utd-pstnet-pde-from-msr-pretrain / trainer_state.json
Zhuoxu Huang
init
10a6b96
{
"best_metric": 0.9558139534883721,
"best_model_checkpoint": "UTD_PSTNet_0.05_32_20_encoder_only_init_from_pstnet-pde-msr-pretrain/checkpoint-1500",
"epoch": 10.0,
"global_step": 1500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07,
"learning_rate": 0.0016666666666666668,
"loss": 3.3244,
"step": 10
},
{
"epoch": 0.13,
"learning_rate": 0.0033333333333333335,
"loss": 3.2871,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 0.005000000000000001,
"loss": 3.2338,
"step": 30
},
{
"epoch": 0.27,
"learning_rate": 0.006666666666666667,
"loss": 3.139,
"step": 40
},
{
"epoch": 0.33,
"learning_rate": 0.008333333333333333,
"loss": 3.1182,
"step": 50
},
{
"epoch": 0.4,
"learning_rate": 0.010000000000000002,
"loss": 3.0352,
"step": 60
},
{
"epoch": 0.47,
"learning_rate": 0.011666666666666667,
"loss": 2.9244,
"step": 70
},
{
"epoch": 0.53,
"learning_rate": 0.013333333333333334,
"loss": 2.793,
"step": 80
},
{
"epoch": 0.6,
"learning_rate": 0.015,
"loss": 2.7002,
"step": 90
},
{
"epoch": 0.67,
"learning_rate": 0.016666666666666666,
"loss": 2.4439,
"step": 100
},
{
"epoch": 0.73,
"learning_rate": 0.018333333333333333,
"loss": 2.2782,
"step": 110
},
{
"epoch": 0.8,
"learning_rate": 0.020000000000000004,
"loss": 2.1274,
"step": 120
},
{
"epoch": 0.87,
"learning_rate": 0.021666666666666667,
"loss": 2.0017,
"step": 130
},
{
"epoch": 0.93,
"learning_rate": 0.023333333333333334,
"loss": 1.8383,
"step": 140
},
{
"epoch": 1.0,
"learning_rate": 0.025,
"loss": 1.6579,
"step": 150
},
{
"epoch": 1.0,
"eval_acc": 0.4616838127230737,
"eval_class_acc": [
0.4074074074074074,
0.0,
0.0,
0.0,
0.0,
0.6875,
0.7333333333333333,
0.0,
1.0,
0.0,
0.9375,
1.0,
0.5,
0.9375,
0.0,
0.0,
0.625,
0.125,
0.0,
0.0,
1.0,
0.0,
0.9333333333333333,
0.0,
0.06666666666666667,
0.9375,
0.9333333333333333
],
"eval_loss": 1.956584095954895,
"eval_video_acc": 0.4069767441860465,
"step": 150
},
{
"epoch": 1.07,
"learning_rate": 0.02666666666666667,
"loss": 1.4516,
"step": 160
},
{
"epoch": 1.13,
"learning_rate": 0.028333333333333335,
"loss": 1.4985,
"step": 170
},
{
"epoch": 1.2,
"learning_rate": 0.03,
"loss": 1.2078,
"step": 180
},
{
"epoch": 1.27,
"learning_rate": 0.03166666666666667,
"loss": 1.1453,
"step": 190
},
{
"epoch": 1.33,
"learning_rate": 0.03333333333333333,
"loss": 1.0496,
"step": 200
},
{
"epoch": 1.4,
"learning_rate": 0.034999999999999996,
"loss": 0.885,
"step": 210
},
{
"epoch": 1.47,
"learning_rate": 0.03666666666666667,
"loss": 0.9587,
"step": 220
},
{
"epoch": 1.53,
"learning_rate": 0.03833333333333334,
"loss": 0.8476,
"step": 230
},
{
"epoch": 1.6,
"learning_rate": 0.04000000000000001,
"loss": 0.8108,
"step": 240
},
{
"epoch": 1.67,
"learning_rate": 0.04166666666666667,
"loss": 0.6405,
"step": 250
},
{
"epoch": 1.73,
"learning_rate": 0.043333333333333335,
