hushell's picture
Model save
68f6b06 verified
{
"best_metric": 1.1089974641799927,
"best_model_checkpoint": "data/tinyllama_moe_sft_ultrachat200k_v2_epochs5/checkpoint-3400",
"epoch": 4.9978098992553655,
"eval_steps": 100,
"global_step": 5705,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 1.7391304347826088e-07,
"loss": 2.7477,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 8.695652173913044e-07,
"loss": 2.6989,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 1.7391304347826088e-06,
"loss": 2.7026,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 2.6086956521739132e-06,
"loss": 2.5994,
"step": 15
},
{
"epoch": 0.02,
"learning_rate": 3.4782608695652175e-06,
"loss": 2.4352,
"step": 20
},
{
"epoch": 0.02,
"learning_rate": 4.347826086956522e-06,
"loss": 2.0773,
"step": 25
},
{
"epoch": 0.03,
"learning_rate": 5.2173913043478265e-06,
"loss": 1.8706,
"step": 30
},
{
"epoch": 0.03,
"learning_rate": 6.086956521739132e-06,
"loss": 1.7352,
"step": 35
},
{
"epoch": 0.04,
"learning_rate": 6.956521739130435e-06,
"loss": 1.6574,
"step": 40
},
{
"epoch": 0.04,
"learning_rate": 7.82608695652174e-06,
"loss": 1.588,
"step": 45
},
{
"epoch": 0.04,
"learning_rate": 8.695652173913044e-06,
"loss": 1.5429,
"step": 50
},
{
"epoch": 0.05,
"learning_rate": 9.565217391304349e-06,
"loss": 1.4804,
"step": 55
},
{
"epoch": 0.05,
"learning_rate": 1.0434782608695653e-05,
"loss": 1.4633,
"step": 60
},
{
"epoch": 0.06,
"learning_rate": 1.1304347826086957e-05,
"loss": 1.42,
"step": 65
},
{
"epoch": 0.06,
"learning_rate": 1.2173913043478263e-05,
"loss": 1.4117,
"step": 70
},
{
"epoch": 0.07,
"learning_rate": 1.3043478260869566e-05,
"loss": 1.3668,
"step": 75
},
{
"epoch": 0.07,
"learning_rate": 1.391304347826087e-05,
"loss": 1.359,
"step": 80
},
{
"epoch": 0.07,
"learning_rate": 1.4782608695652174e-05,
"loss": 1.3563,
"step": 85
},
{
"epoch": 0.08,
"learning_rate": 1.565217391304348e-05,
"loss": 1.3425,
"step": 90
},
{
"epoch": 0.08,
"learning_rate": 1.6521739130434785e-05,
"loss": 1.3117,
"step": 95
},
{
"epoch": 0.09,
"learning_rate": 1.739130434782609e-05,
"loss": 1.3359,
"step": 100
},
{
"epoch": 0.09,
"eval_loss": 1.3128533363342285,
"eval_runtime": 428.862,
"eval_samples_per_second": 37.695,
"eval_steps_per_second": 1.18,
"step": 100
},
{
"epoch": 0.09,
"learning_rate": 1.8260869565217393e-05,
"loss": 1.3087,
"step": 105
},
{
"epoch": 0.1,
"learning_rate": 1.9130434782608697e-05,
"loss": 1.2993,
"step": 110
},
{
"epoch": 0.1,
"learning_rate": 2e-05,
"loss": 1.2999,
"step": 115
},
{
"epoch": 0.11,
"learning_rate": 1.9999960519207943e-05,
"loss": 1.2738,
"step": 120
},
{
"epoch": 0.11,
"learning_rate": 1.999984207714351e-05,
"loss": 1.2769,
"step": 125
},
{
"epoch": 0.11,
"learning_rate": 1.9999644674741944e-05,
"loss": 1.255,
"step": 130
},
{
"epoch": 0.12,
"learning_rate": 1.9999368313561967e-05,
"loss": 1.271,
"step": 135
},
{
"epoch": 0.12,
"learning_rate": 1.9999012995785765e-05,
"loss": 1.282,
"step": 140
},
{
"epoch": 0.13,
"learning_rate": 1.9998578724218986e-05,
"loss": 1.2598,
"step": 145
},
{
"epoch": 0.13,
"learning_rate": 1.999806550229071e-05,
"loss": 1.2622,
"step": 150
},
{
"epoch": 0.14,
"learning_rate": 1.9997473334053415e-05,
"loss": 1.2541,
"step": 155
},
{
"epoch": 0.14,
"learning_rate": 1.9996802224182953e-05,
"loss": 1.248,
"step": 160
},
{
"epoch": 0.14,
"learning_rate": 1.999605217797852e-05,
"loss": 1.2444,
"step": 165
},
{
"epoch": 0.15,
"learning_rate": 1.9995223201362596e-05,
"loss": 1.2482,
"step": 170
},
{
"epoch": 0.15,
"learning_rate": 1.9994315300880913e-05,
"loss": 1.2575,
"step": 175
},
{
"epoch": 0.16,
"learning_rate": 1.9993328483702393e-05,
"loss": 1.2306,
"step": 180
},
{
"epoch": 0.16,
"learning_rate": 1.999226275761911e-05,
"loss": 1.2327,
"step": 185
},
{
"epoch": 0.17,
"learning_rate": 1.9991118131046197e-05,
"loss": 1.2423,
"step": 190
},
{
"epoch": 0.17,
"learning_rate": 1.998989461302181e-05,
"loss": 1.2272,
"step": 195
},
{
"epoch": 0.18,
"learning_rate": 1.998859221320704e-05,
"loss": 1.2425,
"step": 200
},
{
"epoch": 0.18,
"eval_loss": 1.2362912893295288,
"eval_runtime": 428.6396,
"eval_samples_per_second": 37.715,
"eval_steps_per_second": 1.18,
"step": 200
},
{
"epoch": 0.18,
"learning_rate": 1.9987210941885843e-05,
"loss": 1.2403,
"step": 205
},
{
"epoch": 0.18,
"learning_rate": 1.998575080996496e-05,
"loss": 1.2136,
"step": 210
},
{
"epoch": 0.19,
"learning_rate": 1.998421182897382e-05,
"loss": 1.246,
"step": 215
},
{
"epoch": 0.19,
"learning_rate": 1.9982594011064457e-05,
"loss": 1.2197,
"step": 220
},
{
"epoch": 0.2,
"learning_rate": 1.9980897369011424e-05,
"loss": 1.2035,
"step": 225
},
{
"epoch": 0.2,
"learning_rate": 1.9979121916211674e-05,
"loss": 1.2221,
"step": 230
},
{
"epoch": 0.21,
"learning_rate": 1.997726766668446e-05,
"loss": 1.2323,
"step": 235
},
{
"epoch": 0.21,
"learning_rate": 1.9975334635071234e-05,
"loss": 1.2047,
"step": 240
},
{
"epoch": 0.21,
"learning_rate": 1.9973322836635517e-05,
"loss": 1.2338,
"step": 245
},
{
"epoch": 0.22,
"learning_rate": 1.997123228726279e-05,
"loss": 1.2085,
"step": 250
},
{
"epoch": 0.22,
"learning_rate": 1.9969063003460364e-05,
"loss": 1.2169,
"step": 255
},
{
"epoch": 0.23,
"learning_rate": 1.9966815002357243e-05,
"loss": 1.2208,
"step": 260
},
{
"epoch": 0.23,
"learning_rate": 1.9964488301704004e-05,
"loss": 1.2016,
"step": 265
},
{
"epoch": 0.24,
"learning_rate": 1.996208291987264e-05,
"loss": 1.2027,
"step": 270
},
{
"epoch": 0.24,
"learning_rate": 1.995959887585643e-05,
"loss": 1.2171,
"step": 275
},
{
"epoch": 0.25,
"learning_rate": 1.9957036189269777e-05,
"loss": 1.1921,
"step": 280
},
{
"epoch": 0.25,
"learning_rate": 1.995439488034806e-05,
"loss": 1.224,
"step": 285
},
{
"epoch": 0.25,
"learning_rate": 1.9951674969947477e-05,
"loss": 1.1924,
"step": 290
},
{
"epoch": 0.26,
"learning_rate": 1.9948876479544866e-05,
"loss": 1.2121,
"step": 295
},
{
"epoch": 0.26,
"learning_rate": 1.994599943123755e-05,
"loss": 1.2079,
"step": 300
},
{
"epoch": 0.26,
"eval_loss": 1.208325743675232,
"eval_runtime": 428.2751,
"eval_samples_per_second": 37.747,
