younggi's picture
Training in progress, epoch 0
503a110
raw
history blame
58.8 kB
{
"best_metric": 1.0,
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-1950",
"epoch": 49.02,
"global_step": 3750,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 1.3333333333333334e-06,
"loss": 2.3504,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 2.666666666666667e-06,
"loss": 2.409,
"step": 20
},
{
"epoch": 0.01,
"learning_rate": 4.000000000000001e-06,
"loss": 2.3102,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 5.333333333333334e-06,
"loss": 2.3537,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 6.666666666666667e-06,
"loss": 2.2566,
"step": 50
},
{
"epoch": 0.02,
"learning_rate": 8.000000000000001e-06,
"loss": 2.2599,
"step": 60
},
{
"epoch": 0.02,
"learning_rate": 9.333333333333334e-06,
"loss": 2.3212,
"step": 70
},
{
"epoch": 0.02,
"eval_accuracy": 0.25806451612903225,
"eval_loss": 2.2295100688934326,
"eval_runtime": 2.0436,
"eval_samples_per_second": 15.169,
"eval_steps_per_second": 3.915,
"step": 75
},
{
"epoch": 1.0,
"learning_rate": 1.0666666666666667e-05,
"loss": 2.235,
"step": 80
},
{
"epoch": 1.0,
"learning_rate": 1.2e-05,
"loss": 2.1721,
"step": 90
},
{
"epoch": 1.01,
"learning_rate": 1.3333333333333333e-05,
"loss": 2.1171,
"step": 100
},
{
"epoch": 1.01,
"learning_rate": 1.4666666666666668e-05,
"loss": 2.1702,
"step": 110
},
{
"epoch": 1.01,
"learning_rate": 1.6000000000000003e-05,
"loss": 2.0646,
"step": 120
},
{
"epoch": 1.01,
"learning_rate": 1.7333333333333336e-05,
"loss": 1.9602,
"step": 130
},
{
"epoch": 1.02,
"learning_rate": 1.866666666666667e-05,
"loss": 1.9113,
"step": 140
},
{
"epoch": 1.02,
"learning_rate": 2e-05,
"loss": 1.7775,
"step": 150
},
{
"epoch": 1.02,
"eval_accuracy": 0.3870967741935484,
"eval_loss": 1.780012845993042,
"eval_runtime": 2.1066,
"eval_samples_per_second": 14.716,
"eval_steps_per_second": 3.798,
"step": 150
},
{
"epoch": 2.0,
"learning_rate": 2.1333333333333335e-05,
"loss": 1.5096,
"step": 160
},
{
"epoch": 2.01,
"learning_rate": 2.2666666666666668e-05,
"loss": 1.4483,
"step": 170
},
{
"epoch": 2.01,
"learning_rate": 2.4e-05,
"loss": 1.4243,
"step": 180
},
{
"epoch": 2.01,
"learning_rate": 2.5333333333333337e-05,
"loss": 1.1883,
"step": 190
},
{
"epoch": 2.01,
"learning_rate": 2.6666666666666667e-05,
"loss": 0.9991,
"step": 200
},
{
"epoch": 2.02,
"learning_rate": 2.8000000000000003e-05,
"loss": 1.0043,
"step": 210
},
{
"epoch": 2.02,
"learning_rate": 2.9333333333333336e-05,
"loss": 1.0633,
"step": 220
},
{
"epoch": 2.02,
"eval_accuracy": 0.5483870967741935,
"eval_loss": 0.9655246734619141,
"eval_runtime": 2.0203,
"eval_samples_per_second": 15.344,
"eval_steps_per_second": 3.96,
"step": 225
},
{
"epoch": 3.0,
"learning_rate": 3.066666666666667e-05,
"loss": 0.7053,
"step": 230
},
{
"epoch": 3.0,
"learning_rate": 3.2000000000000005e-05,
"loss": 0.7653,
"step": 240
},
{
"epoch": 3.01,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.5797,
"step": 250
},
{
"epoch": 3.01,
"learning_rate": 3.466666666666667e-05,
"loss": 0.5341,
"step": 260
},
{
"epoch": 3.01,
"learning_rate": 3.6e-05,
"loss": 0.7323,
"step": 270
},
{
"epoch": 3.01,
"learning_rate": 3.733333333333334e-05,
"loss": 0.4958,
"step": 280
},
{
"epoch": 3.02,
"learning_rate": 3.866666666666667e-05,
"loss": 0.6096,
"step": 290
},
{
"epoch": 3.02,
"learning_rate": 4e-05,
"loss": 0.3783,
"step": 300
},
{
"epoch": 3.02,
"eval_accuracy": 0.7419354838709677,
"eval_loss": 0.5900937914848328,
"eval_runtime": 2.0722,
"eval_samples_per_second": 14.96,
"eval_steps_per_second": 3.861,
"step": 300
},
{
"epoch": 4.0,
"learning_rate": 4.133333333333333e-05,
"loss": 0.2128,
"step": 310
},
{
"epoch": 4.01,
"learning_rate": 4.266666666666667e-05,
"loss": 0.5407,
"step": 320
},
{
"epoch": 4.01,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.3279,
"step": 330
},
{
"epoch": 4.01,
"learning_rate": 4.5333333333333335e-05,
"loss": 0.4606,
"step": 340
},
{
"epoch": 4.01,
"learning_rate": 4.666666666666667e-05,
"loss": 0.4715,
"step": 350
},
{
"epoch": 4.02,
"learning_rate": 4.8e-05,
"loss": 0.3578,
"step": 360
},
{
"epoch": 4.02,
"learning_rate": 4.933333333333334e-05,
"loss": 0.756,
"step": 370
},
{
"epoch": 4.02,
"eval_accuracy": 0.6774193548387096,
"eval_loss": 0.9142343401908875,
"eval_runtime": 2.1263,
"eval_samples_per_second": 14.579,
"eval_steps_per_second": 3.762,
"step": 375
},
{
"epoch": 5.0,
"learning_rate": 4.9925925925925926e-05,
"loss": 0.2887,
"step": 380
},
{
"epoch": 5.0,
"learning_rate": 4.977777777777778e-05,
"loss": 1.0616,
"step": 390
},
{
"epoch": 5.01,
"learning_rate": 4.962962962962963e-05,
"loss": 0.2231,
"step": 400
},
{
"epoch": 5.01,
"learning_rate": 4.9481481481481485e-05,
"loss": 0.4401,
"step": 410
},
{
"epoch": 5.01,
"learning_rate": 4.933333333333334e-05,
"loss": 0.1261,
"step": 420
},
{
"epoch": 5.01,
"learning_rate": 4.918518518518519e-05,
"loss": 0.655,
"step": 430
},
{
"epoch": 5.02,
"learning_rate": 4.903703703703704e-05,
"loss": 0.4523,
"step": 440
},
{
"epoch": 5.02,
"learning_rate": 4.888888888888889e-05,
"loss": 0.5186,
"step": 450
},
{
"epoch": 5.02,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 0.7383989095687866,
"eval_runtime": 2.1403,
"eval_samples_per_second": 14.484,
"eval_steps_per_second": 3.738,
"step": 450
},
{
"epoch": 6.0,
"learning_rate": 4.874074074074074e-05,
"loss": 0.1326,
"step": 460
},
{
"epoch": 6.01,
"learning_rate": 4.8592592592592596e-05,
"loss": 0.0233,
"step": 470
},
{
"epoch": 6.01,
"learning_rate": 4.844444444444445e-05,
"loss": 0.3254,