"loss": 0.6028,
"step": 260
},
{
"epoch": 1.8,
"learning_rate": 0.045000000000000005,
"loss": 0.605,
"step": 270
},
{
"epoch": 1.87,
"learning_rate": 0.04666666666666667,
"loss": 0.5063,
"step": 280
},
{
"epoch": 1.93,
"learning_rate": 0.04833333333333334,
"loss": 0.4979,
"step": 290
},
{
"epoch": 2.0,
"learning_rate": 0.05,
"loss": 0.4003,
"step": 300
},
{
"epoch": 2.0,
"eval_acc": 0.7606550493386521,
"eval_class_acc": [
0.8888888888888888,
0.9375,
0.625,
0.875,
0.6666666666666666,
0.3125,
0.7333333333333333,
1.0,
0.5,
1.0,
0.75,
0.9375,
1.0,
0.75,
0.5714285714285714,
1.0,
0.75,
0.8125,
0.9375,
0.6153846153846154,
0.8125,
1.0,
0.2,
0.75,
0.9333333333333333,
1.0,
1.0
],
"eval_loss": 0.8617399334907532,
"eval_video_acc": 0.7953488372093023,
"step": 300
},
{
"epoch": 2.07,
"learning_rate": 0.04981481481481482,
"loss": 0.3683,
"step": 310
},
{
"epoch": 2.13,
"learning_rate": 0.049629629629629635,
"loss": 0.307,
"step": 320
},
{
"epoch": 2.2,
"learning_rate": 0.04944444444444445,
"loss": 0.3476,
"step": 330
},
{
"epoch": 2.27,
"learning_rate": 0.04925925925925926,
"loss": 0.3378,
"step": 340
},
{
"epoch": 2.33,
"learning_rate": 0.049074074074074076,
"loss": 0.2597,
"step": 350
},
{
"epoch": 2.4,
"learning_rate": 0.04888888888888889,
"loss": 0.2054,
"step": 360
},
{
"epoch": 2.47,
"learning_rate": 0.04870370370370371,
"loss": 0.2538,
"step": 370
},
{
"epoch": 2.53,
"learning_rate": 0.048518518518518516,
"loss": 0.1798,
"step": 380
},
{
"epoch": 2.6,
"learning_rate": 0.04833333333333334,
"loss": 0.1361,
"step": 390
},
{
"epoch": 2.67,
"learning_rate": 0.04814814814814815,
"loss": 0.1726,
"step": 400
},
{
"epoch": 2.73,
"learning_rate": 0.047962962962962964,
"loss": 0.1387,
"step": 410
},
{
"epoch": 2.8,
"learning_rate": 0.04777777777777778,
"loss": 0.1443,
"step": 420
},
{
"epoch": 2.87,
"learning_rate": 0.047592592592592596,
"loss": 0.1652,
"step": 430
},
{
"epoch": 2.93,
"learning_rate": 0.04740740740740741,
"loss": 0.176,
"step": 440
},
{
"epoch": 3.0,
"learning_rate": 0.04722222222222222,
"loss": 0.1585,
"step": 450
},
{
"epoch": 3.0,
"eval_acc": 0.8013856812933026,
"eval_class_acc": [
0.8148148148148148,
0.9375,
0.3125,
0.0625,
0.5333333333333333,
1.0,
0.26666666666666666,
1.0,
1.0,
0.625,
0.9375,
1.0,
0.9375,
0.75,
0.5,
0.9230769230769231,
0.75,
0.8125,
0.9375,
0.6923076923076923,
0.9375,
0.6875,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.9217666387557983,
"eval_video_acc": 0.7930232558139535,
"step": 450
},
{
"epoch": 3.07,
"learning_rate": 0.047037037037037044,
"loss": 0.1146,
"step": 460
},
{
"epoch": 3.13,
"learning_rate": 0.04685185185185185,
"loss": 0.1312,
"step": 470
},
{
"epoch": 3.2,
"learning_rate": 0.04666666666666667,
"loss": 0.1586,
"step": 480
},
{
"epoch": 3.27,
"learning_rate": 0.046481481481481485,
"loss": 0.0861,
"step": 490
},
{
"epoch": 3.33,
"learning_rate": 0.0462962962962963,
