
{
"epoch": 11.52,
"learning_rate": 3.405017921146954e-05,
"loss": 0.4306,
"step": 720
},
{
"epoch": 11.68,
"learning_rate": 3.375149342891278e-05,
"loss": 0.4261,
"step": 730
},
{
"epoch": 11.84,
"learning_rate": 3.3452807646356034e-05,
"loss": 0.4127,
"step": 740
},
{
"epoch": 12.0,
"learning_rate": 3.3154121863799286e-05,
"loss": 0.4071,
"step": 750
},
{
"epoch": 12.0,
"eval_accuracy": 0.8376550169109357,
"eval_f1": 0.8331467588806277,
"eval_loss": 0.5311676859855652,
"eval_precision": 0.8355591516110524,
"eval_recall": 0.8376550169109357,
"eval_runtime": 11.2728,
"eval_samples_per_second": 157.369,
"eval_steps_per_second": 2.484,
"step": 750
},
{
"epoch": 12.16,
"learning_rate": 3.285543608124254e-05,
"loss": 0.4095,
"step": 760
},
{
"epoch": 12.32,
"learning_rate": 3.255675029868578e-05,
"loss": 0.3963,
"step": 770
},
{
"epoch": 12.48,
"learning_rate": 3.2258064516129034e-05,
"loss": 0.4089,
"step": 780
},
{
"epoch": 12.64,
"learning_rate": 3.195937873357228e-05,
"loss": 0.4132,
"step": 790
},
{
"epoch": 12.8,
"learning_rate": 3.1660692951015537e-05,
"loss": 0.4453,
"step": 800
},
{
"epoch": 12.96,
"learning_rate": 3.136200716845878e-05,
"loss": 0.4057,
"step": 810
},
{
"epoch": 12.99,
"eval_accuracy": 0.8432919954904171,
"eval_f1": 0.8416137806009663,
"eval_loss": 0.4954121708869934,
"eval_precision": 0.8448799019254453,
"eval_recall": 0.8432919954904171,
"eval_runtime": 11.2621,
"eval_samples_per_second": 157.519,
"eval_steps_per_second": 2.486,
"step": 812
},
{
"epoch": 13.12,
"learning_rate": 3.106332138590203e-05,
"loss": 0.3744,
"step": 820
},
{
"epoch": 13.28,
"learning_rate": 3.0764635603345284e-05,
"loss": 0.3736,
"step": 830
},
{
"epoch": 13.44,
"learning_rate": 3.046594982078853e-05,
"loss": 0.417,
"step": 840
},
{
"epoch": 13.6,
"learning_rate": 3.016726403823178e-05,
"loss": 0.3653,
"step": 850
},
{
"epoch": 13.76,
"learning_rate": 2.9868578255675032e-05,
"loss": 0.3976,
"step": 860
},
{
"epoch": 13.92,
"learning_rate": 2.9569892473118284e-05,
"loss": 0.4074,
"step": 870
},
{
"epoch": 14.0,
"eval_accuracy": 0.8534385569334837,
"eval_f1": 0.8492870108565496,
"eval_loss": 0.47353193163871765,
"eval_precision": 0.8509134567052327,
"eval_recall": 0.8534385569334837,
"eval_runtime": 11.3202,
"eval_samples_per_second": 156.711,
"eval_steps_per_second": 2.473,
"step": 875
},
{
"epoch": 14.08,
"learning_rate": 2.9271206690561532e-05,
"loss": 0.3713,
"step": 880
},
{
"epoch": 14.24,
"learning_rate": 2.897252090800478e-05,
"loss": 0.3466,
"step": 890
},
{
"epoch": 14.4,
"learning_rate": 2.8673835125448028e-05,
"loss": 0.3751,
"step": 900
},
{
"epoch": 14.56,
"learning_rate": 2.8375149342891276e-05,
"loss": 0.3886,
"step": 910
},
{
"epoch": 14.72,
"learning_rate": 2.807646356033453e-05,