"eval_steps_per_second": 1.181,
"step": 300
},
{
"epoch": 0.27,
"learning_rate": 1.9943043847743164e-05,
"loss": 1.2103,
"step": 305
},
{
"epoch": 0.27,
"learning_rate": 1.9940009752399462e-05,
"loss": 1.1775,
"step": 310
},
{
"epoch": 0.28,
"learning_rate": 1.993689716916414e-05,
"loss": 1.2003,
"step": 315
},
{
"epoch": 0.28,
"learning_rate": 1.9933706122614645e-05,
"loss": 1.1945,
"step": 320
},
{
"epoch": 0.28,
"learning_rate": 1.993043663794799e-05,
"loss": 1.1911,
"step": 325
},
{
"epoch": 0.29,
"learning_rate": 1.992708874098054e-05,
"loss": 1.1859,
"step": 330
},
{
"epoch": 0.29,
"learning_rate": 1.992366245814783e-05,
"loss": 1.2373,
"step": 335
},
{
"epoch": 0.3,
"learning_rate": 1.992015781650432e-05,
"loss": 1.1896,
"step": 340
},
{
"epoch": 0.3,
"learning_rate": 1.991657484372322e-05,
"loss": 1.1971,
"step": 345
},
{
"epoch": 0.31,
"learning_rate": 1.9912913568096252e-05,
"loss": 1.1909,
"step": 350
},
{
"epoch": 0.31,
"learning_rate": 1.9909174018533428e-05,
"loss": 1.19,
"step": 355
},
{
"epoch": 0.32,
"learning_rate": 1.9905356224562824e-05,
"loss": 1.1955,
"step": 360
},
{
"epoch": 0.32,
"learning_rate": 1.9901460216330344e-05,
"loss": 1.192,
"step": 365
},
{
"epoch": 0.32,
"learning_rate": 1.989748602459949e-05,
"loss": 1.1862,
"step": 370
},
{
"epoch": 0.33,
"learning_rate": 1.9893433680751105e-05,
"loss": 1.1777,
"step": 375
},
{
"epoch": 0.33,
"learning_rate": 1.988930321678314e-05,
"loss": 1.1933,
"step": 380
},
{
"epoch": 0.34,
"learning_rate": 1.988509466531039e-05,
"loss": 1.1831,
"step": 385
},
{
"epoch": 0.34,
"learning_rate": 1.988080805956425e-05,
"loss": 1.1763,
"step": 390
},
{
"epoch": 0.35,
"learning_rate": 1.9876443433392433e-05,
"loss": 1.1777,
"step": 395
},
{
"epoch": 0.35,
"learning_rate": 1.9872000821258723e-05,
"loss": 1.1849,
"step": 400
},
{
"epoch": 0.35,
"eval_loss": 1.1909927129745483,
"eval_runtime": 426.8078,
"eval_samples_per_second": 37.877,
"eval_steps_per_second": 1.186,
"step": 400
},
{
"epoch": 0.35,
"learning_rate": 1.986748025824268e-05,
"loss": 1.206,
"step": 405
},
{
"epoch": 0.36,
"learning_rate": 1.9862881780039398e-05,
"loss": 1.1976,
"step": 410
},
{
"epoch": 0.36,
"learning_rate": 1.9858205422959183e-05,
"loss": 1.184,
"step": 415
},
{
"epoch": 0.37,
"learning_rate": 1.9853451223927293e-05,
"loss": 1.1855,
"step": 420
},
{
"epoch": 0.37,
"learning_rate": 1.9848619220483632e-05,
"loss": 1.1769,
"step": 425
},
{
"epoch": 0.38,
"learning_rate": 1.984370945078247e-05,
"loss": 1.1842,
"step": 430
},
{
"epoch": 0.38,
"learning_rate": 1.9838721953592122e-05,
"loss": 1.1709,
"step": 435
},
{
"epoch": 0.39,
"learning_rate": 1.983365676829466e-05,
"loss": 1.1851,
"step": 440
},
{
"epoch": 0.39,
"learning_rate": 1.9828513934885592e-05,
"loss": 1.1748,
"step": 445
},
{
"epoch": 0.39,
"learning_rate": 1.982329349397354e-05,
"loss": 1.1996,
"step": 450
},
{
"epoch": 0.4,
"learning_rate": 1.981799548677993e-05,
"loss": 1.1701,
"step": 455
},
{
"epoch": 0.4,
"learning_rate": 1.981261995513867e-05,
"loss": 1.1689,
"step": 460
},
{
"epoch": 0.41,
"learning_rate": 1.9807166941495813e-05,
"loss": 1.1793,
"step": 465
},
{
"epoch": 0.41,
"learning_rate": 1.9801636488909208e-05,
"loss": 1.1832,
"step": 470
},
{
"epoch": 0.42,
"learning_rate": 1.9796028641048197e-05,
"loss": 1.1771,
"step": 475
},
{
"epoch": 0.42,
"learning_rate": 1.979034344219323e-05,
"loss": 1.1846,
"step": 480
},
{
"epoch": 0.42,
"learning_rate": 1.9784580937235535e-05,
"loss": 1.1509,
"step": 485
},
{
"epoch": 0.43,
"learning_rate": 1.9778741171676766e-05,
"loss": 1.1836,
"step": 490
},
{
"epoch": 0.43,
"learning_rate": 1.9772824191628635e-05,
"loss": 1.1671,
"step": 495
},
{
"epoch": 0.44,
"learning_rate": 1.976683004381256e-05,
"loss": 1.1545,
"step": 500
},
{
"epoch": 0.44,
"eval_loss": 1.178593397140503,
"eval_runtime": 426.2866,
"eval_samples_per_second": 37.923,
"eval_steps_per_second": 1.187,
"step": 500
},
{
"epoch": 0.44,
"learning_rate": 1.9760758775559275e-05,
"loss": 1.1682,
"step": 505
},
{
"epoch": 0.45,
"learning_rate": 1.9754610434808476e-05,
"loss": 1.1715,
"step": 510
},
{
"epoch": 0.45,
"learning_rate": 1.9748385070108444e-05,
"loss": 1.1791,
"step": 515
},
{
"epoch": 0.46,
"learning_rate": 1.9742082730615634e-05,
"loss": 1.1492,
"step": 520
},
{
"epoch": 0.46,
"learning_rate": 1.9735703466094326e-05,
"loss": 1.1689,
"step": 525
},
{
"epoch": 0.46,
"learning_rate": 1.9729247326916194e-05,
"loss": 1.1593,
"step": 530
},
{
"epoch": 0.47,
"learning_rate": 1.9722714364059943e-05,
"loss": 1.1557,
"step": 535
},
{
"epoch": 0.47,
"learning_rate": 1.9716104629110882e-05,
"loss": 1.1702,
"step": 540
},
{
"epoch": 0.48,
"learning_rate": 1.9709418174260523e-05,
"loss": 1.1591,
"step": 545
},
{
"epoch": 0.48,
"learning_rate": 1.970265505230617e-05,
"loss": 1.1714,
"step": 550
},
{
"epoch": 0.49,
"learning_rate": 1.9695815316650514e-05,
"loss": 1.1721,
"step": 555
},
{
"epoch": 0.49,
"learning_rate": 1.9688899021301182e-05,
"loss": 1.1788,
"step": 560
},
{
"epoch": 0.49,
"learning_rate": 1.9681906220870345e-05,
"loss": 1.1759,
"step": 565
},
{
"epoch": 0.5,
"learning_rate": 1.9674836970574253e-05,
"loss": 1.1674,
"step": 570
},
{
"epoch": 0.5,
"learning_rate": 1.966769132623284e-05,
"loss": 1.1684,
"step": 575
},
{
"epoch": 0.51,
"learning_rate": 1.9660469344269234e-05,
"loss": 1.1528,
"step": 580
},
{
"epoch": 0.51,
"learning_rate": 1.9653171081709353e-05,
"loss": 1.1686,
"step": 585
},
{
"epoch": 0.52,
"learning_rate": 1.9645796596181436e-05,
"loss": 1.1598,
"step": 590
},
{
"epoch": 0.52,
"learning_rate": 1.9638345945915588e-05,
"loss": 1.1847,
"step": 595
},
{
"epoch": 0.53,
"learning_rate": 1.963081918974332e-05,
"loss": 1.174,
"step": 600
},
{
"epoch": 0.53,
"eval_loss": 1.169019341468811,
"eval_runtime": 427.0041,
"eval_samples_per_second": 37.859,
"eval_steps_per_second": 1.185,
"step": 600
},
{
"epoch": 0.53,
"learning_rate": 1.9623216387097095e-05,
"loss": 1.1445,
"step": 605
},
{
"epoch": 0.53,
"learning_rate": 1.9615537598009845e-05,
"loss": 1.1567,
"step": 610
},
{
"epoch": 0.54,