"loss": 0.0667,
"step": 500
},
{
"epoch": 3.4,
"learning_rate": 0.04611111111111112,
"loss": 0.0614,
"step": 510
},
{
"epoch": 3.47,
"learning_rate": 0.045925925925925926,
"loss": 0.0803,
"step": 520
},
{
"epoch": 3.53,
"learning_rate": 0.04574074074074075,
"loss": 0.1192,
"step": 530
},
{
"epoch": 3.6,
"learning_rate": 0.04555555555555556,
"loss": 0.0755,
"step": 540
},
{
"epoch": 3.67,
"learning_rate": 0.04537037037037037,
"loss": 0.034,
"step": 550
},
{
"epoch": 3.73,
"learning_rate": 0.04518518518518519,
"loss": 0.0742,
"step": 560
},
{
"epoch": 3.8,
"learning_rate": 0.045000000000000005,
"loss": 0.046,
"step": 570
},
{
"epoch": 3.87,
"learning_rate": 0.044814814814814814,
"loss": 0.094,
"step": 580
},
{
"epoch": 3.93,
"learning_rate": 0.04462962962962963,
"loss": 0.0881,
"step": 590
},
{
"epoch": 4.0,
"learning_rate": 0.044444444444444446,
"loss": 0.0825,
"step": 600
},
{
"epoch": 4.0,
"eval_acc": 0.8351879067814403,
"eval_class_acc": [
0.9259259259259259,
0.875,
0.6875,
1.0,
0.7333333333333333,
1.0,
0.2,
1.0,
1.0,
0.6875,
1.0,
0.9375,
1.0,
1.0,
0.35714285714285715,
0.9230769230769231,
0.6875,
0.8125,
1.0,
0.5384615384615384,
0.625,
1.0,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.9301585555076599,
"eval_video_acc": 0.8558139534883721,
"step": 600
},
{
"epoch": 4.07,
"learning_rate": 0.04425925925925926,
"loss": 0.0465,
"step": 610
},
{
"epoch": 4.13,
"learning_rate": 0.04407407407407408,
"loss": 0.031,
"step": 620
},
{
"epoch": 4.2,
"learning_rate": 0.043888888888888894,
"loss": 0.0846,
"step": 630
},
{
"epoch": 4.27,
"learning_rate": 0.04370370370370371,
"loss": 0.024,
"step": 640
},
{
"epoch": 4.33,
"learning_rate": 0.04351851851851852,
"loss": 0.0452,
"step": 650
},
{
"epoch": 4.4,
"learning_rate": 0.043333333333333335,
"loss": 0.0375,
"step": 660
},
{
"epoch": 4.47,
"learning_rate": 0.04314814814814815,
"loss": 0.0389,
"step": 670
},
{
"epoch": 4.53,
"learning_rate": 0.04296296296296297,
"loss": 0.0495,
"step": 680
},
{
"epoch": 4.6,
"learning_rate": 0.042777777777777776,
"loss": 0.0427,
"step": 690
},
{
"epoch": 4.67,
"learning_rate": 0.0425925925925926,
"loss": 0.032,
"step": 700
},
{
"epoch": 4.73,
"learning_rate": 0.04240740740740741,
"loss": 0.0321,
"step": 710
},
{
"epoch": 4.8,
"learning_rate": 0.042222222222222223,
"loss": 0.0371,
"step": 720
},
{
"epoch": 4.87,
"learning_rate": 0.04203703703703704,
"loss": 0.0271,
"step": 730
},
{
"epoch": 4.93,
"learning_rate": 0.041851851851851855,
"loss": 0.0231,
"step": 740
},
{
"epoch": 5.0,
"learning_rate": 0.04166666666666667,
"loss": 0.0542,
"step": 750
},
{
"epoch": 5.0,
"eval_acc": 0.876968297291623,
"eval_class_acc": [
0.9629629629629629,
1.0,
0.6875,
1.0,
0.6666666666666666,
0.8125,
0.6666666666666666,
1.0,
1.0,
0.875,
0.9375,
0.9375,
0.9375,
0.9375,
0.5,
1.0,
0.8125,
1.0,
0.8125,
0.9230769230769231,
1.0,
