vit-base-skin / trainer_state.json
khleeloo's picture
Training in progress, epoch 1
2d9529c
raw
history blame
31.8 kB
{
"best_metric": 0.8911917098445595,
"best_model_checkpoint": "./vit-base-skin/checkpoint-1252",
"epoch": 4.0,
"global_step": 2504,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 0.00019920127795527157,
"loss": 1.3607,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 0.00019840255591054313,
"loss": 1.0863,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 0.00019760383386581472,
"loss": 0.8928,
"step": 30
},
{
"epoch": 0.06,
"learning_rate": 0.00019680511182108628,
"loss": 1.0954,
"step": 40
},
{
"epoch": 0.08,
"learning_rate": 0.00019600638977635784,
"loss": 0.7844,
"step": 50
},
{
"epoch": 0.1,
"learning_rate": 0.0001952076677316294,
"loss": 0.8456,
"step": 60
},
{
"epoch": 0.11,
"learning_rate": 0.00019440894568690097,
"loss": 0.8827,
"step": 70
},
{
"epoch": 0.13,
"learning_rate": 0.00019361022364217253,
"loss": 0.8214,
"step": 80
},
{
"epoch": 0.14,
"learning_rate": 0.0001928115015974441,
"loss": 0.9063,
"step": 90
},
{
"epoch": 0.16,
"learning_rate": 0.00019201277955271565,
"loss": 0.8956,
"step": 100
},
{
"epoch": 0.18,
"learning_rate": 0.0001912140575079872,
"loss": 0.9014,
"step": 110
},
{
"epoch": 0.19,
"learning_rate": 0.0001904153354632588,
"loss": 0.83,
"step": 120
},
{
"epoch": 0.21,
"learning_rate": 0.00018961661341853036,
"loss": 0.7877,
"step": 130
},
{
"epoch": 0.22,
"learning_rate": 0.00018881789137380192,
"loss": 0.6501,
"step": 140
},
{
"epoch": 0.24,
"learning_rate": 0.00018801916932907348,
"loss": 0.8166,
"step": 150
},
{
"epoch": 0.26,
"learning_rate": 0.00018722044728434505,
"loss": 0.8288,
"step": 160
},
{
"epoch": 0.27,
"learning_rate": 0.00018642172523961663,
"loss": 0.662,
"step": 170
},
{
"epoch": 0.29,
"learning_rate": 0.0001856230031948882,
"loss": 0.7948,
"step": 180
},
{
"epoch": 0.3,
"learning_rate": 0.00018482428115015976,
"loss": 0.6333,
"step": 190
},
{
"epoch": 0.32,
"learning_rate": 0.00018402555910543132,
"loss": 0.7313,
"step": 200
},
{
"epoch": 0.34,
"learning_rate": 0.00018322683706070288,
"loss": 0.7396,
"step": 210
},
{
"epoch": 0.35,
"learning_rate": 0.00018242811501597444,
"loss": 0.683,
"step": 220
},
{
"epoch": 0.37,
"learning_rate": 0.000181629392971246,
"loss": 0.6597,
"step": 230
},
{
"epoch": 0.38,
"learning_rate": 0.00018083067092651756,
"loss": 0.6867,
"step": 240
},
{
"epoch": 0.4,
"learning_rate": 0.00018003194888178913,
"loss": 0.5824,
"step": 250
},
{
"epoch": 0.42,
"learning_rate": 0.00017923322683706071,
"loss": 0.7605,
"step": 260
},
{
"epoch": 0.43,
"learning_rate": 0.00017843450479233228,
"loss": 0.7816,
"step": 270
},
{
"epoch": 0.45,
"learning_rate": 0.00017763578274760384,
"loss": 0.7451,
"step": 280
},
{
"epoch": 0.46,
"learning_rate": 0.0001768370607028754,
"loss": 0.7161,
"step": 290
},
{
"epoch": 0.48,
"learning_rate": 0.000176038338658147,
"loss": 0.6315,
"step": 300
},
{
"epoch": 0.5,
