classify-ISIN-STEP7_binary / trainer_state.json
calculito's picture
calculito/classify-ISIN-STEP7_binary
a4546e8 verified
raw
history blame
22 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 20.0,
"eval_steps": 100,
"global_step": 960,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.20833333333333334,
"grad_norm": 21.270000457763672,
"learning_rate": 4.0000000000000003e-07,
"loss": 1.573,
"step": 10
},
{
"epoch": 0.4166666666666667,
"grad_norm": 23.03444480895996,
"learning_rate": 8.000000000000001e-07,
"loss": 1.511,
"step": 20
},
{
"epoch": 0.625,
"grad_norm": 25.605567932128906,
"learning_rate": 1.2000000000000002e-06,
"loss": 1.38,
"step": 30
},
{
"epoch": 0.8333333333333334,
"grad_norm": 27.936298370361328,
"learning_rate": 1.6000000000000001e-06,
"loss": 1.2262,
"step": 40
},
{
"epoch": 1.0416666666666667,
"grad_norm": 27.172657012939453,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.0138,
"step": 50
},
{
"epoch": 1.25,
"grad_norm": 15.924922943115234,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.8283,
"step": 60
},
{
"epoch": 1.4583333333333333,
"grad_norm": 15.164562225341797,
"learning_rate": 2.8000000000000003e-06,
"loss": 0.6043,
"step": 70
},
{
"epoch": 1.6666666666666665,
"grad_norm": 12.560202598571777,
"learning_rate": 3.2000000000000003e-06,
"loss": 0.4287,
"step": 80
},
{
"epoch": 1.875,
"grad_norm": 12.39704418182373,
"learning_rate": 3.6000000000000003e-06,
"loss": 0.3089,
"step": 90
},
{
"epoch": 2.0833333333333335,
"grad_norm": 6.223243236541748,
"learning_rate": 4.000000000000001e-06,
"loss": 0.2172,
"step": 100
},
{
"epoch": 2.0833333333333335,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.1748288869857788,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.8468,
"eval_samples_per_second": 135.809,
"eval_steps_per_second": 9.448,
"step": 100
},
{
"epoch": 2.2916666666666665,
"grad_norm": 2.8637752532958984,
"learning_rate": 4.4e-06,
"loss": 0.1346,
"step": 110
},
{
"epoch": 2.5,
"grad_norm": 2.1358628273010254,
"learning_rate": 4.800000000000001e-06,
"loss": 0.0979,
"step": 120
},
{
"epoch": 2.7083333333333335,
"grad_norm": 0.8700501918792725,
"learning_rate": 5.2e-06,
"loss": 0.0466,
"step": 130
},
{
"epoch": 2.9166666666666665,
"grad_norm": 0.4395294487476349,
"learning_rate": 5.600000000000001e-06,
"loss": 0.0223,
"step": 140
},
{
"epoch": 3.125,
"grad_norm": 0.23942218720912933,
"learning_rate": 6e-06,
"loss": 0.0469,
"step": 150
},
{
"epoch": 3.3333333333333335,
"grad_norm": 0.16919788718223572,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.009,
"step": 160
},
{
"epoch": 3.5416666666666665,
"grad_norm": 0.1501808762550354,
"learning_rate": 6.800000000000001e-06,
"loss": 0.0069,
"step": 170
},
{
"epoch": 3.75,
"grad_norm": 0.08838535845279694,
"learning_rate": 7.2000000000000005e-06,
"loss": 0.0053,
"step": 180
},
{
"epoch": 3.9583333333333335,
"grad_norm": 0.08624881505966187,
"learning_rate": 7.600000000000001e-06,
"loss": 0.0041,
"step": 190
},
{
"epoch": 4.166666666666667,
"grad_norm": 0.1038636863231659,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0224,
"step": 200
},
{