"learning_rate": 1.9607782883114507e-05,
"loss": 1.1774,
"step": 615
},
{
"epoch": 0.54,
"learning_rate": 1.9599952303643538e-05,
"loss": 1.1534,
"step": 620
},
{
"epoch": 0.55,
"learning_rate": 1.9592045921428435e-05,
"loss": 1.1571,
"step": 625
},
{
"epoch": 0.55,
"learning_rate": 1.958406379889924e-05,
"loss": 1.1573,
"step": 630
},
{
"epoch": 0.56,
"learning_rate": 1.957600599908406e-05,
"loss": 1.1769,
"step": 635
},
{
"epoch": 0.56,
"learning_rate": 1.956787258560856e-05,
"loss": 1.1678,
"step": 640
},
{
"epoch": 0.57,
"learning_rate": 1.955966362269546e-05,
"loss": 1.1648,
"step": 645
},
{
"epoch": 0.57,
"learning_rate": 1.955137917516403e-05,
"loss": 1.1546,
"step": 650
},
{
"epoch": 0.57,
"learning_rate": 1.954301930842958e-05,
"loss": 1.1602,
"step": 655
},
{
"epoch": 0.58,
"learning_rate": 1.9534584088502946e-05,
"loss": 1.1648,
"step": 660
},
{
"epoch": 0.58,
"learning_rate": 1.952607358198996e-05,
"loss": 1.1559,
"step": 665
},
{
"epoch": 0.59,
"learning_rate": 1.9517487856090923e-05,
"loss": 1.1431,
"step": 670
},
{
"epoch": 0.59,
"learning_rate": 1.9508826978600095e-05,
"loss": 1.1665,
"step": 675
},
{
"epoch": 0.6,
"learning_rate": 1.9500091017905133e-05,
"loss": 1.1734,
"step": 680
},
{
"epoch": 0.6,
"learning_rate": 1.9491280042986566e-05,
"loss": 1.154,
"step": 685
},
{
"epoch": 0.6,
"learning_rate": 1.9482394123417246e-05,
"loss": 1.1555,
"step": 690
},
{
"epoch": 0.61,
"learning_rate": 1.9473433329361805e-05,
"loss": 1.1486,
"step": 695
},
{
"epoch": 0.61,
"learning_rate": 1.9464397731576093e-05,
"loss": 1.1609,
"step": 700
},
{
"epoch": 0.61,
"eval_loss": 1.160956859588623,
"eval_runtime": 426.6054,
"eval_samples_per_second": 37.895,
"eval_steps_per_second": 1.186,
"step": 700
},
{
"epoch": 0.62,
"learning_rate": 1.945528740140662e-05,
"loss": 1.1584,
"step": 705
},
{
"epoch": 0.62,
"learning_rate": 1.944610241079e-05,
"loss": 1.1311,
"step": 710
},
{
"epoch": 0.63,
"learning_rate": 1.9436842832252363e-05,
"loss": 1.1384,
"step": 715
},
{
"epoch": 0.63,
"learning_rate": 1.9427508738908817e-05,
"loss": 1.1683,
"step": 720
},
{
"epoch": 0.64,
"learning_rate": 1.9418100204462843e-05,
"loss": 1.1368,
"step": 725
},
{
"epoch": 0.64,
"learning_rate": 1.9408617303205712e-05,
"loss": 1.1778,
"step": 730
},
{
"epoch": 0.64,
"learning_rate": 1.9399060110015918e-05,
"loss": 1.1624,
"step": 735
},
{
"epoch": 0.65,
"learning_rate": 1.9389428700358575e-05,
"loss": 1.1505,
"step": 740
},
{
"epoch": 0.65,
"learning_rate": 1.9379723150284817e-05,
"loss": 1.1638,
"step": 745
},
{
"epoch": 0.66,
"learning_rate": 1.9369943536431207e-05,
"loss": 1.1491,
"step": 750
},
{
"epoch": 0.66,
"learning_rate": 1.936008993601912e-05,
"loss": 1.1634,
"step": 755
},
{
"epoch": 0.67,
"learning_rate": 1.9350162426854152e-05,
"loss": 1.152,
"step": 760
},
{
"epoch": 0.67,
"learning_rate": 1.9340161087325483e-05,
"loss": 1.1364,
"step": 765
},
{
"epoch": 0.67,
"learning_rate": 1.933008599640528e-05,
"loss": 1.1371,
"step": 770
},
{
"epoch": 0.68,
"learning_rate": 1.9319937233648044e-05,
"loss": 1.143,
"step": 775
},
{
"epoch": 0.68,
"learning_rate": 1.9309714879190032e-05,
"loss": 1.1353,
"step": 780
},
{
"epoch": 0.69,
"learning_rate": 1.929941901374856e-05,
"loss": 1.1533,
"step": 785
},
{
"epoch": 0.69,
"learning_rate": 1.9289049718621423e-05,
"loss": 1.1305,
"step": 790
},
{
"epoch": 0.7,
"learning_rate": 1.9278607075686207e-05,
"loss": 1.1585,
"step": 795
},
{
"epoch": 0.7,
"learning_rate": 1.9268091167399682e-05,
"loss": 1.1449,
"step": 800
},
{
"epoch": 0.7,
"eval_loss": 1.1543163061141968,
"eval_runtime": 428.7409,
"eval_samples_per_second": 37.706,
"eval_steps_per_second": 1.18,
"step": 800
},
{
"epoch": 0.71,
"learning_rate": 1.9257502076797125e-05,
"loss": 1.1294,
"step": 805
},
{
"epoch": 0.71,
"learning_rate": 1.924683988749167e-05,
"loss": 1.1343,
"step": 810
},
{
"epoch": 0.71,
"learning_rate": 1.9236104683673656e-05,
"loss": 1.1317,
"step": 815
},
{
"epoch": 0.72,
"learning_rate": 1.9225296550109948e-05,
"loss": 1.1248,
"step": 820
},
{
"epoch": 0.72,
"learning_rate": 1.921441557214329e-05,
"loss": 1.1472,
"step": 825
},
{
"epoch": 0.73,
"learning_rate": 1.9203461835691596e-05,
"loss": 1.1421,
"step": 830
},
{
"epoch": 0.73,
"learning_rate": 1.9192435427247313e-05,
"loss": 1.1666,
"step": 835
},
{
"epoch": 0.74,
"learning_rate": 1.9181336433876704e-05,
"loss": 1.1441,
"step": 840
},
{
"epoch": 0.74,
"learning_rate": 1.9170164943219186e-05,
"loss": 1.1591,
"step": 845
},
{
"epoch": 0.74,
"learning_rate": 1.915892104348661e-05,
"loss": 1.1379,
"step": 850
},
{
"epoch": 0.75,
"learning_rate": 1.914760482346259e-05,
"loss": 1.1503,
"step": 855
},
{
"epoch": 0.75,
"learning_rate": 1.91362163725018e-05,
"loss": 1.1451,
"step": 860
},
{
"epoch": 0.76,
"learning_rate": 1.9124755780529243e-05,
"loss": 1.1601,
"step": 865
},
{
"epoch": 0.76,
"learning_rate": 1.911322313803958e-05,
"loss": 1.1516,
"step": 870
},
{
"epoch": 0.77,
"learning_rate": 1.910161853609637e-05,
"loss": 1.133,
"step": 875
},
{
"epoch": 0.77,
"learning_rate": 1.90899420663314e-05,
"loss": 1.1461,
"step": 880
},
{
"epoch": 0.78,
"learning_rate": 1.9078193820943916e-05,
"loss": 1.1315,
"step": 885
},
{
"epoch": 0.78,
"learning_rate": 1.9066373892699933e-05,
"loss": 1.1316,
"step": 890
},
{
"epoch": 0.78,
"learning_rate": 1.905448237493147e-05,
"loss": 1.1372,
"step": 895
},
{
"epoch": 0.79,
"learning_rate": 1.9042519361535835e-05,
"loss": 1.1406,
"step": 900
},
{
"epoch": 0.79,
"eval_loss": 1.148454189300537,
"eval_runtime": 428.8248,
"eval_samples_per_second": 37.698,
"eval_steps_per_second": 1.18,
"step": 900
},
{
"epoch": 0.79,
"learning_rate": 1.903048494697488e-05,
"loss": 1.1461,
"step": 905
},
{
"epoch": 0.8,
"learning_rate": 1.901837922627425e-05,
"loss": 1.143,
"step": 910
},
{
"epoch": 0.8,
"learning_rate": 1.9006202295022632e-05,
"loss": 1.1558,
"step": 915
},
{
"epoch": 0.81,
"learning_rate": 1.8993954249371e-05,
"loss": 1.1318,
"step": 920
},
{
"epoch": 0.81,
"learning_rate": 1.898163518603187e-05,
"loss": 1.1409,
"step": 925
},
{
"epoch": 0.81,