"loss": 0.3793,
"step": 920
},
{
"epoch": 14.88,
"learning_rate": 2.777777777777778e-05,
"loss": 0.3709,
"step": 930
},
{
"epoch": 14.99,
"eval_accuracy": 0.8461104847801578,
"eval_f1": 0.8442085878019076,
"eval_loss": 0.497653990983963,
"eval_precision": 0.844988014104576,
"eval_recall": 0.8461104847801578,
"eval_runtime": 11.295,
"eval_samples_per_second": 157.061,
"eval_steps_per_second": 2.479,
"step": 937
},
{
"epoch": 15.04,
"learning_rate": 2.747909199522103e-05,
"loss": 0.3604,
"step": 940
},
{
"epoch": 15.2,
"learning_rate": 2.718040621266428e-05,
"loss": 0.3384,
"step": 950
},
{
"epoch": 15.36,
"learning_rate": 2.6881720430107527e-05,
"loss": 0.337,
"step": 960
},
{
"epoch": 15.52,
"learning_rate": 2.6583034647550775e-05,
"loss": 0.3563,
"step": 970
},
{
"epoch": 15.68,
"learning_rate": 2.628434886499403e-05,
"loss": 0.341,
"step": 980
},
{
"epoch": 15.84,
"learning_rate": 2.5985663082437278e-05,
"loss": 0.3561,
"step": 990
},
{
"epoch": 16.0,
"learning_rate": 2.5686977299880526e-05,
"loss": 0.3467,
"step": 1000
},
{
"epoch": 16.0,
"eval_accuracy": 0.8286358511837655,
"eval_f1": 0.8273552077905264,
"eval_loss": 0.5363720655441284,
"eval_precision": 0.8277942397724607,
"eval_recall": 0.8286358511837655,
"eval_runtime": 11.2632,
"eval_samples_per_second": 157.504,
"eval_steps_per_second": 2.486,
"step": 1000
},
{
"epoch": 16.16,
"learning_rate": 2.5388291517323774e-05,
"loss": 0.3488,
"step": 1010
},
{
"epoch": 16.32,
"learning_rate": 2.5089605734767026e-05,
"loss": 0.3147,
"step": 1020
},
{
"epoch": 16.48,
"learning_rate": 2.4790919952210277e-05,
"loss": 0.3281,
"step": 1030
},
{
"epoch": 16.64,
"learning_rate": 2.4492234169653525e-05,
"loss": 0.3676,
"step": 1040
},
{
"epoch": 16.8,
"learning_rate": 2.4193548387096777e-05,
"loss": 0.3334,
"step": 1050
},
{
"epoch": 16.96,
"learning_rate": 2.3894862604540025e-05,
"loss": 0.3129,
"step": 1060
},
{
"epoch": 16.99,
"eval_accuracy": 0.8421645997745209,
"eval_f1": 0.8375729961422652,
"eval_loss": 0.5695042610168457,
"eval_precision": 0.8412955682829938,
"eval_recall": 0.8421645997745209,
"eval_runtime": 11.3044,
"eval_samples_per_second": 156.931,
"eval_steps_per_second": 2.477,
"step": 1062
},
{
"epoch": 17.12,
"learning_rate": 2.3596176821983273e-05,
"loss": 0.3371,
"step": 1070
},
{
"epoch": 17.28,
"learning_rate": 2.3297491039426525e-05,
"loss": 0.3281,
"step": 1080
},
{
"epoch": 17.44,
"learning_rate": 2.2998805256869773e-05,
"loss": 0.36,
"step": 1090
},
{
"epoch": 17.6,
"learning_rate": 2.270011947431302e-05,
"loss": 0.3199,
"step": 1100
},
{
"epoch": 17.76,
"learning_rate": 2.2401433691756272e-05,
"loss": 0.3108,
"step": 1110
},
{
"epoch": 17.92,
"learning_rate": 2.2102747909199524e-05,
"loss": 0.3242,
"step": 1120
},
{