"epoch": 4.166666666666667,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.003508708206936717,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6622,
"eval_samples_per_second": 173.658,
"eval_steps_per_second": 12.081,
"step": 200
},
{
"epoch": 4.375,
"grad_norm": 0.08752647787332535,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0033,
"step": 210
},
{
"epoch": 4.583333333333333,
"grad_norm": 0.10564388334751129,
"learning_rate": 8.8e-06,
"loss": 0.023,
"step": 220
},
{
"epoch": 4.791666666666667,
"grad_norm": 0.08224104344844818,
"learning_rate": 9.200000000000002e-06,
"loss": 0.0026,
"step": 230
},
{
"epoch": 5.0,
"grad_norm": 0.04869069904088974,
"learning_rate": 9.600000000000001e-06,
"loss": 0.0023,
"step": 240
},
{
"epoch": 5.208333333333333,
"grad_norm": 0.04047730192542076,
"learning_rate": 1e-05,
"loss": 0.002,
"step": 250
},
{
"epoch": 5.416666666666667,
"grad_norm": 0.046379927545785904,
"learning_rate": 1.04e-05,
"loss": 0.0018,
"step": 260
},
{
"epoch": 5.625,
"grad_norm": 0.03858022019267082,
"learning_rate": 1.0800000000000002e-05,
"loss": 0.0016,
"step": 270
},
{
"epoch": 5.833333333333333,
"grad_norm": 0.05792458727955818,
"learning_rate": 1.1200000000000001e-05,
"loss": 0.0225,
"step": 280
},
{
"epoch": 6.041666666666667,
"grad_norm": 0.08217403292655945,
"learning_rate": 1.16e-05,
"loss": 0.0197,
"step": 290
},
{
"epoch": 6.25,
"grad_norm": 0.03509116172790527,
"learning_rate": 1.2e-05,
"loss": 0.0015,
"step": 300
},
{
"epoch": 6.25,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.0013682404533028603,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6551,
"eval_samples_per_second": 175.543,
"eval_steps_per_second": 12.212,
"step": 300
},
{
"epoch": 6.458333333333333,
"grad_norm": 0.08348393440246582,
"learning_rate": 1.2400000000000002e-05,
"loss": 0.0263,
"step": 310
},
{
"epoch": 6.666666666666667,
"grad_norm": 0.0798223465681076,
"learning_rate": 1.2800000000000001e-05,
"loss": 0.0014,
"step": 320
},
{
"epoch": 6.875,
"grad_norm": 0.02793905697762966,
"learning_rate": 1.3200000000000002e-05,
"loss": 0.0011,
"step": 330
},
{
"epoch": 7.083333333333333,
"grad_norm": 0.027305051684379578,
"learning_rate": 1.3600000000000002e-05,
"loss": 0.0011,
"step": 340
},
{
"epoch": 7.291666666666667,
"grad_norm": 0.026127709075808525,
"learning_rate": 1.4e-05,
"loss": 0.001,
"step": 350
},
{
"epoch": 7.5,
"grad_norm": 0.01849101483821869,
"learning_rate": 1.4400000000000001e-05,
"loss": 0.0008,
"step": 360
},
{
"epoch": 7.708333333333333,
"grad_norm": 0.13365192711353302,
"learning_rate": 1.48e-05,
"loss": 0.0199,
"step": 370
},
{
"epoch": 7.916666666666667,
"grad_norm": 0.11528883129358292,
"learning_rate": 1.5200000000000002e-05,
"loss": 0.001,
"step": 380
},
{
"epoch": 8.125,
"grad_norm": 0.02463056519627571,
"learning_rate": 1.5600000000000003e-05,
"loss": 0.0007,
"step": 390
},
{
"epoch": 8.333333333333334,
"grad_norm": 0.03092743456363678,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.0098,
"step": 400
},
{
"epoch": 8.333333333333334,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.0006972891860641539,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6482,