"learning_rate": 1.896924520227851e-05,
"loss": 1.1313,
"step": 930
},
{
"epoch": 0.82,
"learning_rate": 1.89567843959442e-05,
"loss": 1.1545,
"step": 935
},
{
"epoch": 0.82,
"learning_rate": 1.894425286542144e-05,
"loss": 1.1453,
"step": 940
},
{
"epoch": 0.83,
"learning_rate": 1.8931650709661178e-05,
"loss": 1.1461,
"step": 945
},
{
"epoch": 0.83,
"learning_rate": 1.8918978028172034e-05,
"loss": 1.1554,
"step": 950
},
{
"epoch": 0.84,
"learning_rate": 1.8906234921019505e-05,
"loss": 1.1522,
"step": 955
},
{
"epoch": 0.84,
"learning_rate": 1.889342148882519e-05,
"loss": 1.1349,
"step": 960
},
{
"epoch": 0.85,
"learning_rate": 1.8880537832765976e-05,
"loss": 1.1301,
"step": 965
},
{
"epoch": 0.85,
"learning_rate": 1.886758405457325e-05,
"loss": 1.1382,
"step": 970
},
{
"epoch": 0.85,
"learning_rate": 1.8854560256532098e-05,
"loss": 1.152,
"step": 975
},
{
"epoch": 0.86,
"learning_rate": 1.88414665414805e-05,
"loss": 1.1468,
"step": 980
},
{
"epoch": 0.86,
"learning_rate": 1.8828303012808492e-05,
"loss": 1.1304,
"step": 985
},
{
"epoch": 0.87,
"learning_rate": 1.8815069774457386e-05,
"loss": 1.1337,
"step": 990
},
{
"epoch": 0.87,
"learning_rate": 1.8801766930918934e-05,
"loss": 1.1495,
"step": 995
},
{
"epoch": 0.88,
"learning_rate": 1.8788394587234486e-05,
"loss": 1.1241,
"step": 1000
},
{
"epoch": 0.88,
"eval_loss": 1.1431844234466553,
"eval_runtime": 426.7827,
"eval_samples_per_second": 37.879,
"eval_steps_per_second": 1.186,
"step": 1000
},
{
"epoch": 0.88,
"learning_rate": 1.8774952848994195e-05,
"loss": 1.1251,
"step": 1005
},
{
"epoch": 0.88,
"learning_rate": 1.8761441822336152e-05,
"loss": 1.131,
"step": 1010
},
{
"epoch": 0.89,
"learning_rate": 1.874786161394556e-05,
"loss": 1.1257,
"step": 1015
},
{
"epoch": 0.89,
"learning_rate": 1.8734212331053904e-05,
"loss": 1.1233,
"step": 1020
},
{
"epoch": 0.9,
"learning_rate": 1.872049408143808e-05,
"loss": 1.1417,
"step": 1025
},
{
"epoch": 0.9,
"learning_rate": 1.8706706973419555e-05,
"loss": 1.1215,
"step": 1030
},
{
"epoch": 0.91,
"learning_rate": 1.8692851115863524e-05,
"loss": 1.1322,
"step": 1035
},
{
"epoch": 0.91,
"learning_rate": 1.8678926618178034e-05,
"loss": 1.1415,
"step": 1040
},
{
"epoch": 0.92,
"learning_rate": 1.866493359031312e-05,
"loss": 1.1396,
"step": 1045
},
{
"epoch": 0.92,
"learning_rate": 1.8650872142759947e-05,
"loss": 1.1322,
"step": 1050
},
{
"epoch": 0.92,
"learning_rate": 1.863674238654994e-05,
"loss": 1.1262,
"step": 1055
},
{
"epoch": 0.93,
"learning_rate": 1.8622544433253883e-05,
"loss": 1.1701,
"step": 1060
},
{
"epoch": 0.93,
"learning_rate": 1.8608278394981068e-05,
"loss": 1.1395,
"step": 1065
},
{
"epoch": 0.94,
"learning_rate": 1.859394438437839e-05,
"loss": 1.1292,
"step": 1070
},
{
"epoch": 0.94,
"learning_rate": 1.8579542514629476e-05,
"loss": 1.1242,
"step": 1075
},
{
"epoch": 0.95,
"learning_rate": 1.8565072899453763e-05,
"loss": 1.1334,
"step": 1080
},
{
"epoch": 0.95,
"learning_rate": 1.8550535653105625e-05,
"loss": 1.1194,
"step": 1085
},
{
"epoch": 0.95,
"learning_rate": 1.8535930890373467e-05,
"loss": 1.118,
"step": 1090
},
{
"epoch": 0.96,
"learning_rate": 1.8521258726578804e-05,
"loss": 1.1354,
"step": 1095
},
{
"epoch": 0.96,
"learning_rate": 1.850651927757537e-05,
"loss": 1.1477,
"step": 1100
},
{
"epoch": 0.96,
"eval_loss": 1.138464331626892,
"eval_runtime": 427.2443,
"eval_samples_per_second": 37.838,
"eval_steps_per_second": 1.184,
"step": 1100
},
{
"epoch": 0.97,
"learning_rate": 1.8491712659748183e-05,
"loss": 1.1413,
"step": 1105
},
{
"epoch": 0.97,
"learning_rate": 1.847683899001265e-05,
"loss": 1.1143,
"step": 1110
},
{
"epoch": 0.98,
"learning_rate": 1.8461898385813622e-05,
"loss": 1.109,
"step": 1115
},
{
"epoch": 0.98,
"learning_rate": 1.8446890965124472e-05,
"loss": 1.1249,
"step": 1120
},
{
"epoch": 0.99,
"learning_rate": 1.8431816846446173e-05,
"loss": 1.119,
"step": 1125
},
{
"epoch": 0.99,
"learning_rate": 1.8416676148806356e-05,
"loss": 1.1372,
"step": 1130
},
{
"epoch": 0.99,
"learning_rate": 1.8401468991758368e-05,
"loss": 1.1256,
"step": 1135
},
{
"epoch": 1.0,
"learning_rate": 1.8386195495380324e-05,
"loss": 1.1361,
"step": 1140
},
{
"epoch": 1.0,
"learning_rate": 1.8370855780274182e-05,
"loss": 1.1013,
"step": 1145
},
{
"epoch": 1.01,
"learning_rate": 1.8355449967564754e-05,
"loss": 1.0445,
"step": 1150
},
{
"epoch": 1.01,
"learning_rate": 1.833997817889878e-05,
"loss": 1.0758,
"step": 1155
},
{
"epoch": 1.02,
"learning_rate": 1.8324440536443956e-05,
"loss": 1.084,
"step": 1160
},
{
"epoch": 1.02,
"learning_rate": 1.8308837162887964e-05,
"loss": 1.0677,
"step": 1165
},
{
"epoch": 1.02,
"learning_rate": 1.8293168181437516e-05,
"loss": 1.0827,
"step": 1170
},
{
"epoch": 1.03,
"learning_rate": 1.827743371581737e-05,
"loss": 1.0782,
"step": 1175
},
{
"epoch": 1.03,
"learning_rate": 1.8261633890269363e-05,
"loss": 1.0785,
"step": 1180
},
{
"epoch": 1.04,
"learning_rate": 1.8245768829551417e-05,
"loss": 1.0771,
"step": 1185
},
{
"epoch": 1.04,
"learning_rate": 1.8229838658936566e-05,
"loss": 1.0785,
"step": 1190
},
{
"epoch": 1.05,
"learning_rate": 1.821384350421196e-05,
"loss": 1.0894,
"step": 1195
},
{
"epoch": 1.05,
"learning_rate": 1.8197783491677873e-05,
"loss": 1.0644,
"step": 1200
},
{
"epoch": 1.05,
"eval_loss": 1.1381514072418213,
"eval_runtime": 427.6281,
"eval_samples_per_second": 37.804,
"eval_steps_per_second": 1.183,
"step": 1200
},
{
"epoch": 1.06,
"learning_rate": 1.818165874814671e-05,
"loss": 1.0904,
"step": 1205
},
{
"epoch": 1.06,
"learning_rate": 1.8165469400942e-05,
"loss": 1.0588,
"step": 1210
},
{
"epoch": 1.06,
"learning_rate": 1.8149215577897397e-05,
"loss": 1.0787,
"step": 1215
},
{
"epoch": 1.07,
"learning_rate": 1.8132897407355657e-05,
"loss": 1.0742,
"step": 1220
},
{
"epoch": 1.07,
"learning_rate": 1.8116515018167636e-05,
"loss": 1.0821,
"step": 1225
},
{
"epoch": 1.08,
"learning_rate": 1.8100068539691283e-05,
"loss": 1.0653,
"step": 1230
},
{
"epoch": 1.08,
"learning_rate": 1.8083558101790598e-05,
"loss": 1.0636,
"step": 1235
},
{
"epoch": 1.09,
"learning_rate": 1.8066983834834607e-05,