"step": 480
},
{
"epoch": 6.01,
"learning_rate": 4.82962962962963e-05,
"loss": 0.2502,
"step": 490
},
{
"epoch": 6.01,
"learning_rate": 4.814814814814815e-05,
"loss": 0.1076,
"step": 500
},
{
"epoch": 6.02,
"learning_rate": 4.8e-05,
"loss": 0.5042,
"step": 510
},
{
"epoch": 6.02,
"learning_rate": 4.7851851851851854e-05,
"loss": 0.3714,
"step": 520
},
{
"epoch": 6.02,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 1.1661574840545654,
"eval_runtime": 2.1688,
"eval_samples_per_second": 14.293,
"eval_steps_per_second": 3.689,
"step": 525
},
{
"epoch": 7.0,
"learning_rate": 4.770370370370371e-05,
"loss": 0.8849,
"step": 530
},
{
"epoch": 7.0,
"learning_rate": 4.755555555555556e-05,
"loss": 0.3047,
"step": 540
},
{
"epoch": 7.01,
"learning_rate": 4.740740740740741e-05,
"loss": 0.4302,
"step": 550
},
{
"epoch": 7.01,
"learning_rate": 4.7259259259259266e-05,
"loss": 0.454,
"step": 560
},
{
"epoch": 7.01,
"learning_rate": 4.711111111111111e-05,
"loss": 0.1688,
"step": 570
},
{
"epoch": 7.01,
"learning_rate": 4.6962962962962966e-05,
"loss": 0.1852,
"step": 580
},
{
"epoch": 7.02,
"learning_rate": 4.681481481481482e-05,
"loss": 0.0478,
"step": 590
},
{
"epoch": 7.02,
"learning_rate": 4.666666666666667e-05,
"loss": 0.0263,
"step": 600
},
{
"epoch": 7.02,
"eval_accuracy": 0.8064516129032258,
"eval_loss": 0.910193681716919,
"eval_runtime": 2.227,
"eval_samples_per_second": 13.92,
"eval_steps_per_second": 3.592,
"step": 600
},
{
"epoch": 8.0,
"learning_rate": 4.6518518518518525e-05,
"loss": 0.0189,
"step": 610
},
{
"epoch": 8.01,
"learning_rate": 4.637037037037038e-05,
"loss": 0.0965,
"step": 620
},
{
"epoch": 8.01,
"learning_rate": 4.6222222222222224e-05,
"loss": 0.0689,
"step": 630
},
{
"epoch": 8.01,
"learning_rate": 4.607407407407408e-05,
"loss": 0.0171,
"step": 640
},
{
"epoch": 8.01,
"learning_rate": 4.592592592592593e-05,
"loss": 0.281,
"step": 650
},
{
"epoch": 8.02,
"learning_rate": 4.577777777777778e-05,
"loss": 0.2834,
"step": 660
},
{
"epoch": 8.02,
"learning_rate": 4.5629629629629636e-05,
"loss": 0.0848,
"step": 670
},
{
"epoch": 8.02,
"eval_accuracy": 0.9354838709677419,
"eval_loss": 0.12099169194698334,
"eval_runtime": 2.1891,
"eval_samples_per_second": 14.161,
"eval_steps_per_second": 3.654,
"step": 675
},
{
"epoch": 9.0,
"learning_rate": 4.548148148148149e-05,
"loss": 0.009,
"step": 680
},
{
"epoch": 9.0,
"learning_rate": 4.5333333333333335e-05,
"loss": 0.1009,
"step": 690
},
{
"epoch": 9.01,
"learning_rate": 4.518518518518519e-05,
"loss": 0.1219,
"step": 700
},
{
"epoch": 9.01,
"learning_rate": 4.503703703703704e-05,
"loss": 0.0463,
"step": 710
},
{
"epoch": 9.01,
"learning_rate": 4.4888888888888894e-05,
"loss": 0.0083,
"step": 720
},
{
"epoch": 9.01,
"learning_rate": 4.474074074074075e-05,
"loss": 0.0036,
"step": 730
},
{
"epoch": 9.02,
"learning_rate": 4.4592592592592594e-05,
"loss": 0.0295,
"step": 740
},
{
"epoch": 9.02,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.0028,
"step": 750
},
{
"epoch": 9.02,
"eval_accuracy": 0.9354838709677419,
"eval_loss": 0.2373807281255722,
"eval_runtime": 2.1663,
"eval_samples_per_second": 14.31,
"eval_steps_per_second": 3.693,
"step": 750
},
{
"epoch": 10.0,
"learning_rate": 4.42962962962963e-05,
"loss": 0.005,
"step": 760
},
{
"epoch": 10.01,
"learning_rate": 4.414814814814815e-05,
"loss": 0.0043,
"step": 770
},
{
"epoch": 10.01,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.0094,
"step": 780
},
{
"epoch": 10.01,
"learning_rate": 4.385185185185185e-05,
"loss": 0.0021,
"step": 790
},
{
"epoch": 10.01,
"learning_rate": 4.3703703703703705e-05,
"loss": 0.0051,
"step": 800
},
{
"epoch": 10.02,
"learning_rate": 4.355555555555556e-05,
"loss": 0.0024,
"step": 810
},
{
"epoch": 10.02,
"learning_rate": 4.340740740740741e-05,
"loss": 0.0863,
"step": 820
},
{
"epoch": 10.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.17276477813720703,
"eval_runtime": 2.132,
"eval_samples_per_second": 14.54,
"eval_steps_per_second": 3.752,
"step": 825
},
{
"epoch": 11.0,
"learning_rate": 4.325925925925926e-05,
"loss": 0.0026,
"step": 830
},
{
"epoch": 11.0,
"learning_rate": 4.311111111111111e-05,
"loss": 0.0022,
"step": 840
},
{
"epoch": 11.01,
"learning_rate": 4.296296296296296e-05,
"loss": 0.0023,
"step": 850
},
{
"epoch": 11.01,
"learning_rate": 4.2814814814814816e-05,
"loss": 0.002,
"step": 860
},
{
"epoch": 11.01,
"learning_rate": 4.266666666666667e-05,
"loss": 0.002,
"step": 870
},
{
"epoch": 11.01,
"learning_rate": 4.2518518518518515e-05,
"loss": 0.0033,
"step": 880
},
{
"epoch": 11.02,
"learning_rate": 4.237037037037037e-05,
"loss": 0.1644,
"step": 890
},
{
"epoch": 11.02,
"learning_rate": 4.222222222222222e-05,
"loss": 0.0018,
"step": 900
},
{
"epoch": 11.02,
"eval_accuracy": 0.9354838709677419,
"eval_loss": 0.3042505085468292,
"eval_runtime": 2.1005,
"eval_samples_per_second": 14.759,
"eval_steps_per_second": 3.809,
"step": 900
},
{
"epoch": 12.0,
"learning_rate": 4.2074074074074075e-05,
"loss": 0.002,
"step": 910
},
{
"epoch": 12.01,
"learning_rate": 4.192592592592593e-05,
"loss": 0.0018,
"step": 920
},
{
"epoch": 12.01,
"learning_rate": 4.177777777777778e-05,
"loss": 0.0043,
"step": 930
},
{
"epoch": 12.01,
"learning_rate": 4.162962962962963e-05,
"loss": 0.0028,
"step": 940
},
{
"epoch": 12.01,
"learning_rate": 4.148148148148148e-05,
"loss": 0.002,
"step": 950
},
{
"epoch": 12.02,
"learning_rate": 4.133333333333333e-05,
"loss": 0.0027,
"step": 960
},
{
"epoch": 12.02,
"learning_rate": 4.1185185185185186e-05,
"loss": 0.1662,
"step": 970
},
{
"epoch": 12.02,