"eval_samples_per_second": 177.428,
"eval_steps_per_second": 12.343,
"step": 400
},
{
"epoch": 8.541666666666666,
"grad_norm": 0.03988618031144142,
"learning_rate": 1.64e-05,
"loss": 0.0066,
"step": 410
},
{
"epoch": 8.75,
"grad_norm": 0.02140761725604534,
"learning_rate": 1.6800000000000002e-05,
"loss": 0.0007,
"step": 420
},
{
"epoch": 8.958333333333334,
"grad_norm": 0.021611498668789864,
"learning_rate": 1.72e-05,
"loss": 0.0006,
"step": 430
},
{
"epoch": 9.166666666666666,
"grad_norm": 0.010292780585587025,
"learning_rate": 1.76e-05,
"loss": 0.0005,
"step": 440
},
{
"epoch": 9.375,
"grad_norm": 9.613543510437012,
"learning_rate": 1.8e-05,
"loss": 0.0059,
"step": 450
},
{
"epoch": 9.583333333333334,
"grad_norm": 0.018242070451378822,
"learning_rate": 1.8400000000000003e-05,
"loss": 0.0004,
"step": 460
},
{
"epoch": 9.791666666666666,
"grad_norm": 0.015562301501631737,
"learning_rate": 1.88e-05,
"loss": 0.0072,
"step": 470
},
{
"epoch": 10.0,
"grad_norm": 0.0493791326880455,
"learning_rate": 1.9200000000000003e-05,
"loss": 0.0005,
"step": 480
},
{
"epoch": 10.208333333333334,
"grad_norm": 0.017059756442904472,
"learning_rate": 1.9600000000000002e-05,
"loss": 0.0004,
"step": 490
},
{
"epoch": 10.416666666666666,
"grad_norm": 0.019885728135704994,
"learning_rate": 2e-05,
"loss": 0.0094,
"step": 500
},
{
"epoch": 10.416666666666666,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.000431541062425822,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6315,
"eval_samples_per_second": 182.109,
"eval_steps_per_second": 12.668,
"step": 500
},
{
"epoch": 10.625,
"grad_norm": 0.018004899844527245,
"learning_rate": 1.956521739130435e-05,
"loss": 0.0004,
"step": 510
},
{
"epoch": 10.833333333333334,
"grad_norm": 0.01476901862770319,
"learning_rate": 1.9130434782608697e-05,
"loss": 0.0004,
"step": 520
},
{
"epoch": 11.041666666666666,
"grad_norm": 0.009996396489441395,
"learning_rate": 1.8695652173913045e-05,
"loss": 0.0004,
"step": 530
},
{
"epoch": 11.25,
"grad_norm": 0.014687100425362587,
"learning_rate": 1.8260869565217393e-05,
"loss": 0.0173,
"step": 540
},
{
"epoch": 11.458333333333334,
"grad_norm": 0.014822542667388916,
"learning_rate": 1.782608695652174e-05,
"loss": 0.0004,
"step": 550
},
{
"epoch": 11.666666666666666,
"grad_norm": 0.012569064274430275,
"learning_rate": 1.739130434782609e-05,
"loss": 0.0034,
"step": 560
},
{
"epoch": 11.875,
"grad_norm": 0.02039659582078457,
"learning_rate": 1.6956521739130437e-05,
"loss": 0.0003,
"step": 570
},
{
"epoch": 12.083333333333334,
"grad_norm": 0.015721727162599564,
"learning_rate": 1.6521739130434785e-05,
"loss": 0.0003,
"step": 580
},
{
"epoch": 12.291666666666666,
"grad_norm": 0.016610363498330116,
"learning_rate": 1.6086956521739132e-05,
"loss": 0.0089,
"step": 590
},
{
"epoch": 12.5,
"grad_norm": 0.013905448839068413,
"learning_rate": 1.565217391304348e-05,
"loss": 0.0003,
"step": 600
},
{
"epoch": 12.5,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.00028381560696288943,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.623,
"eval_samples_per_second": 184.591,
"eval_steps_per_second": 12.841,