"loss": 1.0761,
"step": 1240
},
{
"epoch": 1.09,
"learning_rate": 1.805034586969635e-05,
"loss": 1.0729,
"step": 1245
},
{
"epoch": 1.1,
"learning_rate": 1.8033644337751833e-05,
"loss": 1.077,
"step": 1250
},
{
"epoch": 1.1,
"learning_rate": 1.8016879370879006e-05,
"loss": 1.0904,
"step": 1255
},
{
"epoch": 1.1,
"learning_rate": 1.800005110145669e-05,
"loss": 1.0804,
"step": 1260
},
{
"epoch": 1.11,
"learning_rate": 1.7983159662363582e-05,
"loss": 1.0639,
"step": 1265
},
{
"epoch": 1.11,
"learning_rate": 1.7966205186977154e-05,
"loss": 1.0558,
"step": 1270
},
{
"epoch": 1.12,
"learning_rate": 1.7949187809172623e-05,
"loss": 1.0719,
"step": 1275
},
{
"epoch": 1.12,
"learning_rate": 1.793210766332191e-05,
"loss": 1.0852,
"step": 1280
},
{
"epoch": 1.13,
"learning_rate": 1.7914964884292543e-05,
"loss": 1.0774,
"step": 1285
},
{
"epoch": 1.13,
"learning_rate": 1.7897759607446627e-05,
"loss": 1.0782,
"step": 1290
},
{
"epoch": 1.13,
"learning_rate": 1.788049196863975e-05,
"loss": 1.0647,
"step": 1295
},
{
"epoch": 1.14,
"learning_rate": 1.786316210421993e-05,
"loss": 1.067,
"step": 1300
},
{
"epoch": 1.14,
"eval_loss": 1.1359283924102783,
"eval_runtime": 427.5945,
"eval_samples_per_second": 37.807,
"eval_steps_per_second": 1.183,
"step": 1300
},
{
"epoch": 1.14,
"learning_rate": 1.7845770151026516e-05,
"loss": 1.0729,
"step": 1305
},
{
"epoch": 1.15,
"learning_rate": 1.7828316246389125e-05,
"loss": 1.0597,
"step": 1310
},
{
"epoch": 1.15,
"learning_rate": 1.7810800528126556e-05,
"loss": 1.0795,
"step": 1315
},
{
"epoch": 1.16,
"learning_rate": 1.7793223134545692e-05,
"loss": 1.0637,
"step": 1320
},
{
"epoch": 1.16,
"learning_rate": 1.777558420444042e-05,
"loss": 1.0759,
"step": 1325
},
{
"epoch": 1.17,
"learning_rate": 1.7757883877090526e-05,
"loss": 1.0615,
"step": 1330
},
{
"epoch": 1.17,
"learning_rate": 1.7740122292260594e-05,
"loss": 1.0725,
"step": 1335
},
{
"epoch": 1.17,
"learning_rate": 1.772229959019892e-05,
"loss": 1.0734,
"step": 1340
},
{
"epoch": 1.18,
"learning_rate": 1.7704415911636378e-05,
"loss": 1.0969,
"step": 1345
},
{
"epoch": 1.18,
"learning_rate": 1.7686471397785322e-05,
"loss": 1.0553,
"step": 1350
},
{
"epoch": 1.19,
"learning_rate": 1.7668466190338484e-05,
"loss": 1.0795,
"step": 1355
},
{
"epoch": 1.19,
"learning_rate": 1.7650400431467832e-05,
"loss": 1.0608,
"step": 1360
},
{
"epoch": 1.2,
"learning_rate": 1.7632274263823458e-05,
"loss": 1.0866,
"step": 1365
},
{
"epoch": 1.2,
"learning_rate": 1.7614087830532453e-05,
"loss": 1.0733,
"step": 1370
},
{
"epoch": 1.2,
"learning_rate": 1.7595841275197783e-05,
"loss": 1.0775,
"step": 1375
},
{
"epoch": 1.21,
"learning_rate": 1.7577534741897126e-05,
"loss": 1.0755,
"step": 1380
},
{
"epoch": 1.21,
"learning_rate": 1.7559168375181777e-05,
"loss": 1.0684,
"step": 1385
},
{
"epoch": 1.22,
"learning_rate": 1.7540742320075476e-05,
"loss": 1.0711,
"step": 1390
},
{
"epoch": 1.22,
"learning_rate": 1.7522256722073273e-05,
"loss": 1.0724,
"step": 1395
},
{
"epoch": 1.23,
"learning_rate": 1.7503711727140384e-05,
"loss": 1.0791,
"step": 1400
},
{
"epoch": 1.23,
"eval_loss": 1.1332117319107056,
"eval_runtime": 427.2793,
"eval_samples_per_second": 37.835,
"eval_steps_per_second": 1.184,
"step": 1400
},
{
"epoch": 1.23,
"learning_rate": 1.7485107481711014e-05,
"loss": 1.0694,
"step": 1405
},
{
"epoch": 1.24,
"learning_rate": 1.746644413268724e-05,
"loss": 1.1044,
"step": 1410
},
{
"epoch": 1.24,
"learning_rate": 1.744772182743782e-05,
"loss": 1.0771,
"step": 1415
},
{
"epoch": 1.24,
"learning_rate": 1.7428940713797044e-05,
"loss": 1.077,
"step": 1420
},
{
"epoch": 1.25,
"learning_rate": 1.741010094006356e-05,
"loss": 1.0789,
"step": 1425
},
{
"epoch": 1.25,
"learning_rate": 1.7391202654999207e-05,
"loss": 1.0627,
"step": 1430
},
{
"epoch": 1.26,
"learning_rate": 1.7372246007827837e-05,
"loss": 1.0809,
"step": 1435
},
{
"epoch": 1.26,
"learning_rate": 1.7353231148234134e-05,
"loss": 1.0634,
"step": 1440
},
{
"epoch": 1.27,
"learning_rate": 1.7334158226362446e-05,
"loss": 1.0771,
"step": 1445
},
{
"epoch": 1.27,
"learning_rate": 1.7315027392815588e-05,
"loss": 1.0551,
"step": 1450
},
{
"epoch": 1.27,
"learning_rate": 1.729583879865365e-05,
"loss": 1.0719,
"step": 1455
},
{
"epoch": 1.28,
"learning_rate": 1.727659259539281e-05,
"loss": 1.0804,
"step": 1460
},
{
"epoch": 1.28,
"learning_rate": 1.7257288935004134e-05,
"loss": 1.0687,
"step": 1465
},
{
"epoch": 1.29,
"learning_rate": 1.7237927969912393e-05,
"loss": 1.064,
"step": 1470
},
{
"epoch": 1.29,
"learning_rate": 1.7218509852994825e-05,
"loss": 1.075,
"step": 1475
},
{
"epoch": 1.3,
"learning_rate": 1.7199034737579962e-05,
"loss": 1.0597,
"step": 1480
},
{
"epoch": 1.3,
"learning_rate": 1.7179502777446395e-05,
"loss": 1.0519,
"step": 1485
},
{
"epoch": 1.31,
"learning_rate": 1.715991412682158e-05,
"loss": 1.0667,
"step": 1490
},
{
"epoch": 1.31,
"learning_rate": 1.7140268940380608e-05,
"loss": 1.075,
"step": 1495
},
{
"epoch": 1.31,
"learning_rate": 1.7120567373244976e-05,
"loss": 1.0702,
"step": 1500
},
{
"epoch": 1.31,
"eval_loss": 1.130350947380066,
"eval_runtime": 436.0673,
"eval_samples_per_second": 37.072,
"eval_steps_per_second": 1.16,
"step": 1500
},
{
"epoch": 1.32,
"learning_rate": 1.7100809580981384e-05,
"loss": 1.0609,
"step": 1505
},
{
"epoch": 1.32,
"learning_rate": 1.708099571960049e-05,
"loss": 1.0808,
"step": 1510
},
{
"epoch": 1.33,
"learning_rate": 1.706112594555568e-05,
"loss": 1.0828,
"step": 1515
},
{
"epoch": 1.33,
"learning_rate": 1.7041200415741838e-05,
"loss": 1.0941,
"step": 1520
},
{
"epoch": 1.34,
"learning_rate": 1.7021219287494103e-05,
"loss": 1.0698,
"step": 1525
},
{
"epoch": 1.34,
"learning_rate": 1.7001182718586625e-05,
"loss": 1.0732,
"step": 1530
},
{
"epoch": 1.34,
"learning_rate": 1.6981090867231337e-05,
"loss": 1.0748,
"step": 1535
},
{
"epoch": 1.35,
"learning_rate": 1.6960943892076672e-05,
"loss": 1.0659,
"step": 1540
},
{
"epoch": 1.35,
"learning_rate": 1.6940741952206342e-05,
"loss": 1.0568,
"step": 1545
},
{
"epoch": 1.36,
"learning_rate": 1.692048520713806e-05,
"loss": 1.067,
"step": 1550