"learning_rate": 0.00017523961661341855,
"loss": 0.6246,
"step": 310
},
{
"epoch": 0.51,
"learning_rate": 0.0001744408945686901,
"loss": 0.5847,
"step": 320
},
{
"epoch": 0.53,
"learning_rate": 0.00017364217252396167,
"loss": 0.632,
"step": 330
},
{
"epoch": 0.54,
"learning_rate": 0.00017284345047923323,
"loss": 0.7901,
"step": 340
},
{
"epoch": 0.56,
"learning_rate": 0.00017204472843450482,
"loss": 0.6926,
"step": 350
},
{
"epoch": 0.58,
"learning_rate": 0.00017124600638977638,
"loss": 0.6079,
"step": 360
},
{
"epoch": 0.59,
"learning_rate": 0.00017044728434504792,
"loss": 0.6323,
"step": 370
},
{
"epoch": 0.61,
"learning_rate": 0.00016964856230031948,
"loss": 0.6733,
"step": 380
},
{
"epoch": 0.62,
"learning_rate": 0.00016884984025559104,
"loss": 0.6248,
"step": 390
},
{
"epoch": 0.64,
"learning_rate": 0.00016805111821086263,
"loss": 0.7995,
"step": 400
},
{
"epoch": 0.65,
"learning_rate": 0.0001672523961661342,
"loss": 0.7554,
"step": 410
},
{
"epoch": 0.67,
"learning_rate": 0.00016645367412140575,
"loss": 0.6262,
"step": 420
},
{
"epoch": 0.69,
"learning_rate": 0.0001656549520766773,
"loss": 0.71,
"step": 430
},
{
"epoch": 0.7,
"learning_rate": 0.0001648562300319489,
"loss": 0.6281,
"step": 440
},
{
"epoch": 0.72,
"learning_rate": 0.00016405750798722046,
"loss": 0.6005,
"step": 450
},
{
"epoch": 0.73,
"learning_rate": 0.00016325878594249202,
"loss": 0.5122,
"step": 460
},
{
"epoch": 0.75,
"learning_rate": 0.00016246006389776359,
"loss": 0.5836,
"step": 470
},
{
"epoch": 0.77,
"learning_rate": 0.00016166134185303515,
"loss": 0.5401,
"step": 480
},
{
"epoch": 0.78,
"learning_rate": 0.00016086261980830674,
"loss": 0.8159,
"step": 490
},
{
"epoch": 0.8,
"learning_rate": 0.0001600638977635783,
"loss": 0.632,
"step": 500
},
{
"epoch": 0.81,
"learning_rate": 0.00015926517571884986,
"loss": 0.6293,
"step": 510
},
{
"epoch": 0.83,
"learning_rate": 0.00015846645367412142,
"loss": 0.5831,
"step": 520
},
{
"epoch": 0.85,
"learning_rate": 0.00015766773162939298,
"loss": 0.4642,
"step": 530
},
{
"epoch": 0.86,
"learning_rate": 0.00015686900958466454,
"loss": 0.6678,
"step": 540
},
{
"epoch": 0.88,
"learning_rate": 0.0001560702875399361,
"loss": 0.6439,
"step": 550
},
{
"epoch": 0.89,
"learning_rate": 0.00015527156549520767,
"loss": 0.6224,
"step": 560
},
{
"epoch": 0.91,
"learning_rate": 0.00015447284345047923,
"loss": 0.5306,
"step": 570
},
{
"epoch": 0.93,
"learning_rate": 0.00015367412140575082,
"loss": 0.4909,
"step": 580
},
{
"epoch": 0.94,
"learning_rate": 0.00015287539936102238,
"loss": 0.4913,
"step": 590
},
{
"epoch": 0.96,
"learning_rate": 0.00015207667731629394,
"loss": 0.5959,
"step": 600
},
{
"epoch": 0.97,
"learning_rate": 0.0001512779552715655,
"loss": 0.7232,
"step": 610
},
{
"epoch": 0.99,
"learning_rate": 0.00015047923322683706,
"loss": 0.4987,
"step": 620
},
{
"epoch": 1.0,
"eval_accuracy": 0.8031088082901554,
"eval_loss": 0.5480978488922119,
"eval_runtime": 1.2438,
"eval_samples_per_second": 155.169,