"step": 600
},
{
"epoch": 12.708333333333334,
"grad_norm": 0.008639817126095295,
"learning_rate": 1.5217391304347828e-05,
"loss": 0.0003,
"step": 610
},
{
"epoch": 12.916666666666666,
"grad_norm": 0.013035294599831104,
"learning_rate": 1.4782608695652174e-05,
"loss": 0.0002,
"step": 620
},
{
"epoch": 13.125,
"grad_norm": 0.005680867005139589,
"learning_rate": 1.4347826086956522e-05,
"loss": 0.0002,
"step": 630
},
{
"epoch": 13.333333333333334,
"grad_norm": 0.007329205516725779,
"learning_rate": 1.391304347826087e-05,
"loss": 0.0042,
"step": 640
},
{
"epoch": 13.541666666666666,
"grad_norm": 0.006812490988522768,
"learning_rate": 1.3478260869565218e-05,
"loss": 0.0002,
"step": 650
},
{
"epoch": 13.75,
"grad_norm": 0.009631389752030373,
"learning_rate": 1.3043478260869566e-05,
"loss": 0.0018,
"step": 660
},
{
"epoch": 13.958333333333334,
"grad_norm": 0.010884142480790615,
"learning_rate": 1.2608695652173915e-05,
"loss": 0.0002,
"step": 670
},
{
"epoch": 14.166666666666666,
"grad_norm": 0.012448645196855068,
"learning_rate": 1.2173913043478263e-05,
"loss": 0.0029,
"step": 680
},
{
"epoch": 14.375,
"grad_norm": 0.14938339591026306,
"learning_rate": 1.1739130434782611e-05,
"loss": 0.0014,
"step": 690
},
{
"epoch": 14.583333333333334,
"grad_norm": 0.014957024715840816,
"learning_rate": 1.1304347826086957e-05,
"loss": 0.0002,
"step": 700
},
{
"epoch": 14.583333333333334,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.00022231435286812484,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6111,
"eval_samples_per_second": 188.179,
"eval_steps_per_second": 13.091,
"step": 700
},
{
"epoch": 14.791666666666666,
"grad_norm": 0.00558474101126194,
"learning_rate": 1.0869565217391305e-05,
"loss": 0.0002,
"step": 710
},
{
"epoch": 15.0,
"grad_norm": 0.006383685395121574,
"learning_rate": 1.0434782608695653e-05,
"loss": 0.0002,
"step": 720
},
{
"epoch": 15.208333333333334,
"grad_norm": 0.007420627865940332,
"learning_rate": 1e-05,
"loss": 0.0009,
"step": 730
},
{
"epoch": 15.416666666666666,
"grad_norm": 0.006337632890790701,
"learning_rate": 9.565217391304349e-06,
"loss": 0.002,
"step": 740
},
{
"epoch": 15.625,
"grad_norm": 0.005708240903913975,
"learning_rate": 9.130434782608697e-06,
"loss": 0.0002,
"step": 750
},
{
"epoch": 15.833333333333334,
"grad_norm": 0.006120636593550444,
"learning_rate": 8.695652173913044e-06,
"loss": 0.0002,
"step": 760
},
{
"epoch": 16.041666666666668,
"grad_norm": 0.004273953847587109,
"learning_rate": 8.260869565217392e-06,
"loss": 0.0002,
"step": 770
},
{
"epoch": 16.25,
"grad_norm": 0.00433018384501338,
"learning_rate": 7.82608695652174e-06,
"loss": 0.0002,
"step": 780
},
{
"epoch": 16.458333333333332,
"grad_norm": 0.004073740914463997,
"learning_rate": 7.391304347826087e-06,
"loss": 0.0013,
"step": 790
},
{
"epoch": 16.666666666666668,
"grad_norm": 0.004012993536889553,
"learning_rate": 6.956521739130435e-06,
"loss": 0.0002,
"step": 800
},
{
"epoch": 16.666666666666668,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.00017283624038100243,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6058,
"eval_samples_per_second": 189.819,