},
{
"epoch": 1.36,
"learning_rate": 1.690017381682229e-05,
"loss": 1.0706,
"step": 1555
},
{
"epoch": 1.37,
"learning_rate": 1.6879807941641e-05,
"loss": 1.0869,
"step": 1560
},
{
"epoch": 1.37,
"learning_rate": 1.6859387742406358e-05,
"loss": 1.0698,
"step": 1565
},
{
"epoch": 1.38,
"learning_rate": 1.6838913380359492e-05,
"loss": 1.0719,
"step": 1570
},
{
"epoch": 1.38,
"learning_rate": 1.6818385017169213e-05,
"loss": 1.0812,
"step": 1575
},
{
"epoch": 1.38,
"learning_rate": 1.6797802814930723e-05,
"loss": 1.082,
"step": 1580
},
{
"epoch": 1.39,
"learning_rate": 1.6777166936164354e-05,
"loss": 1.0631,
"step": 1585
},
{
"epoch": 1.39,
"learning_rate": 1.6756477543814276e-05,
"loss": 1.0807,
"step": 1590
},
{
"epoch": 1.4,
"learning_rate": 1.6735734801247204e-05,
"loss": 1.0718,
"step": 1595
},
{
"epoch": 1.4,
"learning_rate": 1.671493887225112e-05,
"loss": 1.0741,
"step": 1600
},
{
"epoch": 1.4,
"eval_loss": 1.1276899576187134,
"eval_runtime": 427.1768,
"eval_samples_per_second": 37.844,
"eval_steps_per_second": 1.185,
"step": 1600
},
{
"epoch": 1.41,
"learning_rate": 1.6694089921033977e-05,
"loss": 1.0777,
"step": 1605
},
{
"epoch": 1.41,
"learning_rate": 1.6673188112222394e-05,
"loss": 1.0657,
"step": 1610
},
{
"epoch": 1.41,
"learning_rate": 1.6652233610860368e-05,
"loss": 1.0724,
"step": 1615
},
{
"epoch": 1.42,
"learning_rate": 1.6631226582407954e-05,
"loss": 1.0614,
"step": 1620
},
{
"epoch": 1.42,
"learning_rate": 1.661016719273998e-05,
"loss": 1.0787,
"step": 1625
},
{
"epoch": 1.43,
"learning_rate": 1.6589055608144727e-05,
"loss": 1.0711,
"step": 1630
},
{
"epoch": 1.43,
"learning_rate": 1.6567891995322605e-05,
"loss": 1.0648,
"step": 1635
},
{
"epoch": 1.44,
"learning_rate": 1.6546676521384856e-05,
"loss": 1.0663,
"step": 1640
},
{
"epoch": 1.44,
"learning_rate": 1.6525409353852226e-05,
"loss": 1.066,
"step": 1645
},
{
"epoch": 1.45,
"learning_rate": 1.650409066065363e-05,
"loss": 1.0708,
"step": 1650
},
{
"epoch": 1.45,
"learning_rate": 1.6482720610124858e-05,
"loss": 1.0597,
"step": 1655
},
{
"epoch": 1.45,
"learning_rate": 1.646129937100721e-05,
"loss": 1.0627,
"step": 1660
},
{
"epoch": 1.46,
"learning_rate": 1.6439827112446177e-05,
"loss": 1.0841,
"step": 1665
},
{
"epoch": 1.46,
"learning_rate": 1.641830400399012e-05,
"loss": 1.068,
"step": 1670
},
{
"epoch": 1.47,
"learning_rate": 1.6396730215588913e-05,
"loss": 1.0662,
"step": 1675
},
{
"epoch": 1.47,
"learning_rate": 1.6375105917592607e-05,
"loss": 1.0736,
"step": 1680
},
{
"epoch": 1.48,
"learning_rate": 1.6353431280750083e-05,
"loss": 1.0515,
"step": 1685
},
{
"epoch": 1.48,
"learning_rate": 1.6331706476207708e-05,
"loss": 1.0649,
"step": 1690
},
{
"epoch": 1.48,
"learning_rate": 1.630993167550798e-05,
"loss": 1.0531,
"step": 1695
},
{
"epoch": 1.49,
"learning_rate": 1.6288107050588172e-05,
"loss": 1.0701,
"step": 1700
},
{
"epoch": 1.49,
"eval_loss": 1.1250778436660767,
"eval_runtime": 427.8614,
"eval_samples_per_second": 37.783,
"eval_steps_per_second": 1.183,
"step": 1700
},
{
"epoch": 1.49,
"learning_rate": 1.6266232773778984e-05,
"loss": 1.0658,
"step": 1705
},
{
"epoch": 1.5,
"learning_rate": 1.624430901780317e-05,
"loss": 1.0689,
"step": 1710
},
{
"epoch": 1.5,
"learning_rate": 1.6222335955774178e-05,
"loss": 1.0546,
"step": 1715
},
{
"epoch": 1.51,
"learning_rate": 1.6200313761194785e-05,
"loss": 1.083,
"step": 1720
},
{
"epoch": 1.51,
"learning_rate": 1.617824260795573e-05,
"loss": 1.0769,
"step": 1725
},
{
"epoch": 1.52,
"learning_rate": 1.615612267033434e-05,
"loss": 1.0731,
"step": 1730
},
{
"epoch": 1.52,
"learning_rate": 1.613395412299314e-05,
"loss": 1.057,
"step": 1735
},
{
"epoch": 1.52,
"learning_rate": 1.6111737140978495e-05,
"loss": 1.0554,
"step": 1740
},
{
"epoch": 1.53,
"learning_rate": 1.6089471899719213e-05,
"loss": 1.0803,
"step": 1745
},
{
"epoch": 1.53,
"learning_rate": 1.6067158575025165e-05,
"loss": 1.0519,
"step": 1750
},
{
"epoch": 1.54,
"learning_rate": 1.6044797343085902e-05,
"loss": 1.0697,
"step": 1755
},
{
"epoch": 1.54,
"learning_rate": 1.6022388380469248e-05,
"loss": 1.0799,
"step": 1760
},
{
"epoch": 1.55,
"learning_rate": 1.5999931864119923e-05,
"loss": 1.0697,
"step": 1765
},
{
"epoch": 1.55,
"learning_rate": 1.597742797135814e-05,
"loss": 1.0532,
"step": 1770
},
{
"epoch": 1.55,
"learning_rate": 1.59548768798782e-05,
"loss": 1.0733,
"step": 1775
},
{
"epoch": 1.56,
"learning_rate": 1.5932278767747092e-05,
"loss": 1.0767,
"step": 1780
},
{
"epoch": 1.56,
"learning_rate": 1.5909633813403094e-05,
"loss": 1.0705,
"step": 1785
},
{
"epoch": 1.57,
"learning_rate": 1.588694219565435e-05,
"loss": 1.0599,
"step": 1790
},
{
"epoch": 1.57,
"learning_rate": 1.5864204093677464e-05,
"loss": 1.0655,
"step": 1795
},
{
"epoch": 1.58,
"learning_rate": 1.5841419687016097e-05,
"loss": 1.0529,
"step": 1800
},
{
"epoch": 1.58,
"eval_loss": 1.122536540031433,
"eval_runtime": 426.9183,
"eval_samples_per_second": 37.867,
"eval_steps_per_second": 1.185,
"step": 1800
},
{
"epoch": 1.58,
"learning_rate": 1.581858915557953e-05,
"loss": 1.0673,
"step": 1805
},
{
"epoch": 1.59,
"learning_rate": 1.5795712679641257e-05,
"loss": 1.0501,
"step": 1810
},
{
"epoch": 1.59,
"learning_rate": 1.5772790439837557e-05,
"loss": 1.0541,
"step": 1815
},
{
"epoch": 1.59,
"learning_rate": 1.5749822617166067e-05,
"loss": 1.083,
"step": 1820
},
{
"epoch": 1.6,
"learning_rate": 1.572680939298435e-05,
"loss": 1.0544,
"step": 1825
},
{
"epoch": 1.6,
"learning_rate": 1.570375094900847e-05,
"loss": 1.0502,
"step": 1830
},
{
"epoch": 1.61,
"learning_rate": 1.568064746731156e-05,
"loss": 1.0839,
"step": 1835
},
{
"epoch": 1.61,
"learning_rate": 1.5657499130322366e-05,
"loss": 1.0615,
"step": 1840
},
{
"epoch": 1.62,
"learning_rate": 1.5634306120823822e-05,
"loss": 1.0754,
"step": 1845
},
{
"epoch": 1.62,
"learning_rate": 1.5611068621951603e-05,
"loss": 1.0466,
"step": 1850
},
{
"epoch": 1.63,
"learning_rate": 1.558778681719269e-05,
"loss": 1.0547,
"step": 1855
},
{
"epoch": 1.63,
"learning_rate": 1.5564460890383897e-05,
"loss": 1.054,
"step": 1860
},
{
"epoch": 1.63,