"epoch": 18.0,
"eval_accuracy": 0.8455467869222097,
"eval_f1": 0.8450233747368677,
"eval_loss": 0.5130738615989685,
"eval_precision": 0.8468561808534232,
"eval_recall": 0.8455467869222097,
"eval_runtime": 11.1586,
"eval_samples_per_second": 158.981,
"eval_steps_per_second": 2.509,
"step": 1125
},
{
"epoch": 18.08,
"learning_rate": 2.1804062126642775e-05,
"loss": 0.3123,
"step": 1130
},
{
"epoch": 18.24,
"learning_rate": 2.1505376344086024e-05,
"loss": 0.2948,
"step": 1140
},
{
"epoch": 18.4,
"learning_rate": 2.120669056152927e-05,
"loss": 0.2817,
"step": 1150
},
{
"epoch": 18.56,
"learning_rate": 2.0908004778972523e-05,
"loss": 0.3027,
"step": 1160
},
{
"epoch": 18.72,
"learning_rate": 2.060931899641577e-05,
"loss": 0.3025,
"step": 1170
},
{
"epoch": 18.88,
"learning_rate": 2.031063321385902e-05,
"loss": 0.3046,
"step": 1180
},
{
"epoch": 18.99,
"eval_accuracy": 0.8399098083427283,
"eval_f1": 0.8371411676075852,
"eval_loss": 0.5552954077720642,
"eval_precision": 0.8381896001479403,
"eval_recall": 0.8399098083427283,
"eval_runtime": 11.3483,
"eval_samples_per_second": 156.323,
"eval_steps_per_second": 2.467,
"step": 1187
},
{
"epoch": 19.04,
"learning_rate": 2.001194743130227e-05,
"loss": 0.3158,
"step": 1190
},
{
"epoch": 19.2,
"learning_rate": 1.9713261648745522e-05,
"loss": 0.2836,
"step": 1200
},
{
"epoch": 19.36,
"learning_rate": 1.941457586618877e-05,
"loss": 0.2912,
"step": 1210
},
{
"epoch": 19.52,
"learning_rate": 1.9115890083632022e-05,
"loss": 0.3061,
"step": 1220
},
{
"epoch": 19.68,
"learning_rate": 1.881720430107527e-05,
"loss": 0.285,
"step": 1230
},
{
"epoch": 19.84,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.2913,
"step": 1240
},
{
"epoch": 20.0,
"learning_rate": 1.821983273596177e-05,
"loss": 0.2805,
"step": 1250
},
{
"epoch": 20.0,
"eval_accuracy": 0.8523111612175873,
"eval_f1": 0.8456439029191984,
"eval_loss": 0.587050199508667,
"eval_precision": 0.8531689423038871,
"eval_recall": 0.8523111612175873,
"eval_runtime": 11.3595,
"eval_samples_per_second": 156.169,
"eval_steps_per_second": 2.465,
"step": 1250
},
{
"epoch": 20.16,
"learning_rate": 1.7921146953405018e-05,
"loss": 0.2785,
"step": 1260
},
{
"epoch": 20.32,
"learning_rate": 1.7622461170848266e-05,
"loss": 0.2796,
"step": 1270
},
{
"epoch": 20.48,
"learning_rate": 1.7323775388291518e-05,
"loss": 0.2834,
"step": 1280
},
{
"epoch": 20.64,
"learning_rate": 1.702508960573477e-05,
"loss": 0.2845,
"step": 1290
},
{
"epoch": 20.8,
"learning_rate": 1.6726403823178017e-05,
"loss": 0.3042,
"step": 1300
},
{
"epoch": 20.96,
"learning_rate": 1.642771804062127e-05,
"loss": 0.2776,
"step": 1310
},
{
"epoch": 20.99,
"eval_accuracy": 0.8432919954904171,
"eval_f1": 0.8403811306080587,
"eval_loss": 0.542815625667572,