1.0,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.5839924812316895,
"eval_video_acc": 0.9069767441860465,
"step": 750
},
{
"epoch": 5.07,
"learning_rate": 0.04148148148148148,
"loss": 0.0426,
"step": 760
},
{
"epoch": 5.13,
"learning_rate": 0.0412962962962963,
"loss": 0.0481,
"step": 770
},
{
"epoch": 5.2,
"learning_rate": 0.04111111111111111,
"loss": 0.0199,
"step": 780
},
{
"epoch": 5.27,
"learning_rate": 0.04092592592592593,
"loss": 0.0139,
"step": 790
},
{
"epoch": 5.33,
"learning_rate": 0.040740740740740744,
"loss": 0.0182,
"step": 800
},
{
"epoch": 5.4,
"learning_rate": 0.04055555555555556,
"loss": 0.0172,
"step": 810
},
{
"epoch": 5.47,
"learning_rate": 0.040370370370370376,
"loss": 0.0074,
"step": 820
},
{
"epoch": 5.53,
"learning_rate": 0.040185185185185185,
"loss": 0.0201,
"step": 830
},
{
"epoch": 5.6,
"learning_rate": 0.04000000000000001,
"loss": 0.0314,
"step": 840
},
{
"epoch": 5.67,
"learning_rate": 0.03981481481481482,
"loss": 0.0405,
"step": 850
},
{
"epoch": 5.73,
"learning_rate": 0.03962962962962963,
"loss": 0.0281,
"step": 860
},
{
"epoch": 5.8,
"learning_rate": 0.03944444444444445,
"loss": 0.0157,
"step": 870
},
{
"epoch": 5.87,
"learning_rate": 0.039259259259259265,
"loss": 0.0155,
"step": 880
},
{
"epoch": 5.93,
"learning_rate": 0.039074074074074074,
"loss": 0.0132,
"step": 890
},
{
"epoch": 6.0,
"learning_rate": 0.03888888888888889,
"loss": 0.0148,
"step": 900
},
{
"epoch": 6.0,
"eval_acc": 0.8639512912030233,
"eval_class_acc": [
0.9259259259259259,
1.0,
0.75,
1.0,
0.4,
0.5,
1.0,
1.0,
1.0,
0.8125,
1.0,
1.0,
1.0,
1.0,
0.5714285714285714,
0.8461538461538461,
0.8125,
1.0,
0.9375,
0.7692307692307693,
1.0,
1.0,
0.7333333333333333,
1.0,
0.9333333333333333,
1.0,
1.0
],
"eval_loss": 0.6891928315162659,
"eval_video_acc": 0.8930232558139535,
"step": 900
},
{
"epoch": 6.07,
"learning_rate": 0.038703703703703705,
"loss": 0.0333,
"step": 910
},
{
"epoch": 6.13,
"learning_rate": 0.03851851851851852,
"loss": 0.0233,
"step": 920
},
{
"epoch": 6.2,
"learning_rate": 0.03833333333333334,
"loss": 0.009,
"step": 930
},
{
"epoch": 6.27,
"learning_rate": 0.03814814814814815,
"loss": 0.0209,
"step": 940
},
{
"epoch": 6.33,
"learning_rate": 0.03796296296296297,
"loss": 0.0204,
"step": 950
},
{
"epoch": 6.4,
"learning_rate": 0.03777777777777778,
"loss": 0.0115,
"step": 960
},
{
"epoch": 6.47,
"learning_rate": 0.037592592592592594,
"loss": 0.0102,
"step": 970
},
{
"epoch": 6.53,
"learning_rate": 0.03740740740740741,
"loss": 0.0084,
"step": 980
},
{
"epoch": 6.6,
"learning_rate": 0.037222222222222226,
"loss": 0.0143,
"step": 990
},
{
"epoch": 6.67,
"learning_rate": 0.037037037037037035,
"loss": 0.0063,
"step": 1000
},
{
"epoch": 6.73,
"learning_rate": 0.03685185185185185,
"loss": 0.0119,
"step": 1010
},
{
"epoch": 6.8,
"learning_rate": 0.03666666666666667,
"loss": 0.0193,
"step": 1020
},
{
"epoch": 6.87,
"learning_rate": 0.03648148148148148,