"eval_accuracy": 0.9032258064516129,
"eval_loss": 0.42439064383506775,
"eval_runtime": 2.1089,
"eval_samples_per_second": 14.699,
"eval_steps_per_second": 3.793,
"step": 975
},
{
"epoch": 13.0,
"learning_rate": 4.103703703703704e-05,
"loss": 0.0018,
"step": 980
},
{
"epoch": 13.0,
"learning_rate": 4.088888888888889e-05,
"loss": 0.041,
"step": 990
},
{
"epoch": 13.01,
"learning_rate": 4.074074074074074e-05,
"loss": 0.0018,
"step": 1000
},
{
"epoch": 13.01,
"learning_rate": 4.059259259259259e-05,
"loss": 0.0107,
"step": 1010
},
{
"epoch": 13.01,
"learning_rate": 4.0444444444444444e-05,
"loss": 0.0019,
"step": 1020
},
{
"epoch": 13.01,
"learning_rate": 4.02962962962963e-05,
"loss": 0.0016,
"step": 1030
},
{
"epoch": 13.02,
"learning_rate": 4.014814814814815e-05,
"loss": 0.0016,
"step": 1040
},
{
"epoch": 13.02,
"learning_rate": 4e-05,
"loss": 0.0132,
"step": 1050
},
{
"epoch": 13.02,
"eval_accuracy": 0.8709677419354839,
"eval_loss": 0.6965717673301697,
"eval_runtime": 2.0555,
"eval_samples_per_second": 15.082,
"eval_steps_per_second": 3.892,
"step": 1050
},
{
"epoch": 14.0,
"learning_rate": 3.985185185185185e-05,
"loss": 0.0015,
"step": 1060
},
{
"epoch": 14.01,
"learning_rate": 3.97037037037037e-05,
"loss": 0.0444,
"step": 1070
},
{
"epoch": 14.01,
"learning_rate": 3.9555555555555556e-05,
"loss": 0.0015,
"step": 1080
},
{
"epoch": 14.01,
"learning_rate": 3.940740740740741e-05,
"loss": 0.0015,
"step": 1090
},
{
"epoch": 14.01,
"learning_rate": 3.925925925925926e-05,
"loss": 0.0014,
"step": 1100
},
{
"epoch": 14.02,
"learning_rate": 3.9111111111111115e-05,
"loss": 0.0013,
"step": 1110
},
{
"epoch": 14.02,
"learning_rate": 3.896296296296296e-05,
"loss": 0.0019,
"step": 1120
},
{
"epoch": 14.02,
"eval_accuracy": 0.8709677419354839,
"eval_loss": 0.5602208971977234,
"eval_runtime": 2.0688,
"eval_samples_per_second": 14.985,
"eval_steps_per_second": 3.867,
"step": 1125
},
{
"epoch": 15.0,
"learning_rate": 3.8814814814814814e-05,
"loss": 0.0013,
"step": 1130
},
{
"epoch": 15.0,
"learning_rate": 3.866666666666667e-05,
"loss": 0.2104,
"step": 1140
},
{
"epoch": 15.01,
"learning_rate": 3.851851851851852e-05,
"loss": 0.0014,
"step": 1150
},
{
"epoch": 15.01,
"learning_rate": 3.837037037037037e-05,
"loss": 0.0015,
"step": 1160
},
{
"epoch": 15.01,
"learning_rate": 3.8222222222222226e-05,
"loss": 0.0011,
"step": 1170
},
{
"epoch": 15.01,
"learning_rate": 3.807407407407408e-05,
"loss": 0.0013,
"step": 1180
},
{
"epoch": 15.02,
"learning_rate": 3.7925925925925925e-05,
"loss": 0.0011,
"step": 1190
},
{
"epoch": 15.02,
"learning_rate": 3.777777777777778e-05,
"loss": 0.0012,
"step": 1200
},
{
"epoch": 15.02,
"eval_accuracy": 0.8709677419354839,
"eval_loss": 0.26489391922950745,
"eval_runtime": 2.1064,
"eval_samples_per_second": 14.717,
"eval_steps_per_second": 3.798,
"step": 1200
},
{
"epoch": 16.0,
"learning_rate": 3.762962962962963e-05,
"loss": 0.0012,
"step": 1210
},
{
"epoch": 16.01,
"learning_rate": 3.7481481481481484e-05,
"loss": 0.0014,
"step": 1220
},
{
"epoch": 16.01,
"learning_rate": 3.733333333333334e-05,
"loss": 0.0012,
"step": 1230
},
{
"epoch": 16.01,
"learning_rate": 3.718518518518519e-05,
"loss": 0.0011,
"step": 1240
},
{
"epoch": 16.01,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.0013,
"step": 1250
},
{
"epoch": 16.02,
"learning_rate": 3.688888888888889e-05,
"loss": 0.003,
"step": 1260
},
{
"epoch": 16.02,
"learning_rate": 3.674074074074074e-05,
"loss": 0.0061,
"step": 1270
},
{
"epoch": 16.02,
"eval_accuracy": 0.8709677419354839,
"eval_loss": 0.7361043095588684,
"eval_runtime": 2.0949,
"eval_samples_per_second": 14.798,
"eval_steps_per_second": 3.819,
"step": 1275
},
{
"epoch": 17.0,
"learning_rate": 3.6592592592592596e-05,
"loss": 0.1429,
"step": 1280
},
{
"epoch": 17.0,
"learning_rate": 3.644444444444445e-05,
"loss": 0.0012,
"step": 1290
},
{
"epoch": 17.01,
"learning_rate": 3.62962962962963e-05,
"loss": 0.0011,
"step": 1300
},
{
"epoch": 17.01,
"learning_rate": 3.614814814814815e-05,
"loss": 0.2073,
"step": 1310
},
{
"epoch": 17.01,
"learning_rate": 3.6e-05,
"loss": 0.0425,
"step": 1320
},
{
"epoch": 17.01,
"learning_rate": 3.5851851851851854e-05,
"loss": 0.0014,
"step": 1330
},
{
"epoch": 17.02,
"learning_rate": 3.570370370370371e-05,
"loss": 0.0016,
"step": 1340
},
{
"epoch": 17.02,
"learning_rate": 3.555555555555556e-05,
"loss": 0.0012,
"step": 1350
},
{
"epoch": 17.02,
"eval_accuracy": 0.9354838709677419,
"eval_loss": 0.2821272313594818,
"eval_runtime": 2.1139,
"eval_samples_per_second": 14.665,
"eval_steps_per_second": 3.784,
"step": 1350
},
{
"epoch": 18.0,
"learning_rate": 3.540740740740741e-05,
"loss": 0.0011,
"step": 1360
},
{
"epoch": 18.01,
"learning_rate": 3.525925925925926e-05,
"loss": 0.0951,
"step": 1370
},
{
"epoch": 18.01,
"learning_rate": 3.511111111111111e-05,
"loss": 0.0011,
"step": 1380
},
{
"epoch": 18.01,
"learning_rate": 3.4962962962962965e-05,
"loss": 0.0013,
"step": 1390
},
{
"epoch": 18.01,
"learning_rate": 3.481481481481482e-05,
"loss": 0.001,
"step": 1400
},
{
"epoch": 18.02,
"learning_rate": 3.466666666666667e-05,
"loss": 0.0094,
"step": 1410
},
{
"epoch": 18.02,
"learning_rate": 3.4518518518518524e-05,
"loss": 0.001,
"step": 1420
},
{
"epoch": 18.02,
"eval_accuracy": 0.9354838709677419,
"eval_loss": 0.3117120563983917,
"eval_runtime": 2.0573,
"eval_samples_per_second": 15.068,
"eval_steps_per_second": 3.889,
"step": 1425
},
{
"epoch": 19.0,
"learning_rate": 3.437037037037037e-05,
"loss": 0.001,
"step": 1430
},
{
"epoch": 19.0,
"learning_rate": 3.4222222222222224e-05,