"eval_steps_per_second": 10.452,
"step": 626
},
{
"epoch": 1.01,
"learning_rate": 0.00014968051118210865,
"loss": 0.5424,
"step": 630
},
{
"epoch": 1.02,
"learning_rate": 0.0001488817891373802,
"loss": 0.3795,
"step": 640
},
{
"epoch": 1.04,
"learning_rate": 0.00014808306709265177,
"loss": 0.5057,
"step": 650
},
{
"epoch": 1.05,
"learning_rate": 0.00014728434504792333,
"loss": 0.5375,
"step": 660
},
{
"epoch": 1.07,
"learning_rate": 0.0001464856230031949,
"loss": 0.5068,
"step": 670
},
{
"epoch": 1.09,
"learning_rate": 0.00014568690095846646,
"loss": 0.5207,
"step": 680
},
{
"epoch": 1.1,
"learning_rate": 0.00014488817891373802,
"loss": 0.4394,
"step": 690
},
{
"epoch": 1.12,
"learning_rate": 0.00014408945686900958,
"loss": 0.3614,
"step": 700
},
{
"epoch": 1.13,
"learning_rate": 0.00014329073482428114,
"loss": 0.5282,
"step": 710
},
{
"epoch": 1.15,
"learning_rate": 0.00014249201277955273,
"loss": 0.4466,
"step": 720
},
{
"epoch": 1.17,
"learning_rate": 0.0001416932907348243,
"loss": 0.4081,
"step": 730
},
{
"epoch": 1.18,
"learning_rate": 0.00014089456869009585,
"loss": 0.4235,
"step": 740
},
{
"epoch": 1.2,
"learning_rate": 0.0001400958466453674,
"loss": 0.3791,
"step": 750
},
{
"epoch": 1.21,
"learning_rate": 0.000139297124600639,
"loss": 0.4024,
"step": 760
},
{
"epoch": 1.23,
"learning_rate": 0.00013849840255591056,
"loss": 0.3729,
"step": 770
},
{
"epoch": 1.25,
"learning_rate": 0.00013769968051118212,
"loss": 0.4715,
"step": 780
},
{
"epoch": 1.26,
"learning_rate": 0.00013690095846645369,
"loss": 0.4664,
"step": 790
},
{
"epoch": 1.28,
"learning_rate": 0.00013610223642172525,
"loss": 0.4945,
"step": 800
},
{
"epoch": 1.29,
"learning_rate": 0.0001353035143769968,
"loss": 0.4936,
"step": 810
},
{
"epoch": 1.31,
"learning_rate": 0.00013450479233226837,
"loss": 0.5814,
"step": 820
},
{
"epoch": 1.33,
"learning_rate": 0.00013370607028753993,
"loss": 0.3679,
"step": 830
},
{
"epoch": 1.34,
"learning_rate": 0.0001329073482428115,
"loss": 0.4897,
"step": 840
},
{
"epoch": 1.36,
"learning_rate": 0.00013210862619808305,
"loss": 0.4264,
"step": 850
},
{
"epoch": 1.37,
"learning_rate": 0.00013130990415335464,
"loss": 0.6188,
"step": 860
},
{
"epoch": 1.39,
"learning_rate": 0.0001305111821086262,
"loss": 0.4897,
"step": 870
},
{
"epoch": 1.41,
"learning_rate": 0.00012971246006389777,
"loss": 0.3392,
"step": 880
},
{
"epoch": 1.42,
"learning_rate": 0.00012891373801916933,
"loss": 0.5327,
"step": 890
},
{
"epoch": 1.44,
"learning_rate": 0.00012811501597444092,
"loss": 0.3651,
"step": 900
},
{
"epoch": 1.45,
"learning_rate": 0.00012731629392971248,
"loss": 0.4061,
"step": 910
},
{
"epoch": 1.47,
"learning_rate": 0.00012651757188498404,
"loss": 0.4701,
"step": 920
},
{
"epoch": 1.49,
"learning_rate": 0.0001257188498402556,
"loss": 0.3794,
"step": 930
},
{
"epoch": 1.5,
"learning_rate": 0.00012492012779552716,
"loss": 0.5627,
"step": 940
},
{
"epoch": 1.52,
"learning_rate": 0.00012412140575079872,