"loss": 0.0078,
"step": 1030
},
{
"epoch": 6.93,
"learning_rate": 0.0362962962962963,
"loss": 0.0042,
"step": 1040
},
{
"epoch": 7.0,
"learning_rate": 0.036111111111111115,
"loss": 0.0066,
"step": 1050
},
{
"epoch": 7.0,
"eval_acc": 0.8853663657358808,
"eval_class_acc": [
0.9629629629629629,
1.0,
0.6875,
1.0,
0.5333333333333333,
0.8125,
0.8666666666666667,
1.0,
0.9375,
0.8125,
0.9375,
0.9375,
1.0,
0.875,
0.7142857142857143,
1.0,
1.0,
1.0,
1.0,
0.7692307692307693,
1.0,
0.875,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.5734891891479492,
"eval_video_acc": 0.9162790697674419,
"step": 1050
},
{
"epoch": 7.07,
"learning_rate": 0.03592592592592593,
"loss": 0.0224,
"step": 1060
},
{
"epoch": 7.13,
"learning_rate": 0.03574074074074074,
"loss": 0.0073,
"step": 1070
},
{
"epoch": 7.2,
"learning_rate": 0.035555555555555556,
"loss": 0.0072,
"step": 1080
},
{
"epoch": 7.27,
"learning_rate": 0.03537037037037037,
"loss": 0.0057,
"step": 1090
},
{
"epoch": 7.33,
"learning_rate": 0.03518518518518519,
"loss": 0.0116,
"step": 1100
},
{
"epoch": 7.4,
"learning_rate": 0.034999999999999996,
"loss": 0.0052,
"step": 1110
},
{
"epoch": 7.47,
"learning_rate": 0.03481481481481482,
"loss": 0.0024,
"step": 1120
},
{
"epoch": 7.53,
"learning_rate": 0.03462962962962963,
"loss": 0.0048,
"step": 1130
},
{
"epoch": 7.6,
"learning_rate": 0.034444444444444444,
"loss": 0.0055,
"step": 1140
},
{
"epoch": 7.67,
"learning_rate": 0.03425925925925926,
"loss": 0.0027,
"step": 1150
},
{
"epoch": 7.73,
"learning_rate": 0.034074074074074076,
"loss": 0.0014,
"step": 1160
},
{
"epoch": 7.8,
"learning_rate": 0.03388888888888889,
"loss": 0.001,
"step": 1170
},
{
"epoch": 7.87,
"learning_rate": 0.0337037037037037,
"loss": 0.0029,
"step": 1180
},
{
"epoch": 7.93,
"learning_rate": 0.033518518518518524,
"loss": 0.0038,
"step": 1190
},
{
"epoch": 8.0,
"learning_rate": 0.03333333333333333,
"loss": 0.0048,
"step": 1200
},
{
"epoch": 8.0,
"eval_acc": 0.8990132269577997,
"eval_class_acc": [
0.8888888888888888,
1.0,
0.5,
1.0,
0.7333333333333333,
1.0,
0.8,
1.0,
1.0,
0.6875,
1.0,
1.0,
1.0,
1.0,
0.5,
1.0,
0.875,
1.0,
1.0,
0.7692307692307693,
1.0,
0.75,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.5563844442367554,
"eval_video_acc": 0.9069767441860465,
"step": 1200
},
{
"epoch": 8.07,
"learning_rate": 0.03314814814814815,
"loss": 0.0043,
"step": 1210
},
{
"epoch": 8.13,
"learning_rate": 0.032962962962962965,
"loss": 0.0048,
"step": 1220
},
{
"epoch": 8.2,
"learning_rate": 0.03277777777777778,
"loss": 0.0175,
"step": 1230
},
{
"epoch": 8.27,
"learning_rate": 0.0325925925925926,
"loss": 0.0032,
"step": 1240
},
{
"epoch": 8.33,
"learning_rate": 0.032407407407407406,
"loss": 0.0137,
"step": 1250
},
{
"epoch": 8.4,
"learning_rate": 0.03222222222222223,
"loss": 0.0078,
"step": 1260
},
{
"epoch": 8.47,
"learning_rate": 0.03203703703703704,
"loss": 0.0132,