"eval_precision": 0.8404231033443561,
"eval_recall": 0.8432919954904171,
"eval_runtime": 11.3482,
"eval_samples_per_second": 156.325,
"eval_steps_per_second": 2.467,
"step": 1312
},
{
"epoch": 21.12,
"learning_rate": 1.6129032258064517e-05,
"loss": 0.2688,
"step": 1320
},
{
"epoch": 21.28,
"learning_rate": 1.5830346475507768e-05,
"loss": 0.286,
"step": 1330
},
{
"epoch": 21.44,
"learning_rate": 1.5531660692951016e-05,
"loss": 0.2629,
"step": 1340
},
{
"epoch": 21.6,
"learning_rate": 1.5232974910394265e-05,
"loss": 0.2447,
"step": 1350
},
{
"epoch": 21.76,
"learning_rate": 1.4934289127837516e-05,
"loss": 0.2826,
"step": 1360
},
{
"epoch": 21.92,
"learning_rate": 1.4635603345280766e-05,
"loss": 0.2975,
"step": 1370
},
{
"epoch": 22.0,
"eval_accuracy": 0.8393461104847801,
"eval_f1": 0.8358819658141899,
"eval_loss": 0.562358021736145,
"eval_precision": 0.834423507730151,
"eval_recall": 0.8393461104847801,
"eval_runtime": 11.3588,
"eval_samples_per_second": 156.178,
"eval_steps_per_second": 2.465,
"step": 1375
},
{
"epoch": 22.08,
"learning_rate": 1.4336917562724014e-05,
"loss": 0.2766,
"step": 1380
},
{
"epoch": 22.24,
"learning_rate": 1.4038231780167265e-05,
"loss": 0.247,
"step": 1390
},
{
"epoch": 22.4,
"learning_rate": 1.3739545997610515e-05,
"loss": 0.271,
"step": 1400
},
{
"epoch": 22.56,
"learning_rate": 1.3440860215053763e-05,
"loss": 0.2532,
"step": 1410
},
{
"epoch": 22.72,
"learning_rate": 1.3142174432497015e-05,
"loss": 0.262,
"step": 1420
},
{
"epoch": 22.88,
"learning_rate": 1.2843488649940263e-05,
"loss": 0.268,
"step": 1430
},
{
"epoch": 22.99,
"eval_accuracy": 0.8494926719278467,
"eval_f1": 0.8497915694347914,
"eval_loss": 0.5484796166419983,
"eval_precision": 0.8517555265951766,
"eval_recall": 0.8494926719278467,
"eval_runtime": 11.387,
"eval_samples_per_second": 155.792,
"eval_steps_per_second": 2.459,
"step": 1437
},
{
"epoch": 23.04,
"learning_rate": 1.2544802867383513e-05,
"loss": 0.2547,
"step": 1440
},
{
"epoch": 23.2,
"learning_rate": 1.2246117084826763e-05,
"loss": 0.2351,
"step": 1450
},
{
"epoch": 23.36,
"learning_rate": 1.1947431302270013e-05,
"loss": 0.244,
"step": 1460
},
{
"epoch": 23.52,
"learning_rate": 1.1648745519713262e-05,
"loss": 0.2348,
"step": 1470
},
{
"epoch": 23.68,
"learning_rate": 1.135005973715651e-05,
"loss": 0.2807,
"step": 1480
},
{
"epoch": 23.84,
"learning_rate": 1.1051373954599762e-05,
"loss": 0.2266,
"step": 1490
},
{
"epoch": 24.0,
"learning_rate": 1.0752688172043012e-05,
"loss": 0.2535,
"step": 1500
},
{
"epoch": 24.0,
"eval_accuracy": 0.8382187147688839,
"eval_f1": 0.8358474613530663,
"eval_loss": 0.6135410666465759,
"eval_precision": 0.8367007977727741,
"eval_recall": 0.8382187147688839,
"eval_runtime": 11.3166,
"eval_samples_per_second": 156.761,