"loss": 0.001,
"step": 1440
},
{
"epoch": 19.01,
"learning_rate": 3.4074074074074077e-05,
"loss": 0.0147,
"step": 1450
},
{
"epoch": 19.01,
"learning_rate": 3.392592592592593e-05,
"loss": 0.0013,
"step": 1460
},
{
"epoch": 19.01,
"learning_rate": 3.377777777777778e-05,
"loss": 0.063,
"step": 1470
},
{
"epoch": 19.01,
"learning_rate": 3.3629629629629636e-05,
"loss": 0.1805,
"step": 1480
},
{
"epoch": 19.02,
"learning_rate": 3.348148148148148e-05,
"loss": 0.1541,
"step": 1490
},
{
"epoch": 19.02,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.001,
"step": 1500
},
{
"epoch": 19.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.11934669315814972,
"eval_runtime": 2.125,
"eval_samples_per_second": 14.588,
"eval_steps_per_second": 3.765,
"step": 1500
},
{
"epoch": 20.0,
"learning_rate": 3.318518518518519e-05,
"loss": 0.0022,
"step": 1510
},
{
"epoch": 20.01,
"learning_rate": 3.303703703703704e-05,
"loss": 0.0009,
"step": 1520
},
{
"epoch": 20.01,
"learning_rate": 3.2888888888888894e-05,
"loss": 0.0016,
"step": 1530
},
{
"epoch": 20.01,
"learning_rate": 3.274074074074075e-05,
"loss": 0.1235,
"step": 1540
},
{
"epoch": 20.01,
"learning_rate": 3.25925925925926e-05,
"loss": 0.0055,
"step": 1550
},
{
"epoch": 20.02,
"learning_rate": 3.2444444444444446e-05,
"loss": 0.001,
"step": 1560
},
{
"epoch": 20.02,
"learning_rate": 3.22962962962963e-05,
"loss": 0.0021,
"step": 1570
},
{
"epoch": 20.02,
"eval_accuracy": 0.9354838709677419,
"eval_loss": 0.24128369987010956,
"eval_runtime": 2.0867,
"eval_samples_per_second": 14.856,
"eval_steps_per_second": 3.834,
"step": 1575
},
{
"epoch": 21.0,
"learning_rate": 3.214814814814815e-05,
"loss": 0.001,
"step": 1580
},
{
"epoch": 21.0,
"learning_rate": 3.2000000000000005e-05,
"loss": 0.0009,
"step": 1590
},
{
"epoch": 21.01,
"learning_rate": 3.185185185185185e-05,
"loss": 0.0012,
"step": 1600
},
{
"epoch": 21.01,
"learning_rate": 3.1703703703703705e-05,
"loss": 0.001,
"step": 1610
},
{
"epoch": 21.01,
"learning_rate": 3.155555555555556e-05,
"loss": 0.0009,
"step": 1620
},
{
"epoch": 21.01,
"learning_rate": 3.140740740740741e-05,
"loss": 0.0008,
"step": 1630
},
{
"epoch": 21.02,
"learning_rate": 3.1259259259259264e-05,
"loss": 0.0009,
"step": 1640
},
{
"epoch": 21.02,
"learning_rate": 3.111111111111111e-05,
"loss": 0.0009,
"step": 1650
},
{
"epoch": 21.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1641165167093277,
"eval_runtime": 2.0769,
"eval_samples_per_second": 14.926,
"eval_steps_per_second": 3.852,
"step": 1650
},
{
"epoch": 22.0,
"learning_rate": 3.096296296296296e-05,
"loss": 0.0009,
"step": 1660
},
{
"epoch": 22.01,
"learning_rate": 3.0814814814814816e-05,
"loss": 0.0289,
"step": 1670
},
{
"epoch": 22.01,
"learning_rate": 3.066666666666667e-05,
"loss": 0.0008,
"step": 1680
},
{
"epoch": 22.01,
"learning_rate": 3.0518518518518515e-05,
"loss": 0.0008,
"step": 1690
},
{
"epoch": 22.01,
"learning_rate": 3.037037037037037e-05,
"loss": 0.0008,
"step": 1700
},
{
"epoch": 22.02,
"learning_rate": 3.0222222222222225e-05,
"loss": 0.0008,
"step": 1710
},
{
"epoch": 22.02,
"learning_rate": 3.0074074074074078e-05,
"loss": 0.0016,
"step": 1720
},
{
"epoch": 22.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1332567036151886,
"eval_runtime": 2.0589,
"eval_samples_per_second": 15.057,
"eval_steps_per_second": 3.886,
"step": 1725
},
{
"epoch": 23.0,
"learning_rate": 2.992592592592593e-05,
"loss": 0.0008,
"step": 1730
},
{
"epoch": 23.0,
"learning_rate": 2.9777777777777777e-05,
"loss": 0.0008,
"step": 1740
},
{
"epoch": 23.01,
"learning_rate": 2.962962962962963e-05,
"loss": 0.0009,
"step": 1750
},
{
"epoch": 23.01,
"learning_rate": 2.9481481481481483e-05,
"loss": 0.0008,
"step": 1760
},
{
"epoch": 23.01,
"learning_rate": 2.9333333333333336e-05,
"loss": 0.0009,
"step": 1770
},
{
"epoch": 23.01,
"learning_rate": 2.918518518518519e-05,
"loss": 0.0007,
"step": 1780
},
{
"epoch": 23.02,
"learning_rate": 2.9037037037037042e-05,
"loss": 0.0007,
"step": 1790
},
{
"epoch": 23.02,
"learning_rate": 2.8888888888888888e-05,
"loss": 0.0007,
"step": 1800
},
{
"epoch": 23.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.10598693042993546,
"eval_runtime": 2.0414,
"eval_samples_per_second": 15.186,
"eval_steps_per_second": 3.919,
"step": 1800
},
{
"epoch": 24.0,
"learning_rate": 2.874074074074074e-05,
"loss": 0.0007,
"step": 1810
},
{
"epoch": 24.01,
"learning_rate": 2.8592592592592594e-05,
"loss": 0.0007,
"step": 1820
},
{
"epoch": 24.01,
"learning_rate": 2.8444444444444447e-05,
"loss": 0.0007,
"step": 1830
},
{
"epoch": 24.01,
"learning_rate": 2.8296296296296297e-05,
"loss": 0.0007,
"step": 1840
},
{
"epoch": 24.01,
"learning_rate": 2.814814814814815e-05,
"loss": 0.0007,
"step": 1850
},
{
"epoch": 24.02,
"learning_rate": 2.8000000000000003e-05,
"loss": 0.0006,
"step": 1860
},
{
"epoch": 24.02,
"learning_rate": 2.7851851851851853e-05,
"loss": 0.0008,
"step": 1870
},
{
"epoch": 24.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1111554503440857,
"eval_runtime": 2.0546,
"eval_samples_per_second": 15.088,
"eval_steps_per_second": 3.894,
"step": 1875
},
{
"epoch": 25.0,
"learning_rate": 2.7703703703703706e-05,
"loss": 0.0007,
"step": 1880
},
{
"epoch": 25.0,
"learning_rate": 2.7555555555555555e-05,
"loss": 0.0007,
"step": 1890
},
{
"epoch": 25.01,
"learning_rate": 2.7407407407407408e-05,
"loss": 0.0007,
"step": 1900
},
{
"epoch": 25.01,
"learning_rate": 2.725925925925926e-05,
"loss": 0.0007,
"step": 1910
},
{
"epoch": 25.01,
"learning_rate": 2.7111111111111114e-05,
"loss": 0.0006,