"loss": 0.514,
"step": 950
},
{
"epoch": 1.53,
"learning_rate": 0.00012332268370607028,
"loss": 0.424,
"step": 960
},
{
"epoch": 1.55,
"learning_rate": 0.00012252396166134185,
"loss": 0.4829,
"step": 970
},
{
"epoch": 1.57,
"learning_rate": 0.00012172523961661342,
"loss": 0.387,
"step": 980
},
{
"epoch": 1.58,
"learning_rate": 0.000120926517571885,
"loss": 0.472,
"step": 990
},
{
"epoch": 1.6,
"learning_rate": 0.00012012779552715656,
"loss": 0.3891,
"step": 1000
},
{
"epoch": 1.61,
"learning_rate": 0.00011932907348242812,
"loss": 0.4143,
"step": 1010
},
{
"epoch": 1.63,
"learning_rate": 0.00011853035143769968,
"loss": 0.4191,
"step": 1020
},
{
"epoch": 1.65,
"learning_rate": 0.00011773162939297124,
"loss": 0.44,
"step": 1030
},
{
"epoch": 1.66,
"learning_rate": 0.00011693290734824283,
"loss": 0.4942,
"step": 1040
},
{
"epoch": 1.68,
"learning_rate": 0.00011613418530351439,
"loss": 0.4446,
"step": 1050
},
{
"epoch": 1.69,
"learning_rate": 0.00011533546325878595,
"loss": 0.3974,
"step": 1060
},
{
"epoch": 1.71,
"learning_rate": 0.00011453674121405751,
"loss": 0.4112,
"step": 1070
},
{
"epoch": 1.73,
"learning_rate": 0.00011373801916932908,
"loss": 0.4987,
"step": 1080
},
{
"epoch": 1.74,
"learning_rate": 0.00011293929712460065,
"loss": 0.3738,
"step": 1090
},
{
"epoch": 1.76,
"learning_rate": 0.00011214057507987221,
"loss": 0.3729,
"step": 1100
},
{
"epoch": 1.77,
"learning_rate": 0.00011134185303514377,
"loss": 0.582,
"step": 1110
},
{
"epoch": 1.79,
"learning_rate": 0.00011054313099041533,
"loss": 0.3539,
"step": 1120
},
{
"epoch": 1.81,
"learning_rate": 0.00010974440894568691,
"loss": 0.222,
"step": 1130
},
{
"epoch": 1.82,
"learning_rate": 0.00010894568690095847,
"loss": 0.2987,
"step": 1140
},
{
"epoch": 1.84,
"learning_rate": 0.00010814696485623003,
"loss": 0.5448,
"step": 1150
},
{
"epoch": 1.85,
"learning_rate": 0.0001073482428115016,
"loss": 0.377,
"step": 1160
},
{
"epoch": 1.87,
"learning_rate": 0.00010654952076677316,
"loss": 0.3518,
"step": 1170
},
{
"epoch": 1.88,
"learning_rate": 0.00010575079872204474,
"loss": 0.5725,
"step": 1180
},
{
"epoch": 1.9,
"learning_rate": 0.0001049520766773163,
"loss": 0.2406,
"step": 1190
},
{
"epoch": 1.92,
"learning_rate": 0.00010415335463258787,
"loss": 0.3681,
"step": 1200
},
{
"epoch": 1.93,
"learning_rate": 0.00010335463258785943,
"loss": 0.4454,
"step": 1210
},
{
"epoch": 1.95,
"learning_rate": 0.000102555910543131,
"loss": 0.2933,
"step": 1220
},
{
"epoch": 1.96,
"learning_rate": 0.00010175718849840256,
"loss": 0.3093,
"step": 1230
},
{
"epoch": 1.98,
"learning_rate": 0.00010095846645367413,
"loss": 0.4952,
"step": 1240
},
{
"epoch": 2.0,
"learning_rate": 0.00010015974440894569,
"loss": 0.3814,
"step": 1250
},
{
"epoch": 2.0,
"eval_accuracy": 0.8911917098445595,
"eval_loss": 0.38037967681884766,
"eval_runtime": 1.2431,
"eval_samples_per_second": 155.258,
"eval_steps_per_second": 10.458,
"step": 1252
},
{
"epoch": 2.01,
"learning_rate": 9.936102236421726e-05,