"learning_rate": 1.5541091025710436e-05,
"loss": 1.0648,
"step": 1865
},
{
"epoch": 1.64,
"learning_rate": 1.551767740770446e-05,
"loss": 1.0655,
"step": 1870
},
{
"epoch": 1.64,
"learning_rate": 1.549422022124361e-05,
"loss": 1.0511,
"step": 1875
},
{
"epoch": 1.65,
"learning_rate": 1.5470719651549543e-05,
"loss": 1.0671,
"step": 1880
},
{
"epoch": 1.65,
"learning_rate": 1.544717588418648e-05,
"loss": 1.0455,
"step": 1885
},
{
"epoch": 1.66,
"learning_rate": 1.542358910505974e-05,
"loss": 1.0638,
"step": 1890
},
{
"epoch": 1.66,
"learning_rate": 1.5399959500414264e-05,
"loss": 1.0498,
"step": 1895
},
{
"epoch": 1.66,
"learning_rate": 1.5376287256833154e-05,
"loss": 1.072,
"step": 1900
},
{
"epoch": 1.66,
"eval_loss": 1.1199389696121216,
"eval_runtime": 427.9494,
"eval_samples_per_second": 37.775,
"eval_steps_per_second": 1.182,
"step": 1900
},
{
"epoch": 1.67,
"learning_rate": 1.53525725612362e-05,
"loss": 1.0721,
"step": 1905
},
{
"epoch": 1.67,
"learning_rate": 1.532881560087839e-05,
"loss": 1.0395,
"step": 1910
},
{
"epoch": 1.68,
"learning_rate": 1.5305016563348444e-05,
"loss": 1.046,
"step": 1915
},
{
"epoch": 1.68,
"learning_rate": 1.528117563656734e-05,
"loss": 1.0511,
"step": 1920
},
{
"epoch": 1.69,
"learning_rate": 1.525729300878681e-05,
"loss": 1.0655,
"step": 1925
},
{
"epoch": 1.69,
"learning_rate": 1.5233368868587865e-05,
"loss": 1.0438,
"step": 1930
},
{
"epoch": 1.7,
"learning_rate": 1.5209403404879305e-05,
"loss": 1.0682,
"step": 1935
},
{
"epoch": 1.7,
"learning_rate": 1.5185396806896231e-05,
"loss": 1.085,
"step": 1940
},
{
"epoch": 1.7,
"learning_rate": 1.5161349264198537e-05,
"loss": 1.0628,
"step": 1945
},
{
"epoch": 1.71,
"learning_rate": 1.5137260966669436e-05,
"loss": 1.0491,
"step": 1950
},
{
"epoch": 1.71,
"learning_rate": 1.511313210451394e-05,
"loss": 1.0657,
"step": 1955
},
{
"epoch": 1.72,
"learning_rate": 1.5088962868257367e-05,
"loss": 1.0511,
"step": 1960
},
{
"epoch": 1.72,
"learning_rate": 1.5064753448743834e-05,
"loss": 1.0701,
"step": 1965
},
{
"epoch": 1.73,
"learning_rate": 1.5040504037134751e-05,
"loss": 1.04,
"step": 1970
},
{
"epoch": 1.73,
"learning_rate": 1.5016214824907315e-05,
"loss": 1.0782,
"step": 1975
},
{
"epoch": 1.73,
"learning_rate": 1.4991886003852995e-05,
"loss": 1.0524,
"step": 1980
},
{
"epoch": 1.74,
"learning_rate": 1.4967517766076017e-05,
"loss": 1.0656,
"step": 1985
},
{
"epoch": 1.74,
"learning_rate": 1.4943110303991844e-05,
"loss": 1.0691,
"step": 1990
},
{
"epoch": 1.75,
"learning_rate": 1.491866381032566e-05,
"loss": 1.0636,
"step": 1995
},
{
"epoch": 1.75,
"learning_rate": 1.4894178478110856e-05,
"loss": 1.0759,
"step": 2000
},
{
"epoch": 1.75,
"eval_loss": 1.1177895069122314,
"eval_runtime": 437.3653,
"eval_samples_per_second": 36.962,
"eval_steps_per_second": 1.157,
"step": 2000
},
{
"epoch": 1.76,
"learning_rate": 1.4869654500687494e-05,
"loss": 1.0605,
"step": 2005
},
{
"epoch": 1.76,
"learning_rate": 1.4845092071700786e-05,
"loss": 1.0579,
"step": 2010
},
{
"epoch": 1.77,
"learning_rate": 1.4820491385099555e-05,
"loss": 1.066,
"step": 2015
},
{
"epoch": 1.77,
"learning_rate": 1.4795852635134727e-05,
"loss": 1.0677,
"step": 2020
},
{
"epoch": 1.77,
"learning_rate": 1.4771176016357771e-05,
"loss": 1.0636,
"step": 2025
},
{
"epoch": 1.78,
"learning_rate": 1.4746461723619181e-05,
"loss": 1.0701,
"step": 2030
},
{
"epoch": 1.78,
"learning_rate": 1.4721709952066925e-05,
"loss": 1.0632,
"step": 2035
},
{
"epoch": 1.79,
"learning_rate": 1.4696920897144915e-05,
"loss": 1.0444,
"step": 2040
},
{
"epoch": 1.79,
"learning_rate": 1.467209475459145e-05,
"loss": 1.0587,
"step": 2045
},
{
"epoch": 1.8,
"learning_rate": 1.4647231720437687e-05,
"loss": 1.0449,
"step": 2050
},
{
"epoch": 1.8,
"learning_rate": 1.4622331991006085e-05,
"loss": 1.0663,
"step": 2055
},
{
"epoch": 1.8,
"learning_rate": 1.4597395762908848e-05,
"loss": 1.0638,
"step": 2060
},
{
"epoch": 1.81,
"learning_rate": 1.4572423233046386e-05,
"loss": 1.0375,
"step": 2065
},
{
"epoch": 1.81,
"learning_rate": 1.4547414598605752e-05,
"loss": 1.0584,
"step": 2070
},
{
"epoch": 1.82,
"learning_rate": 1.4522370057059082e-05,
"loss": 1.0752,
"step": 2075
},
{
"epoch": 1.82,
"learning_rate": 1.449728980616204e-05,
"loss": 1.071,
"step": 2080
},
{
"epoch": 1.83,
"learning_rate": 1.4472174043952271e-05,
"loss": 1.0735,
"step": 2085
},
{
"epoch": 1.83,
"learning_rate": 1.4447022968747803e-05,
"loss": 1.0624,
"step": 2090
},
{
"epoch": 1.84,
"learning_rate": 1.4421836779145513e-05,
"loss": 1.0553,
"step": 2095
},
{
"epoch": 1.84,
"learning_rate": 1.4396615674019542e-05,
"loss": 1.0618,
"step": 2100
},
{
"epoch": 1.84,
"eval_loss": 1.1152375936508179,
"eval_runtime": 427.0732,
"eval_samples_per_second": 37.853,
"eval_steps_per_second": 1.185,
"step": 2100
},
{
"epoch": 1.84,
"learning_rate": 1.4371359852519736e-05,
"loss": 1.0677,
"step": 2105
},
{
"epoch": 1.85,
"learning_rate": 1.4346069514070055e-05,
"loss": 1.0532,
"step": 2110
},
{
"epoch": 1.85,
"learning_rate": 1.4320744858367026e-05,
"loss": 1.0847,
"step": 2115
},
{
"epoch": 1.86,
"learning_rate": 1.4295386085378138e-05,
"loss": 1.051,
"step": 2120
},
{
"epoch": 1.86,
"learning_rate": 1.4269993395340278e-05,
"loss": 1.0476,
"step": 2125
},
{
"epoch": 1.87,
"learning_rate": 1.4244566988758152e-05,
"loss": 1.0611,
"step": 2130
},
{
"epoch": 1.87,
"learning_rate": 1.4219107066402694e-05,
"loss": 1.0682,
"step": 2135
},
{
"epoch": 1.87,
"learning_rate": 1.419361382930948e-05,
"loss": 1.0576,
"step": 2140
},
{
"epoch": 1.88,
"learning_rate": 1.4168087478777156e-05,
"loss": 1.048,
"step": 2145
},
{
"epoch": 1.88,
"learning_rate": 1.4142528216365822e-05,
"loss": 1.0636,
"step": 2150
},
{
"epoch": 1.89,
"learning_rate": 1.4116936243895467e-05,
"loss": 1.067,
"step": 2155
},
{
"epoch": 1.89,
"learning_rate": 1.4091311763444358e-05,
"loss": 1.0427,
"step": 2160
},
{
"epoch": 1.9,
"learning_rate": 1.4065654977347452e-05,
"loss": 1.0617,
"step": 2165
},
{
"epoch": 1.9,
"learning_rate": 1.4039966088194802e-05,
"loss": 1.0517,
"step": 2170
},
{
"epoch": 1.91,