"step": 1920
},
{
"epoch": 25.01,
"learning_rate": 2.696296296296296e-05,
"loss": 0.0006,
"step": 1930
},
{
"epoch": 25.02,
"learning_rate": 2.6814814814814814e-05,
"loss": 0.0006,
"step": 1940
},
{
"epoch": 25.02,
"learning_rate": 2.6666666666666667e-05,
"loss": 0.0006,
"step": 1950
},
{
"epoch": 25.02,
"eval_accuracy": 1.0,
"eval_loss": 0.01604604721069336,
"eval_runtime": 2.0655,
"eval_samples_per_second": 15.008,
"eval_steps_per_second": 3.873,
"step": 1950
},
{
"epoch": 26.0,
"learning_rate": 2.651851851851852e-05,
"loss": 0.0006,
"step": 1960
},
{
"epoch": 26.01,
"learning_rate": 2.6370370370370373e-05,
"loss": 0.0006,
"step": 1970
},
{
"epoch": 26.01,
"learning_rate": 2.6222222222222226e-05,
"loss": 0.0006,
"step": 1980
},
{
"epoch": 26.01,
"learning_rate": 2.6074074074074072e-05,
"loss": 0.0006,
"step": 1990
},
{
"epoch": 26.01,
"learning_rate": 2.5925925925925925e-05,
"loss": 0.0006,
"step": 2000
},
{
"epoch": 26.02,
"learning_rate": 2.5777777777777778e-05,
"loss": 0.0006,
"step": 2010
},
{
"epoch": 26.02,
"learning_rate": 2.562962962962963e-05,
"loss": 0.0006,
"step": 2020
},
{
"epoch": 26.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.029128754511475563,
"eval_runtime": 2.1074,
"eval_samples_per_second": 14.71,
"eval_steps_per_second": 3.796,
"step": 2025
},
{
"epoch": 27.0,
"learning_rate": 2.5481481481481484e-05,
"loss": 0.0006,
"step": 2030
},
{
"epoch": 27.0,
"learning_rate": 2.5333333333333337e-05,
"loss": 0.0006,
"step": 2040
},
{
"epoch": 27.01,
"learning_rate": 2.5185185185185183e-05,
"loss": 0.0006,
"step": 2050
},
{
"epoch": 27.01,
"learning_rate": 2.5037037037037036e-05,
"loss": 0.0007,
"step": 2060
},
{
"epoch": 27.01,
"learning_rate": 2.488888888888889e-05,
"loss": 0.0006,
"step": 2070
},
{
"epoch": 27.01,
"learning_rate": 2.4740740740740742e-05,
"loss": 0.0006,
"step": 2080
},
{
"epoch": 27.02,
"learning_rate": 2.4592592592592595e-05,
"loss": 0.0006,
"step": 2090
},
{
"epoch": 27.02,
"learning_rate": 2.4444444444444445e-05,
"loss": 0.0006,
"step": 2100
},
{
"epoch": 27.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.08601882308721542,
"eval_runtime": 2.1685,
"eval_samples_per_second": 14.296,
"eval_steps_per_second": 3.689,
"step": 2100
},
{
"epoch": 28.0,
"learning_rate": 2.4296296296296298e-05,
"loss": 0.0006,
"step": 2110
},
{
"epoch": 28.01,
"learning_rate": 2.414814814814815e-05,
"loss": 0.0006,
"step": 2120
},
{
"epoch": 28.01,
"learning_rate": 2.4e-05,
"loss": 0.0006,
"step": 2130
},
{
"epoch": 28.01,
"learning_rate": 2.3851851851851854e-05,
"loss": 0.0006,
"step": 2140
},
{
"epoch": 28.01,
"learning_rate": 2.3703703703703707e-05,
"loss": 0.0006,
"step": 2150
},
{
"epoch": 28.02,
"learning_rate": 2.3555555555555556e-05,
"loss": 0.0006,
"step": 2160
},
{
"epoch": 28.02,
"learning_rate": 2.340740740740741e-05,
"loss": 0.0005,
"step": 2170
},
{
"epoch": 28.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.10801166296005249,
"eval_runtime": 2.1529,
"eval_samples_per_second": 14.399,
"eval_steps_per_second": 3.716,
"step": 2175
},
{
"epoch": 29.0,
"learning_rate": 2.3259259259259262e-05,
"loss": 0.0006,
"step": 2180
},
{
"epoch": 29.0,
"learning_rate": 2.3111111111111112e-05,
"loss": 0.0005,
"step": 2190
},
{
"epoch": 29.01,
"learning_rate": 2.2962962962962965e-05,
"loss": 0.0006,
"step": 2200
},
{
"epoch": 29.01,
"learning_rate": 2.2814814814814818e-05,
"loss": 0.0006,
"step": 2210
},
{
"epoch": 29.01,
"learning_rate": 2.2666666666666668e-05,
"loss": 0.0005,
"step": 2220
},
{
"epoch": 29.01,
"learning_rate": 2.251851851851852e-05,
"loss": 0.0006,
"step": 2230
},
{
"epoch": 29.02,
"learning_rate": 2.2370370370370374e-05,
"loss": 0.0005,
"step": 2240
},
{
"epoch": 29.02,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.0006,
"step": 2250
},
{
"epoch": 29.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.11195220053195953,
"eval_runtime": 2.1574,
"eval_samples_per_second": 14.369,
"eval_steps_per_second": 3.708,
"step": 2250
},
{
"epoch": 30.0,
"learning_rate": 2.2074074074074076e-05,
"loss": 0.0005,
"step": 2260
},
{
"epoch": 30.01,
"learning_rate": 2.1925925925925926e-05,
"loss": 0.0006,
"step": 2270
},
{
"epoch": 30.01,
"learning_rate": 2.177777777777778e-05,
"loss": 0.1392,
"step": 2280
},
{
"epoch": 30.01,
"learning_rate": 2.162962962962963e-05,
"loss": 0.0005,
"step": 2290
},
{
"epoch": 30.01,
"learning_rate": 2.148148148148148e-05,
"loss": 0.0564,
"step": 2300
},
{
"epoch": 30.02,
"learning_rate": 2.1333333333333335e-05,
"loss": 0.0006,
"step": 2310
},
{
"epoch": 30.02,
"learning_rate": 2.1185185185185184e-05,
"loss": 0.0006,
"step": 2320
},
{
"epoch": 30.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.05927049741148949,
"eval_runtime": 2.1955,
"eval_samples_per_second": 14.12,
"eval_steps_per_second": 3.644,
"step": 2325
},
{
"epoch": 31.0,
"learning_rate": 2.1037037037037037e-05,
"loss": 0.0007,
"step": 2330
},
{
"epoch": 31.0,
"learning_rate": 2.088888888888889e-05,
"loss": 0.0379,
"step": 2340
},
{
"epoch": 31.01,
"learning_rate": 2.074074074074074e-05,
"loss": 0.0006,
"step": 2350
},
{
"epoch": 31.01,
"learning_rate": 2.0592592592592593e-05,
"loss": 0.0005,
"step": 2360
},
{
"epoch": 31.01,
"learning_rate": 2.0444444444444446e-05,
"loss": 0.0006,
"step": 2370
},
{
"epoch": 31.01,
"learning_rate": 2.0296296296296296e-05,
"loss": 0.0005,
"step": 2380
},
{
"epoch": 31.02,
"learning_rate": 2.014814814814815e-05,
"loss": 0.0005,
"step": 2390
},
{
"epoch": 31.02,
"learning_rate": 2e-05,
"loss": 0.0005,
"step": 2400
},
{