"loss": 0.347,
"step": 1260
},
{
"epoch": 2.03,
"learning_rate": 9.856230031948882e-05,
"loss": 0.1718,
"step": 1270
},
{
"epoch": 2.04,
"learning_rate": 9.77635782747604e-05,
"loss": 0.2688,
"step": 1280
},
{
"epoch": 2.06,
"learning_rate": 9.696485623003195e-05,
"loss": 0.2019,
"step": 1290
},
{
"epoch": 2.08,
"learning_rate": 9.616613418530351e-05,
"loss": 0.2318,
"step": 1300
},
{
"epoch": 2.09,
"learning_rate": 9.536741214057508e-05,
"loss": 0.2364,
"step": 1310
},
{
"epoch": 2.11,
"learning_rate": 9.456869009584664e-05,
"loss": 0.2729,
"step": 1320
},
{
"epoch": 2.12,
"learning_rate": 9.376996805111822e-05,
"loss": 0.2046,
"step": 1330
},
{
"epoch": 2.14,
"learning_rate": 9.297124600638978e-05,
"loss": 0.2405,
"step": 1340
},
{
"epoch": 2.16,
"learning_rate": 9.217252396166136e-05,
"loss": 0.1687,
"step": 1350
},
{
"epoch": 2.17,
"learning_rate": 9.137380191693292e-05,
"loss": 0.2252,
"step": 1360
},
{
"epoch": 2.19,
"learning_rate": 9.057507987220448e-05,
"loss": 0.1544,
"step": 1370
},
{
"epoch": 2.2,
"learning_rate": 8.977635782747604e-05,
"loss": 0.2436,
"step": 1380
},
{
"epoch": 2.22,
"learning_rate": 8.89776357827476e-05,
"loss": 0.2051,
"step": 1390
},
{
"epoch": 2.24,
"learning_rate": 8.817891373801918e-05,
"loss": 0.163,
"step": 1400
},
{
"epoch": 2.25,
"learning_rate": 8.738019169329074e-05,
"loss": 0.2489,
"step": 1410
},
{
"epoch": 2.27,
"learning_rate": 8.658146964856231e-05,
"loss": 0.2738,
"step": 1420
},
{
"epoch": 2.28,
"learning_rate": 8.578274760383387e-05,
"loss": 0.1508,
"step": 1430
},
{
"epoch": 2.3,
"learning_rate": 8.498402555910544e-05,
"loss": 0.1544,
"step": 1440
},
{
"epoch": 2.32,
"learning_rate": 8.4185303514377e-05,
"loss": 0.2727,
"step": 1450
},
{
"epoch": 2.33,
"learning_rate": 8.338658146964856e-05,
"loss": 0.1686,
"step": 1460
},
{
"epoch": 2.35,
"learning_rate": 8.258785942492013e-05,
"loss": 0.1416,
"step": 1470
},
{
"epoch": 2.36,
"learning_rate": 8.17891373801917e-05,
"loss": 0.4586,
"step": 1480
},
{
"epoch": 2.38,
"learning_rate": 8.099041533546327e-05,
"loss": 0.1554,
"step": 1490
},
{
"epoch": 2.4,
"learning_rate": 8.019169329073483e-05,
"loss": 0.2235,
"step": 1500
},
{
"epoch": 2.41,
"learning_rate": 7.939297124600639e-05,
"loss": 0.1812,
"step": 1510
},
{
"epoch": 2.43,
"learning_rate": 7.859424920127795e-05,
"loss": 0.1648,
"step": 1520
},
{
"epoch": 2.44,
"learning_rate": 7.779552715654951e-05,
"loss": 0.1314,
"step": 1530
},
{
"epoch": 2.46,
"learning_rate": 7.699680511182109e-05,
"loss": 0.2276,
"step": 1540
},
{
"epoch": 2.48,
"learning_rate": 7.619808306709265e-05,
"loss": 0.1624,
"step": 1550
},
{
"epoch": 2.49,
"learning_rate": 7.539936102236423e-05,
"loss": 0.1603,
"step": 1560
},
{
"epoch": 2.51,
"learning_rate": 7.460063897763579e-05,
"loss": 0.2533,
"step": 1570
},
{
"epoch": 2.52,
"learning_rate": 7.380191693290735e-05,
"loss": 0.1245,
"step": 1580
},
{
"epoch": 2.54,
"learning_rate": 7.300319488817891e-05,