"eval_steps_per_second": 2.474,
"step": 1500
},
{
"epoch": 24.16,
"learning_rate": 1.0454002389486262e-05,
"loss": 0.2557,
"step": 1510
},
{
"epoch": 24.32,
"learning_rate": 1.015531660692951e-05,
"loss": 0.2767,
"step": 1520
},
{
"epoch": 24.48,
"learning_rate": 9.856630824372761e-06,
"loss": 0.2525,
"step": 1530
},
{
"epoch": 24.64,
"learning_rate": 9.557945041816011e-06,
"loss": 0.2234,
"step": 1540
},
{
"epoch": 24.8,
"learning_rate": 9.259259259259259e-06,
"loss": 0.2335,
"step": 1550
},
{
"epoch": 24.96,
"learning_rate": 8.960573476702509e-06,
"loss": 0.2543,
"step": 1560
},
{
"epoch": 24.99,
"eval_accuracy": 0.8393461104847801,
"eval_f1": 0.8375424065430281,
"eval_loss": 0.6102624535560608,
"eval_precision": 0.8389362458746243,
"eval_recall": 0.8393461104847801,
"eval_runtime": 11.3077,
"eval_samples_per_second": 156.884,
"eval_steps_per_second": 2.476,
"step": 1562
},
{
"epoch": 25.12,
"learning_rate": 8.661887694145759e-06,
"loss": 0.2304,
"step": 1570
},
{
"epoch": 25.28,
"learning_rate": 8.363201911589009e-06,
"loss": 0.2145,
"step": 1580
},
{
"epoch": 25.44,
"learning_rate": 8.064516129032258e-06,
"loss": 0.2143,
"step": 1590
},
{
"epoch": 25.6,
"learning_rate": 7.765830346475508e-06,
"loss": 0.2245,
"step": 1600
},
{
"epoch": 25.76,
"learning_rate": 7.467144563918758e-06,
"loss": 0.2199,
"step": 1610
},
{
"epoch": 25.92,
"learning_rate": 7.168458781362007e-06,
"loss": 0.2283,
"step": 1620
},
{
"epoch": 26.0,
"eval_accuracy": 0.8483652762119503,
"eval_f1": 0.8479659610328265,
"eval_loss": 0.5638566613197327,
"eval_precision": 0.8499280071208234,
"eval_recall": 0.8483652762119503,
"eval_runtime": 11.3149,
"eval_samples_per_second": 156.785,
"eval_steps_per_second": 2.475,
"step": 1625
},
{
"epoch": 26.08,
"learning_rate": 6.869772998805258e-06,
"loss": 0.243,
"step": 1630
},
{
"epoch": 26.24,
"learning_rate": 6.5710872162485075e-06,
"loss": 0.2348,
"step": 1640
},
{
"epoch": 26.4,
"learning_rate": 6.2724014336917564e-06,
"loss": 0.2338,
"step": 1650
},
{
"epoch": 26.56,
"learning_rate": 5.973715651135006e-06,
"loss": 0.2215,
"step": 1660
},
{
"epoch": 26.72,
"learning_rate": 5.675029868578255e-06,
"loss": 0.2428,
"step": 1670
},
{
"epoch": 26.88,
"learning_rate": 5.376344086021506e-06,
"loss": 0.2341,
"step": 1680
},
{
"epoch": 26.99,
"eval_accuracy": 0.85456595264938,
"eval_f1": 0.8545150298484757,
"eval_loss": 0.5794834494590759,
"eval_precision": 0.8564878833789361,
"eval_recall": 0.85456595264938,
"eval_runtime": 11.3134,
"eval_samples_per_second": 156.806,
"eval_steps_per_second": 2.475,
"step": 1687
},
{
"epoch": 27.04,
"learning_rate": 5.077658303464755e-06,
"loss": 0.2363,
"step": 1690
},
{
"epoch": 27.2,
"learning_rate": 4.7789725209080055e-06,
"loss": 0.2235,