"epoch": 31.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.16602446138858795,
"eval_runtime": 2.1541,
"eval_samples_per_second": 14.391,
"eval_steps_per_second": 3.714,
"step": 2400
},
{
"epoch": 32.0,
"learning_rate": 1.985185185185185e-05,
"loss": 0.0005,
"step": 2410
},
{
"epoch": 32.01,
"learning_rate": 1.9703703703703704e-05,
"loss": 0.0005,
"step": 2420
},
{
"epoch": 32.01,
"learning_rate": 1.9555555555555557e-05,
"loss": 0.0007,
"step": 2430
},
{
"epoch": 32.01,
"learning_rate": 1.9407407407407407e-05,
"loss": 0.0005,
"step": 2440
},
{
"epoch": 32.01,
"learning_rate": 1.925925925925926e-05,
"loss": 0.0005,
"step": 2450
},
{
"epoch": 32.02,
"learning_rate": 1.9111111111111113e-05,
"loss": 0.0005,
"step": 2460
},
{
"epoch": 32.02,
"learning_rate": 1.8962962962962963e-05,
"loss": 0.0005,
"step": 2470
},
{
"epoch": 32.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.04543553665280342,
"eval_runtime": 2.5566,
"eval_samples_per_second": 12.125,
"eval_steps_per_second": 3.129,
"step": 2475
},
{
"epoch": 33.0,
"learning_rate": 1.8814814814814816e-05,
"loss": 0.0005,
"step": 2480
},
{
"epoch": 33.0,
"learning_rate": 1.866666666666667e-05,
"loss": 0.0006,
"step": 2490
},
{
"epoch": 33.01,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.0005,
"step": 2500
},
{
"epoch": 33.01,
"learning_rate": 1.837037037037037e-05,
"loss": 0.0005,
"step": 2510
},
{
"epoch": 33.01,
"learning_rate": 1.8222222222222224e-05,
"loss": 0.0009,
"step": 2520
},
{
"epoch": 33.01,
"learning_rate": 1.8074074074074074e-05,
"loss": 0.0005,
"step": 2530
},
{
"epoch": 33.02,
"learning_rate": 1.7925925925925927e-05,
"loss": 0.0005,
"step": 2540
},
{
"epoch": 33.02,
"learning_rate": 1.777777777777778e-05,
"loss": 0.0007,
"step": 2550
},
{
"epoch": 33.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1014844998717308,
"eval_runtime": 2.1217,
"eval_samples_per_second": 14.611,
"eval_steps_per_second": 3.771,
"step": 2550
},
{
"epoch": 34.0,
"learning_rate": 1.762962962962963e-05,
"loss": 0.0005,
"step": 2560
},
{
"epoch": 34.01,
"learning_rate": 1.7481481481481483e-05,
"loss": 0.0005,
"step": 2570
},
{
"epoch": 34.01,
"learning_rate": 1.7333333333333336e-05,
"loss": 0.0005,
"step": 2580
},
{
"epoch": 34.01,
"learning_rate": 1.7185185185185185e-05,
"loss": 0.0005,
"step": 2590
},
{
"epoch": 34.01,
"learning_rate": 1.7037037037037038e-05,
"loss": 0.0005,
"step": 2600
},
{
"epoch": 34.02,
"learning_rate": 1.688888888888889e-05,
"loss": 0.0005,
"step": 2610
},
{
"epoch": 34.02,
"learning_rate": 1.674074074074074e-05,
"loss": 0.0005,
"step": 2620
},
{
"epoch": 34.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.17122069001197815,
"eval_runtime": 2.1289,
"eval_samples_per_second": 14.561,
"eval_steps_per_second": 3.758,
"step": 2625
},
{
"epoch": 35.0,
"learning_rate": 1.6592592592592594e-05,
"loss": 0.0005,
"step": 2630
},
{
"epoch": 35.0,
"learning_rate": 1.6444444444444447e-05,
"loss": 0.0005,
"step": 2640
},
{
"epoch": 35.01,
"learning_rate": 1.62962962962963e-05,
"loss": 0.0005,
"step": 2650
},
{
"epoch": 35.01,
"learning_rate": 1.614814814814815e-05,
"loss": 0.0005,
"step": 2660
},
{
"epoch": 35.01,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.0005,
"step": 2670
},
{
"epoch": 35.01,
"learning_rate": 1.5851851851851852e-05,
"loss": 0.0005,
"step": 2680
},
{
"epoch": 35.02,
"learning_rate": 1.5703703703703705e-05,
"loss": 0.0004,
"step": 2690
},
{
"epoch": 35.02,
"learning_rate": 1.5555555555555555e-05,
"loss": 0.0005,
"step": 2700
},
{
"epoch": 35.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1599966138601303,
"eval_runtime": 2.0848,
"eval_samples_per_second": 14.87,
"eval_steps_per_second": 3.837,
"step": 2700
},
{
"epoch": 36.0,
"learning_rate": 1.5407407407407408e-05,
"loss": 0.0004,
"step": 2710
},
{
"epoch": 36.01,
"learning_rate": 1.5259259259259258e-05,
"loss": 0.0005,
"step": 2720
},
{
"epoch": 36.01,
"learning_rate": 1.5111111111111112e-05,
"loss": 0.0005,
"step": 2730
},
{
"epoch": 36.01,
"learning_rate": 1.4962962962962965e-05,
"loss": 0.0005,
"step": 2740
},
{
"epoch": 36.01,
"learning_rate": 1.4814814814814815e-05,
"loss": 0.0005,
"step": 2750
},
{
"epoch": 36.02,
"learning_rate": 1.4666666666666668e-05,
"loss": 0.0004,
"step": 2760
},
{
"epoch": 36.02,
"learning_rate": 1.4518518518518521e-05,
"loss": 0.0004,
"step": 2770
},
{
"epoch": 36.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.16178081929683685,
"eval_runtime": 2.0795,
"eval_samples_per_second": 14.908,
"eval_steps_per_second": 3.847,
"step": 2775
},
{
"epoch": 37.0,
"learning_rate": 1.437037037037037e-05,
"loss": 0.0004,
"step": 2780
},
{
"epoch": 37.0,
"learning_rate": 1.4222222222222224e-05,
"loss": 0.0005,
"step": 2790
},
{
"epoch": 37.01,
"learning_rate": 1.4074074074074075e-05,
"loss": 0.0004,
"step": 2800
},
{
"epoch": 37.01,
"learning_rate": 1.3925925925925926e-05,
"loss": 0.0004,
"step": 2810
},
{
"epoch": 37.01,
"learning_rate": 1.3777777777777778e-05,
"loss": 0.0004,
"step": 2820
},
{
"epoch": 37.01,
"learning_rate": 1.362962962962963e-05,
"loss": 0.0006,
"step": 2830
},
{
"epoch": 37.02,
"learning_rate": 1.348148148148148e-05,
"loss": 0.0004,
"step": 2840
},
{
"epoch": 37.02,
"learning_rate": 1.3333333333333333e-05,
"loss": 0.0004,
"step": 2850
},
{
"epoch": 37.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.14675143361091614,
"eval_runtime": 2.0875,
"eval_samples_per_second": 14.851,
"eval_steps_per_second": 3.832,
"step": 2850
},
{
"epoch": 38.0,
"learning_rate": 1.3185185185185186e-05,
"loss": 0.0004,
"step": 2860
},
{
"epoch": 38.01,