"loss": 0.1848,
"step": 1590
},
{
"epoch": 2.56,
"learning_rate": 7.220447284345049e-05,
"loss": 0.223,
"step": 1600
},
{
"epoch": 2.57,
"learning_rate": 7.140575079872205e-05,
"loss": 0.1223,
"step": 1610
},
{
"epoch": 2.59,
"learning_rate": 7.060702875399361e-05,
"loss": 0.309,
"step": 1620
},
{
"epoch": 2.6,
"learning_rate": 6.980830670926518e-05,
"loss": 0.354,
"step": 1630
},
{
"epoch": 2.62,
"learning_rate": 6.900958466453674e-05,
"loss": 0.2051,
"step": 1640
},
{
"epoch": 2.64,
"learning_rate": 6.821086261980832e-05,
"loss": 0.2932,
"step": 1650
},
{
"epoch": 2.65,
"learning_rate": 6.741214057507987e-05,
"loss": 0.1401,
"step": 1660
},
{
"epoch": 2.67,
"learning_rate": 6.661341853035144e-05,
"loss": 0.3012,
"step": 1670
},
{
"epoch": 2.68,
"learning_rate": 6.5814696485623e-05,
"loss": 0.1481,
"step": 1680
},
{
"epoch": 2.7,
"learning_rate": 6.501597444089457e-05,
"loss": 0.3556,
"step": 1690
},
{
"epoch": 2.72,
"learning_rate": 6.421725239616614e-05,
"loss": 0.1193,
"step": 1700
},
{
"epoch": 2.73,
"learning_rate": 6.34185303514377e-05,
"loss": 0.2956,
"step": 1710
},
{
"epoch": 2.75,
"learning_rate": 6.261980830670928e-05,
"loss": 0.164,
"step": 1720
},
{
"epoch": 2.76,
"learning_rate": 6.182108626198084e-05,
"loss": 0.1218,
"step": 1730
},
{
"epoch": 2.78,
"learning_rate": 6.1022364217252406e-05,
"loss": 0.094,
"step": 1740
},
{
"epoch": 2.8,
"learning_rate": 6.022364217252396e-05,
"loss": 0.1787,
"step": 1750
},
{
"epoch": 2.81,
"learning_rate": 5.942492012779552e-05,
"loss": 0.1833,
"step": 1760
},
{
"epoch": 2.83,
"learning_rate": 5.86261980830671e-05,
"loss": 0.1389,
"step": 1770
},
{
"epoch": 2.84,
"learning_rate": 5.782747603833866e-05,
"loss": 0.2109,
"step": 1780
},
{
"epoch": 2.86,
"learning_rate": 5.702875399361023e-05,
"loss": 0.1668,
"step": 1790
},
{
"epoch": 2.88,
"learning_rate": 5.623003194888179e-05,
"loss": 0.1651,
"step": 1800
},
{
"epoch": 2.89,
"learning_rate": 5.543130990415336e-05,
"loss": 0.2081,
"step": 1810
},
{
"epoch": 2.91,
"learning_rate": 5.4632587859424925e-05,
"loss": 0.2506,
"step": 1820
},
{
"epoch": 2.92,
"learning_rate": 5.383386581469649e-05,
"loss": 0.1272,
"step": 1830
},
{
"epoch": 2.94,
"learning_rate": 5.3035143769968054e-05,
"loss": 0.1332,
"step": 1840
},
{
"epoch": 2.96,
"learning_rate": 5.2236421725239616e-05,
"loss": 0.0827,
"step": 1850
},
{
"epoch": 2.97,
"learning_rate": 5.1437699680511184e-05,
"loss": 0.2688,
"step": 1860
},
{
"epoch": 2.99,
"learning_rate": 5.0638977635782745e-05,
"loss": 0.3182,
"step": 1870
},
{
"epoch": 3.0,
"eval_accuracy": 0.8652849740932642,
"eval_loss": 0.3915177285671234,
"eval_runtime": 1.2215,
"eval_samples_per_second": 158.006,
"eval_steps_per_second": 10.643,
"step": 1878
},
{
"epoch": 3.0,
"learning_rate": 4.984025559105431e-05,
"loss": 0.1101,
"step": 1880
},
{
"epoch": 3.02,
"learning_rate": 4.904153354632588e-05,
"loss": 0.08,
"step": 1890
},
{
"epoch": 3.04,
"learning_rate": 4.824281150159744e-05,
"loss": 0.0661,