"step": 1270
},
{
"epoch": 8.53,
"learning_rate": 0.03185185185185185,
"loss": 0.007,
"step": 1280
},
{
"epoch": 8.6,
"learning_rate": 0.03166666666666667,
"loss": 0.0024,
"step": 1290
},
{
"epoch": 8.67,
"learning_rate": 0.031481481481481485,
"loss": 0.0036,
"step": 1300
},
{
"epoch": 8.73,
"learning_rate": 0.031296296296296294,
"loss": 0.004,
"step": 1310
},
{
"epoch": 8.8,
"learning_rate": 0.031111111111111114,
"loss": 0.0026,
"step": 1320
},
{
"epoch": 8.87,
"learning_rate": 0.030925925925925926,
"loss": 0.0039,
"step": 1330
},
{
"epoch": 8.93,
"learning_rate": 0.030740740740740742,
"loss": 0.0014,
"step": 1340
},
{
"epoch": 9.0,
"learning_rate": 0.030555555555555558,
"loss": 0.0055,
"step": 1350
},
{
"epoch": 9.0,
"eval_acc": 0.8975435649800546,
"eval_class_acc": [
1.0,
0.875,
0.75,
1.0,
0.6,
0.75,
1.0,
1.0,
1.0,
1.0,
0.9375,
1.0,
1.0,
0.9375,
0.9285714285714286,
1.0,
0.8125,
1.0,
0.8125,
0.7692307692307693,
0.8125,
0.8125,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.6266849040985107,
"eval_video_acc": 0.9186046511627907,
"step": 1350
},
{
"epoch": 9.07,
"learning_rate": 0.03037037037037037,
"loss": 0.0044,
"step": 1360
},
{
"epoch": 9.13,
"learning_rate": 0.03018518518518519,
"loss": 0.0036,
"step": 1370
},
{
"epoch": 9.2,
"learning_rate": 0.03,
"loss": 0.0055,
"step": 1380
},
{
"epoch": 9.27,
"learning_rate": 0.02981481481481482,
"loss": 0.004,
"step": 1390
},
{
"epoch": 9.33,
"learning_rate": 0.02962962962962963,
"loss": 0.003,
"step": 1400
},
{
"epoch": 9.4,
"learning_rate": 0.029444444444444447,
"loss": 0.0036,
"step": 1410
},
{
"epoch": 9.47,
"learning_rate": 0.02925925925925926,
"loss": 0.018,
"step": 1420
},
{
"epoch": 9.53,
"learning_rate": 0.029074074074074075,
"loss": 0.0061,
"step": 1430
},
{
"epoch": 9.6,
"learning_rate": 0.028888888888888888,
"loss": 0.0029,
"step": 1440
},
{
"epoch": 9.67,
"learning_rate": 0.028703703703703703,
"loss": 0.0038,
"step": 1450
},
{
"epoch": 9.73,
"learning_rate": 0.028518518518518523,
"loss": 0.0024,
"step": 1460
},
{
"epoch": 9.8,
"learning_rate": 0.028333333333333335,
"loss": 0.0047,
"step": 1470
},
{
"epoch": 9.87,
"learning_rate": 0.02814814814814815,
"loss": 0.0033,
"step": 1480
},
{
"epoch": 9.93,
"learning_rate": 0.027962962962962964,
"loss": 0.002,
"step": 1490
},
{
"epoch": 10.0,
"learning_rate": 0.02777777777777778,
"loss": 0.0012,
"step": 1500
},
{
"epoch": 10.0,
"eval_acc": 0.9210581566239765,
"eval_class_acc": [
1.0,
0.9375,
0.75,
1.0,
0.9333333333333333,
0.9375,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.7857142857142857,
1.0,
0.75,
1.0,
1.0,
0.7692307692307693,
1.0,
1.0,
0.9333333333333333,
1.0,
1.0,
1.0,
1.0
],
"eval_loss": 0.42642584443092346,
"eval_video_acc": 0.9558139534883721,
"step": 1500
}
],
"max_steps": 3000,
"num_train_epochs": 20,
"total_flos": 3.588411829911552e+17,
"trial_name": null,
"trial_params": null
}