"learning_rate": 1.3037037037037036e-05,
"loss": 0.0005,
"step": 2870
},
{
"epoch": 38.01,
"learning_rate": 1.2888888888888889e-05,
"loss": 0.0004,
"step": 2880
},
{
"epoch": 38.01,
"learning_rate": 1.2740740740740742e-05,
"loss": 0.0004,
"step": 2890
},
{
"epoch": 38.01,
"learning_rate": 1.2592592592592592e-05,
"loss": 0.0005,
"step": 2900
},
{
"epoch": 38.02,
"learning_rate": 1.2444444444444445e-05,
"loss": 0.0004,
"step": 2910
},
{
"epoch": 38.02,
"learning_rate": 1.2296296296296298e-05,
"loss": 0.0004,
"step": 2920
},
{
"epoch": 38.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.11672930419445038,
"eval_runtime": 2.0961,
"eval_samples_per_second": 14.79,
"eval_steps_per_second": 3.817,
"step": 2925
},
{
"epoch": 39.0,
"learning_rate": 1.2148148148148149e-05,
"loss": 0.0004,
"step": 2930
},
{
"epoch": 39.0,
"learning_rate": 1.2e-05,
"loss": 0.0004,
"step": 2940
},
{
"epoch": 39.01,
"learning_rate": 1.1851851851851853e-05,
"loss": 0.0004,
"step": 2950
},
{
"epoch": 39.01,
"learning_rate": 1.1703703703703705e-05,
"loss": 0.0004,
"step": 2960
},
{
"epoch": 39.01,
"learning_rate": 1.1555555555555556e-05,
"loss": 0.0004,
"step": 2970
},
{
"epoch": 39.01,
"learning_rate": 1.1407407407407409e-05,
"loss": 0.0004,
"step": 2980
},
{
"epoch": 39.02,
"learning_rate": 1.125925925925926e-05,
"loss": 0.0004,
"step": 2990
},
{
"epoch": 39.02,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.0004,
"step": 3000
},
{
"epoch": 39.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.12776704132556915,
"eval_runtime": 2.1378,
"eval_samples_per_second": 14.501,
"eval_steps_per_second": 3.742,
"step": 3000
},
{
"epoch": 40.0,
"learning_rate": 1.0962962962962963e-05,
"loss": 0.0004,
"step": 3010
},
{
"epoch": 40.01,
"learning_rate": 1.0814814814814814e-05,
"loss": 0.0004,
"step": 3020
},
{
"epoch": 40.01,
"learning_rate": 1.0666666666666667e-05,
"loss": 0.0004,
"step": 3030
},
{
"epoch": 40.01,
"learning_rate": 1.0518518518518519e-05,
"loss": 0.0004,
"step": 3040
},
{
"epoch": 40.01,
"learning_rate": 1.037037037037037e-05,
"loss": 0.0004,
"step": 3050
},
{
"epoch": 40.02,
"learning_rate": 1.0222222222222223e-05,
"loss": 0.0004,
"step": 3060
},
{
"epoch": 40.02,
"learning_rate": 1.0074074074074074e-05,
"loss": 0.0004,
"step": 3070
},
{
"epoch": 40.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.12002281844615936,
"eval_runtime": 2.1121,
"eval_samples_per_second": 14.677,
"eval_steps_per_second": 3.788,
"step": 3075
},
{
"epoch": 41.0,
"learning_rate": 9.925925925925926e-06,
"loss": 0.0004,
"step": 3080
},
{
"epoch": 41.0,
"learning_rate": 9.777777777777779e-06,
"loss": 0.0004,
"step": 3090
},
{
"epoch": 41.01,
"learning_rate": 9.62962962962963e-06,
"loss": 0.0004,
"step": 3100
},
{
"epoch": 41.01,
"learning_rate": 9.481481481481481e-06,
"loss": 0.0004,
"step": 3110
},
{
"epoch": 41.01,
"learning_rate": 9.333333333333334e-06,
"loss": 0.0004,
"step": 3120
},
{
"epoch": 41.01,
"learning_rate": 9.185185185185186e-06,
"loss": 0.0004,
"step": 3130
},
{
"epoch": 41.02,
"learning_rate": 9.037037037037037e-06,
"loss": 0.0004,
"step": 3140
},
{
"epoch": 41.02,
"learning_rate": 8.88888888888889e-06,
"loss": 0.0004,
"step": 3150
},
{
"epoch": 41.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.12003983557224274,
"eval_runtime": 2.0588,
"eval_samples_per_second": 15.057,
"eval_steps_per_second": 3.886,
"step": 3150
},
{
"epoch": 42.0,
"learning_rate": 8.740740740740741e-06,
"loss": 0.0004,
"step": 3160
},
{
"epoch": 42.01,
"learning_rate": 8.592592592592593e-06,
"loss": 0.0004,
"step": 3170
},
{
"epoch": 42.01,
"learning_rate": 8.444444444444446e-06,
"loss": 0.0004,
"step": 3180
},
{
"epoch": 42.01,
"learning_rate": 8.296296296296297e-06,
"loss": 0.0004,
"step": 3190
},
{
"epoch": 42.01,
"learning_rate": 8.14814814814815e-06,
"loss": 0.0004,
"step": 3200
},
{
"epoch": 42.02,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0004,
"step": 3210
},
{
"epoch": 42.02,
"learning_rate": 7.851851851851853e-06,
"loss": 0.0004,
"step": 3220
},
{
"epoch": 42.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.12297246605157852,
"eval_runtime": 2.0736,
"eval_samples_per_second": 14.95,
"eval_steps_per_second": 3.858,
"step": 3225
},
{
"epoch": 43.0,
"learning_rate": 7.703703703703704e-06,
"loss": 0.0004,
"step": 3230
},
{
"epoch": 43.0,
"learning_rate": 7.555555555555556e-06,
"loss": 0.0004,
"step": 3240
},
{
"epoch": 43.01,
"learning_rate": 7.4074074074074075e-06,
"loss": 0.0004,
"step": 3250
},
{
"epoch": 43.01,
"learning_rate": 7.2592592592592605e-06,
"loss": 0.0004,
"step": 3260
},
{
"epoch": 43.01,
"learning_rate": 7.111111111111112e-06,
"loss": 0.0004,
"step": 3270
},
{
"epoch": 43.01,
"learning_rate": 6.962962962962963e-06,
"loss": 0.0004,
"step": 3280
},
{
"epoch": 43.02,
"learning_rate": 6.814814814814815e-06,
"loss": 0.0004,
"step": 3290
},
{
"epoch": 43.02,
"learning_rate": 6.666666666666667e-06,
"loss": 0.0004,
"step": 3300
},
{
"epoch": 43.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.13225598633289337,
"eval_runtime": 2.0961,
"eval_samples_per_second": 14.789,
"eval_steps_per_second": 3.817,
"step": 3300
},
{
"epoch": 44.0,
"learning_rate": 6.518518518518518e-06,
"loss": 0.0004,
"step": 3310
},
{
"epoch": 44.01,
"learning_rate": 6.370370370370371e-06,
"loss": 0.0004,
"step": 3320
},
{
"epoch": 44.01,
"learning_rate": 6.222222222222222e-06,
"loss": 0.0004,
"step": 3330
},
{
"epoch": 44.01,
"learning_rate": 6.0740740740740745e-06,
"loss": 0.0004,
"step": 3340
},
{
"epoch": 44.01,
"learning_rate": 5.925925925925927e-06,