"step": 1900
},
{
"epoch": 3.05,
"learning_rate": 4.744408945686901e-05,
"loss": 0.088,
"step": 1910
},
{
"epoch": 3.07,
"learning_rate": 4.664536741214058e-05,
"loss": 0.069,
"step": 1920
},
{
"epoch": 3.08,
"learning_rate": 4.584664536741215e-05,
"loss": 0.0646,
"step": 1930
},
{
"epoch": 3.1,
"learning_rate": 4.504792332268371e-05,
"loss": 0.0334,
"step": 1940
},
{
"epoch": 3.12,
"learning_rate": 4.424920127795527e-05,
"loss": 0.0727,
"step": 1950
},
{
"epoch": 3.13,
"learning_rate": 4.345047923322684e-05,
"loss": 0.0762,
"step": 1960
},
{
"epoch": 3.15,
"learning_rate": 4.265175718849841e-05,
"loss": 0.0199,
"step": 1970
},
{
"epoch": 3.16,
"learning_rate": 4.185303514376997e-05,
"loss": 0.0673,
"step": 1980
},
{
"epoch": 3.18,
"learning_rate": 4.1054313099041536e-05,
"loss": 0.0639,
"step": 1990
},
{
"epoch": 3.19,
"learning_rate": 4.0255591054313104e-05,
"loss": 0.068,
"step": 2000
},
{
"epoch": 3.21,
"learning_rate": 3.9456869009584666e-05,
"loss": 0.034,
"step": 2010
},
{
"epoch": 3.23,
"learning_rate": 3.8658146964856234e-05,
"loss": 0.0384,
"step": 2020
},
{
"epoch": 3.24,
"learning_rate": 3.7859424920127795e-05,
"loss": 0.0797,
"step": 2030
},
{
"epoch": 3.26,
"learning_rate": 3.7060702875399364e-05,
"loss": 0.0842,
"step": 2040
},
{
"epoch": 3.27,
"learning_rate": 3.6261980830670925e-05,
"loss": 0.0483,
"step": 2050
},
{
"epoch": 3.29,
"learning_rate": 3.546325878594249e-05,
"loss": 0.0688,
"step": 2060
},
{
"epoch": 3.31,
"learning_rate": 3.466453674121406e-05,
"loss": 0.0651,
"step": 2070
},
{
"epoch": 3.32,
"learning_rate": 3.386581469648562e-05,
"loss": 0.0215,
"step": 2080
},
{
"epoch": 3.34,
"learning_rate": 3.306709265175719e-05,
"loss": 0.0203,
"step": 2090
},
{
"epoch": 3.35,
"learning_rate": 3.226837060702875e-05,
"loss": 0.035,
"step": 2100
},
{
"epoch": 3.37,
"learning_rate": 3.146964856230032e-05,
"loss": 0.1068,
"step": 2110
},
{
"epoch": 3.39,
"learning_rate": 3.067092651757188e-05,
"loss": 0.0426,
"step": 2120
},
{
"epoch": 3.4,
"learning_rate": 2.987220447284345e-05,
"loss": 0.0234,
"step": 2130
},
{
"epoch": 3.42,
"learning_rate": 2.907348242811502e-05,
"loss": 0.0345,
"step": 2140
},
{
"epoch": 3.43,
"learning_rate": 2.8274760383386583e-05,
"loss": 0.1023,
"step": 2150
},
{
"epoch": 3.45,
"learning_rate": 2.747603833865815e-05,
"loss": 0.0391,
"step": 2160
},
{
"epoch": 3.47,
"learning_rate": 2.6677316293929716e-05,
"loss": 0.0536,
"step": 2170
},
{
"epoch": 3.48,
"learning_rate": 2.5878594249201278e-05,
"loss": 0.0318,
"step": 2180
},
{
"epoch": 3.5,
"learning_rate": 2.5079872204472842e-05,
"loss": 0.0285,
"step": 2190
},
{
"epoch": 3.51,
"learning_rate": 2.428115015974441e-05,
"loss": 0.062,
"step": 2200
},
{
"epoch": 3.53,
"learning_rate": 2.3482428115015975e-05,
"loss": 0.0315,
"step": 2210
},
{
"epoch": 3.55,
"learning_rate": 2.268370607028754e-05,
"loss": 0.0656,
"step": 2220
},
{
"epoch": 3.56,
"learning_rate": 2.188498402555911e-05,
"loss": 0.0152,