"eval_steps_per_second": 13.205,
"step": 800
},
{
"epoch": 16.875,
"grad_norm": 0.00514675909653306,
"learning_rate": 6.521739130434783e-06,
"loss": 0.0002,
"step": 810
},
{
"epoch": 17.083333333333332,
"grad_norm": 0.005228080321103334,
"learning_rate": 6.086956521739132e-06,
"loss": 0.0012,
"step": 820
},
{
"epoch": 17.291666666666668,
"grad_norm": 0.005168806295841932,
"learning_rate": 5.652173913043479e-06,
"loss": 0.0008,
"step": 830
},
{
"epoch": 17.5,
"grad_norm": 0.005291006062179804,
"learning_rate": 5.2173913043478265e-06,
"loss": 0.0002,
"step": 840
},
{
"epoch": 17.708333333333332,
"grad_norm": 0.004714244045317173,
"learning_rate": 4.782608695652174e-06,
"loss": 0.0002,
"step": 850
},
{
"epoch": 17.916666666666668,
"grad_norm": 0.004473408684134483,
"learning_rate": 4.347826086956522e-06,
"loss": 0.0019,
"step": 860
},
{
"epoch": 18.125,
"grad_norm": 0.004324081353843212,
"learning_rate": 3.91304347826087e-06,
"loss": 0.0002,
"step": 870
},
{
"epoch": 18.333333333333332,
"grad_norm": 0.004104109480977058,
"learning_rate": 3.4782608695652175e-06,
"loss": 0.0002,
"step": 880
},
{
"epoch": 18.541666666666668,
"grad_norm": 0.00470446702092886,
"learning_rate": 3.043478260869566e-06,
"loss": 0.0002,
"step": 890
},
{
"epoch": 18.75,
"grad_norm": 0.004495694767683744,
"learning_rate": 2.6086956521739132e-06,
"loss": 0.0002,
"step": 900
},
{
"epoch": 18.75,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.0001600511313881725,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6078,
"eval_samples_per_second": 189.214,
"eval_steps_per_second": 13.163,
"step": 900
},
{
"epoch": 18.958333333333332,
"grad_norm": 0.004032615572214127,
"learning_rate": 2.173913043478261e-06,
"loss": 0.0019,
"step": 910
},
{
"epoch": 19.166666666666668,
"grad_norm": 0.004253485705703497,
"learning_rate": 1.7391304347826088e-06,
"loss": 0.0002,
"step": 920
},
{
"epoch": 19.375,
"grad_norm": 0.004380085505545139,
"learning_rate": 1.3043478260869566e-06,
"loss": 0.0007,
"step": 930
},
{
"epoch": 19.583333333333332,
"grad_norm": 0.0033130357041954994,
"learning_rate": 8.695652173913044e-07,
"loss": 0.0002,
"step": 940
},
{
"epoch": 19.791666666666668,
"grad_norm": 0.0035925493575632572,
"learning_rate": 4.347826086956522e-07,
"loss": 0.0002,
"step": 950
},
{
"epoch": 20.0,
"grad_norm": 0.0037953455466777086,
"learning_rate": 0.0,
"loss": 0.0012,
"step": 960
},
{
"epoch": 20.0,
"step": 960,
"total_flos": 112414246047360.0,
"train_loss": 0.10126596404067338,
"train_runtime": 514.198,
"train_samples_per_second": 59.277,
"train_steps_per_second": 1.867
},
{
"epoch": 20.0,
"eval_accuracy": 1.0,
"eval_accuracy_label_ GD622:Null": 0.0,
"eval_accuracy_label_GD622:NULL": 0.0,
"eval_accuracy_label_GD622:Null": 1.0,
"eval_accuracy_label_GD622:YES": 1.0,
"eval_f1": 1.0,
"eval_loss": 0.00015732948668301105,
"eval_precision": 1.0,
"eval_recall": 1.0,
"eval_runtime": 0.6012,
"eval_samples_per_second": 191.294,
"eval_steps_per_second": 13.307,
"step": 960
}
],
"logging_steps": 10,
"max_steps": 960,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 112414246047360.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}