"loss": 0.0004,
"step": 3350
},
{
"epoch": 44.02,
"learning_rate": 5.777777777777778e-06,
"loss": 0.0004,
"step": 3360
},
{
"epoch": 44.02,
"learning_rate": 5.62962962962963e-06,
"loss": 0.0004,
"step": 3370
},
{
"epoch": 44.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1282513439655304,
"eval_runtime": 2.1105,
"eval_samples_per_second": 14.688,
"eval_steps_per_second": 3.791,
"step": 3375
},
{
"epoch": 45.0,
"learning_rate": 5.4814814814814815e-06,
"loss": 0.0004,
"step": 3380
},
{
"epoch": 45.0,
"learning_rate": 5.333333333333334e-06,
"loss": 0.0004,
"step": 3390
},
{
"epoch": 45.01,
"learning_rate": 5.185185185185185e-06,
"loss": 0.0004,
"step": 3400
},
{
"epoch": 45.01,
"learning_rate": 5.037037037037037e-06,
"loss": 0.0004,
"step": 3410
},
{
"epoch": 45.01,
"learning_rate": 4.888888888888889e-06,
"loss": 0.0004,
"step": 3420
},
{
"epoch": 45.01,
"learning_rate": 4.740740740740741e-06,
"loss": 0.0004,
"step": 3430
},
{
"epoch": 45.02,
"learning_rate": 4.592592592592593e-06,
"loss": 0.0004,
"step": 3440
},
{
"epoch": 45.02,
"learning_rate": 4.444444444444445e-06,
"loss": 0.0004,
"step": 3450
},
{
"epoch": 45.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1329645812511444,
"eval_runtime": 2.079,
"eval_samples_per_second": 14.911,
"eval_steps_per_second": 3.848,
"step": 3450
},
{
"epoch": 46.0,
"learning_rate": 4.296296296296296e-06,
"loss": 0.0004,
"step": 3460
},
{
"epoch": 46.01,
"learning_rate": 4.1481481481481485e-06,
"loss": 0.0004,
"step": 3470
},
{
"epoch": 46.01,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0004,
"step": 3480
},
{
"epoch": 46.01,
"learning_rate": 3.851851851851852e-06,
"loss": 0.0004,
"step": 3490
},
{
"epoch": 46.01,
"learning_rate": 3.7037037037037037e-06,
"loss": 0.0004,
"step": 3500
},
{
"epoch": 46.02,
"learning_rate": 3.555555555555556e-06,
"loss": 0.0004,
"step": 3510
},
{
"epoch": 46.02,
"learning_rate": 3.4074074074074077e-06,
"loss": 0.0004,
"step": 3520
},
{
"epoch": 46.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.13405713438987732,
"eval_runtime": 2.0851,
"eval_samples_per_second": 14.867,
"eval_steps_per_second": 3.837,
"step": 3525
},
{
"epoch": 47.0,
"learning_rate": 3.259259259259259e-06,
"loss": 0.0004,
"step": 3530
},
{
"epoch": 47.0,
"learning_rate": 3.111111111111111e-06,
"loss": 0.0004,
"step": 3540
},
{
"epoch": 47.01,
"learning_rate": 2.9629629629629633e-06,
"loss": 0.0004,
"step": 3550
},
{
"epoch": 47.01,
"learning_rate": 2.814814814814815e-06,
"loss": 0.0004,
"step": 3560
},
{
"epoch": 47.01,
"learning_rate": 2.666666666666667e-06,
"loss": 0.0004,
"step": 3570
},
{
"epoch": 47.01,
"learning_rate": 2.5185185185185186e-06,
"loss": 0.0004,
"step": 3580
},
{
"epoch": 47.02,
"learning_rate": 2.3703703703703703e-06,
"loss": 0.0004,
"step": 3590
},
{
"epoch": 47.02,
"learning_rate": 2.2222222222222225e-06,
"loss": 0.0004,
"step": 3600
},
{
"epoch": 47.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.13054759800434113,
"eval_runtime": 2.2343,
"eval_samples_per_second": 13.875,
"eval_steps_per_second": 3.581,
"step": 3600
},
{
"epoch": 48.0,
"learning_rate": 2.0740740740740742e-06,
"loss": 0.0004,
"step": 3610
},
{
"epoch": 48.01,
"learning_rate": 1.925925925925926e-06,
"loss": 0.0004,
"step": 3620
},
{
"epoch": 48.01,
"learning_rate": 1.777777777777778e-06,
"loss": 0.0004,
"step": 3630
},
{
"epoch": 48.01,
"learning_rate": 1.6296296296296295e-06,
"loss": 0.0003,
"step": 3640
},
{
"epoch": 48.01,
"learning_rate": 1.4814814814814817e-06,
"loss": 0.0004,
"step": 3650
},
{
"epoch": 48.02,
"learning_rate": 1.3333333333333334e-06,
"loss": 0.0004,
"step": 3660
},
{
"epoch": 48.02,
"learning_rate": 1.1851851851851852e-06,
"loss": 0.0004,
"step": 3670
},
{
"epoch": 48.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1309322565793991,
"eval_runtime": 2.2018,
"eval_samples_per_second": 14.079,
"eval_steps_per_second": 3.633,
"step": 3675
},
{
"epoch": 49.0,
"learning_rate": 1.0370370370370371e-06,
"loss": 0.0004,
"step": 3680
},
{
"epoch": 49.0,
"learning_rate": 8.88888888888889e-07,
"loss": 0.0004,
"step": 3690
},
{
"epoch": 49.01,
"learning_rate": 7.407407407407408e-07,
"loss": 0.0004,
"step": 3700
},
{
"epoch": 49.01,
"learning_rate": 5.925925925925926e-07,
"loss": 0.0004,
"step": 3710
},
{
"epoch": 49.01,
"learning_rate": 4.444444444444445e-07,
"loss": 0.0004,
"step": 3720
},
{
"epoch": 49.01,
"learning_rate": 2.962962962962963e-07,
"loss": 0.0004,
"step": 3730
},
{
"epoch": 49.02,
"learning_rate": 1.4814814814814815e-07,
"loss": 0.0004,
"step": 3740
},
{
"epoch": 49.02,
"learning_rate": 0.0,
"loss": 0.0004,
"step": 3750
},
{
"epoch": 49.02,
"eval_accuracy": 0.967741935483871,
"eval_loss": 0.1303798407316208,
"eval_runtime": 2.1527,
"eval_samples_per_second": 14.4,
"eval_steps_per_second": 3.716,
"step": 3750
},
{
"epoch": 49.02,
"step": 3750,
"total_flos": 1.86923023515648e+19,
"train_loss": 0.16177092336689433,
"train_runtime": 3695.0497,
"train_samples_per_second": 4.059,
"train_steps_per_second": 1.015
},
{
"epoch": 49.02,
"eval_accuracy": 0.9078947368421053,
"eval_loss": 0.4171188473701477,
"eval_runtime": 7.7151,
"eval_samples_per_second": 9.851,
"eval_steps_per_second": 2.463,
"step": 3750
},
{
"epoch": 49.02,
"eval_accuracy": 0.9078947368421053,
"eval_loss": 0.41711878776550293,
"eval_runtime": 5.3059,
"eval_samples_per_second": 14.324,
"eval_steps_per_second": 3.581,
"step": 3750
}
],
"max_steps": 3750,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.86923023515648e+19,
"trial_name": null,
"trial_params": null
}