"step": 1700
},
{
"epoch": 27.36,
"learning_rate": 4.4802867383512545e-06,
"loss": 0.2431,
"step": 1710
},
{
"epoch": 27.52,
"learning_rate": 4.181600955794504e-06,
"loss": 0.2245,
"step": 1720
},
{
"epoch": 27.68,
"learning_rate": 3.882915173237754e-06,
"loss": 0.2015,
"step": 1730
},
{
"epoch": 27.84,
"learning_rate": 3.5842293906810035e-06,
"loss": 0.2183,
"step": 1740
},
{
"epoch": 28.0,
"learning_rate": 3.2855436081242537e-06,
"loss": 0.2404,
"step": 1750
},
{
"epoch": 28.0,
"eval_accuracy": 0.8534385569334837,
"eval_f1": 0.8511010449598169,
"eval_loss": 0.5793879628181458,
"eval_precision": 0.8515289875366977,
"eval_recall": 0.8534385569334837,
"eval_runtime": 11.2916,
"eval_samples_per_second": 157.108,
"eval_steps_per_second": 2.48,
"step": 1750
},
{
"epoch": 28.16,
"learning_rate": 2.986857825567503e-06,
"loss": 0.2127,
"step": 1760
},
{
"epoch": 28.32,
"learning_rate": 2.688172043010753e-06,
"loss": 0.2077,
"step": 1770
},
{
"epoch": 28.48,
"learning_rate": 2.3894862604540028e-06,
"loss": 0.2056,
"step": 1780
},
{
"epoch": 28.64,
"learning_rate": 2.090800477897252e-06,
"loss": 0.2144,
"step": 1790
},
{
"epoch": 28.8,
"learning_rate": 1.7921146953405017e-06,
"loss": 0.2202,
"step": 1800
},
{
"epoch": 28.96,
"learning_rate": 1.4934289127837516e-06,
"loss": 0.2168,
"step": 1810
},
{
"epoch": 28.99,
"eval_accuracy": 0.85456595264938,
"eval_f1": 0.8523790480040142,
"eval_loss": 0.5651508569717407,
"eval_precision": 0.8524754972772818,
"eval_recall": 0.85456595264938,
"eval_runtime": 11.2766,
"eval_samples_per_second": 157.317,
"eval_steps_per_second": 2.483,
"step": 1812
},
{
"epoch": 29.12,
"learning_rate": 1.1947431302270014e-06,
"loss": 0.2141,
"step": 1820
},
{
"epoch": 29.28,
"learning_rate": 8.960573476702509e-07,
"loss": 0.2083,
"step": 1830
},
{
"epoch": 29.44,
"learning_rate": 5.973715651135007e-07,
"loss": 0.2043,
"step": 1840
},
{
"epoch": 29.6,
"learning_rate": 2.9868578255675034e-07,
"loss": 0.2162,
"step": 1850
},
{
"epoch": 29.76,
"learning_rate": 0.0,
"loss": 0.2057,
"step": 1860
},
{
"epoch": 29.76,
"eval_accuracy": 0.85456595264938,
"eval_f1": 0.8518371199726075,
"eval_loss": 0.5650473833084106,
"eval_precision": 0.8519237079812747,
"eval_recall": 0.85456595264938,
"eval_runtime": 11.3519,
"eval_samples_per_second": 156.274,
"eval_steps_per_second": 2.467,
"step": 1860
},
{
"epoch": 29.76,
"step": 1860,
"total_flos": 3.672047291926487e+18,
"train_loss": 0.4428144581856266,
"train_runtime": 3659.1539,
"train_samples_per_second": 130.866,
"train_steps_per_second": 0.508
}
],
"logging_steps": 10,
"max_steps": 1860,
"num_input_tokens_seen": 0,
"num_train_epochs": 30,
"save_steps": 500,
"total_flos": 3.672047291926487e+18,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}