"step": 2230
},
{
"epoch": 3.58,
"learning_rate": 2.108626198083067e-05,
"loss": 0.1108,
"step": 2240
},
{
"epoch": 3.59,
"learning_rate": 2.0287539936102238e-05,
"loss": 0.0951,
"step": 2250
},
{
"epoch": 3.61,
"learning_rate": 1.9488817891373803e-05,
"loss": 0.1139,
"step": 2260
},
{
"epoch": 3.63,
"learning_rate": 1.869009584664537e-05,
"loss": 0.027,
"step": 2270
},
{
"epoch": 3.64,
"learning_rate": 1.7891373801916932e-05,
"loss": 0.0577,
"step": 2280
},
{
"epoch": 3.66,
"learning_rate": 1.7172523961661345e-05,
"loss": 0.0411,
"step": 2290
},
{
"epoch": 3.67,
"learning_rate": 1.6373801916932906e-05,
"loss": 0.0706,
"step": 2300
},
{
"epoch": 3.69,
"learning_rate": 1.5575079872204475e-05,
"loss": 0.1357,
"step": 2310
},
{
"epoch": 3.71,
"learning_rate": 1.477635782747604e-05,
"loss": 0.0638,
"step": 2320
},
{
"epoch": 3.72,
"learning_rate": 1.3977635782747606e-05,
"loss": 0.0508,
"step": 2330
},
{
"epoch": 3.74,
"learning_rate": 1.3178913738019169e-05,
"loss": 0.0224,
"step": 2340
},
{
"epoch": 3.75,
"learning_rate": 1.2380191693290735e-05,
"loss": 0.0108,
"step": 2350
},
{
"epoch": 3.77,
"learning_rate": 1.1581469648562302e-05,
"loss": 0.0652,
"step": 2360
},
{
"epoch": 3.79,
"learning_rate": 1.0782747603833867e-05,
"loss": 0.0548,
"step": 2370
},
{
"epoch": 3.8,
"learning_rate": 9.984025559105432e-06,
"loss": 0.0335,
"step": 2380
},
{
"epoch": 3.82,
"learning_rate": 9.185303514376996e-06,
"loss": 0.0372,
"step": 2390
},
{
"epoch": 3.83,
"learning_rate": 8.386581469648563e-06,
"loss": 0.0994,
"step": 2400
},
{
"epoch": 3.85,
"learning_rate": 7.5878594249201285e-06,
"loss": 0.0135,
"step": 2410
},
{
"epoch": 3.87,
"learning_rate": 6.789137380191693e-06,
"loss": 0.0583,
"step": 2420
},
{
"epoch": 3.88,
"learning_rate": 5.990415335463259e-06,
"loss": 0.0961,
"step": 2430
},
{
"epoch": 3.9,
"learning_rate": 5.191693290734825e-06,
"loss": 0.0292,
"step": 2440
},
{
"epoch": 3.91,
"learning_rate": 4.39297124600639e-06,
"loss": 0.0288,
"step": 2450
},
{
"epoch": 3.93,
"learning_rate": 3.5942492012779555e-06,
"loss": 0.0565,
"step": 2460
},
{
"epoch": 3.95,
"learning_rate": 2.7955271565495207e-06,
"loss": 0.0607,
"step": 2470
},
{
"epoch": 3.96,
"learning_rate": 1.9968051118210863e-06,
"loss": 0.0529,
"step": 2480
},
{
"epoch": 3.98,
"learning_rate": 1.1980830670926517e-06,
"loss": 0.0449,
"step": 2490
},
{
"epoch": 3.99,
"learning_rate": 3.9936102236421723e-07,
"loss": 0.067,
"step": 2500
},
{
"epoch": 4.0,
"eval_accuracy": 0.8704663212435233,
"eval_loss": 0.4523600935935974,
"eval_runtime": 1.2991,
"eval_samples_per_second": 148.567,
"eval_steps_per_second": 10.007,
"step": 2504
},
{
"epoch": 4.0,
"step": 2504,
"total_flos": 3.104468219559813e+18,
"train_loss": 0.3521563596071336,
"train_runtime": 855.1502,
"train_samples_per_second": 46.846,
"train_steps_per_second": 2.928
}
],
"max_steps": 2504,
"num_train_epochs": 4,
"total_flos": 3.104468219559813e+18,
"trial_name": null,
"trial_params": null
}