"learning_rate": 1.4014245298829936e-05,
"loss": 1.077,
"step": 2175
},
{
"epoch": 1.91,
"learning_rate": 1.398849281234829e-05,
"loss": 1.0654,
"step": 2180
},
{
"epoch": 1.91,
"learning_rate": 1.396270883209557e-05,
"loss": 1.0646,
"step": 2185
},
{
"epoch": 1.92,
"learning_rate": 1.3936893561666171e-05,
"loss": 1.0542,
"step": 2190
},
{
"epoch": 1.92,
"learning_rate": 1.391104720490156e-05,
"loss": 1.0652,
"step": 2195
},
{
"epoch": 1.93,
"learning_rate": 1.3885169965888667e-05,
"loss": 1.0359,
"step": 2200
},
{
"epoch": 1.93,
"eval_loss": 1.113387107849121,
"eval_runtime": 428.0736,
"eval_samples_per_second": 37.765,
"eval_steps_per_second": 1.182,
"step": 2200
},
{
"epoch": 1.93,
"learning_rate": 1.3859262048958261e-05,
"loss": 1.0768,
"step": 2205
},
{
"epoch": 1.94,
"learning_rate": 1.3833323658683364e-05,
"loss": 1.0545,
"step": 2210
},
{
"epoch": 1.94,
"learning_rate": 1.3807354999877615e-05,
"loss": 1.0601,
"step": 2215
},
{
"epoch": 1.94,
"learning_rate": 1.3781356277593659e-05,
"loss": 1.0707,
"step": 2220
},
{
"epoch": 1.95,
"learning_rate": 1.3755327697121523e-05,
"loss": 1.0516,
"step": 2225
},
{
"epoch": 1.95,
"learning_rate": 1.3729269463987002e-05,
"loss": 1.0575,
"step": 2230
},
{
"epoch": 1.96,
"learning_rate": 1.3703181783950032e-05,
"loss": 1.0548,
"step": 2235
},
{
"epoch": 1.96,
"learning_rate": 1.3677064863003069e-05,
"loss": 1.0801,
"step": 2240
},
{
"epoch": 1.97,
"learning_rate": 1.3650918907369455e-05,
"loss": 1.0647,
"step": 2245
},
{
"epoch": 1.97,
"learning_rate": 1.36247441235018e-05,
"loss": 1.0583,
"step": 2250
},
{
"epoch": 1.98,
"learning_rate": 1.3598540718080346e-05,
"loss": 1.0718,
"step": 2255
},
{
"epoch": 1.98,
"learning_rate": 1.3572308898011328e-05,
"loss": 1.0559,
"step": 2260
},
{
"epoch": 1.98,
"learning_rate": 1.3546048870425356e-05,
"loss": 1.0758,
"step": 2265
},
{
"epoch": 1.99,
"learning_rate": 1.3519760842675767e-05,
"loss": 1.0617,
"step": 2270
},
{
"epoch": 1.99,
"learning_rate": 1.3493445022336996e-05,
"loss": 1.0482,
"step": 2275
},
{
"epoch": 2.0,
"learning_rate": 1.3467101617202922e-05,
"loss": 1.052,
"step": 2280
},
{
"epoch": 2.0,
"learning_rate": 1.344073083528525e-05,
"loss": 1.0294,
"step": 2285
},
{
"epoch": 2.01,
"learning_rate": 1.3414332884811847e-05,
"loss": 0.9769,
"step": 2290
},
{
"epoch": 2.01,
"learning_rate": 1.3387907974225119e-05,
"loss": 0.9721,
"step": 2295
},
{
"epoch": 2.01,
"learning_rate": 1.3361456312180339e-05,
"loss": 0.9918,
"step": 2300
},
{
"epoch": 2.01,
"eval_loss": 1.1195367574691772,
"eval_runtime": 427.4072,
"eval_samples_per_second": 37.823,
"eval_steps_per_second": 1.184,
"step": 2300
},
{
"epoch": 2.02,
"learning_rate": 1.3334978107544026e-05,
"loss": 0.984,
"step": 2305
},
{
"epoch": 2.02,
"learning_rate": 1.3308473569392274e-05,
"loss": 0.9903,
"step": 2310
},
{
"epoch": 2.03,
"learning_rate": 1.3281942907009113e-05,
"loss": 0.9957,
"step": 2315
},
{
"epoch": 2.03,
"learning_rate": 1.3255386329884867e-05,
"loss": 0.9965,
"step": 2320
},
{
"epoch": 2.04,
"learning_rate": 1.3228804047714462e-05,
"loss": 0.9898,
"step": 2325
},
{
"epoch": 2.04,
"learning_rate": 1.320219627039582e-05,
"loss": 0.993,
"step": 2330
},
{
"epoch": 2.05,
"learning_rate": 1.3175563208028161e-05,
"loss": 0.982,
"step": 2335
},
{
"epoch": 2.05,
"learning_rate": 1.3148905070910363e-05,
"loss": 1.0008,
"step": 2340
},
{
"epoch": 2.05,
"learning_rate": 1.31222220695393e-05,
"loss": 0.9914,
"step": 2345
},
{
"epoch": 2.06,
"learning_rate": 1.3095514414608181e-05,
"loss": 0.9918,
"step": 2350
},
{
"epoch": 2.06,
"learning_rate": 1.3068782317004876e-05,
"loss": 0.9959,
"step": 2355
},
{
"epoch": 2.07,
"learning_rate": 1.3042025987810267e-05,
"loss": 0.9924,
"step": 2360
},
{
"epoch": 2.07,
"learning_rate": 1.3015245638296565e-05,
"loss": 0.9918,
"step": 2365
},
{
"epoch": 2.08,
"learning_rate": 1.2988441479925652e-05,
"loss": 0.9896,
"step": 2370
},
{
"epoch": 2.08,
"learning_rate": 1.296161372434741e-05,
"loss": 0.9795,
"step": 2375
},
{
"epoch": 2.08,
"learning_rate": 1.2934762583398043e-05,
"loss": 0.9971,
"step": 2380
},
{
"epoch": 2.09,
"learning_rate": 1.2907888269098418e-05,
"loss": 1.0019,
"step": 2385
},
{
"epoch": 2.09,
"learning_rate": 1.2880990993652379e-05,
"loss": 0.992,
"step": 2390
},
{
"epoch": 2.1,
"learning_rate": 1.2854070969445067e-05,
"loss": 0.9808,
"step": 2395
},
{
"epoch": 2.1,
"learning_rate": 1.2827128409041263e-05,
"loss": 1.002,
"step": 2400
},
{
"epoch": 2.1,
"eval_loss": 1.120516300201416,
"eval_runtime": 427.3661,
"eval_samples_per_second": 37.827,
"eval_steps_per_second": 1.184,
"step": 2400
},
{
"epoch": 2.11,
"learning_rate": 1.2800163525183689e-05,
"loss": 0.9872,
"step": 2405
},
{
"epoch": 2.11,
"learning_rate": 1.2773176530791342e-05,
"loss": 0.9786,
"step": 2410
},
{
"epoch": 2.12,
"learning_rate": 1.2746167638957808e-05,
"loss": 0.9981,
"step": 2415
},
{
"epoch": 2.12,
"learning_rate": 1.2719137062949569e-05,
"loss": 1.0032,
"step": 2420
},
{
"epoch": 2.12,
"learning_rate": 1.2692085016204334e-05,
"loss": 0.9959,
"step": 2425
},
{
"epoch": 2.13,
"learning_rate": 1.2665011712329355e-05,
"loss": 0.996,
"step": 2430
},
{
"epoch": 2.13,
"learning_rate": 1.2637917365099726e-05,
"loss": 0.9809,
"step": 2435
},
{
"epoch": 2.14,
"learning_rate": 1.2610802188456708e-05,
"loss": 0.9971,
"step": 2440
},
{
"epoch": 2.14,
"learning_rate": 1.2583666396506024e-05,
"loss": 0.9926,
"step": 2445
},
{
"epoch": 2.15,
"learning_rate": 1.255651020351619e-05,
"loss": 0.9777,
"step": 2450
},
{
"epoch": 2.15,
"learning_rate": 1.2529333823916807e-05,
"loss": 0.9908,
"step": 2455
},
{
"epoch": 2.16,
"learning_rate": 1.2502137472296879e-05,
"loss": 0.98,
"step": 2460
},
{
"epoch": 2.16,
"learning_rate": 1.2474921363403096e-05,
"loss": 1.009,
"step": 2465
},
{
"epoch": 2.16,
"learning_rate": 1.2447685712138175e-05,
"loss": 0.9895,
"step": 2470
},
{
"epoch": 2.17,
"learning_rate": 1.2420430733559126e-05,
"loss": 1.0016,
"step": 2475
},
{
"epoch": 2.17,
"learning_rate": 1.2393156642875579e-05,
"loss": 0.9825,
"step": 2480
},
{
"epoch": 2.18,
"learning_rate": 1.2365863655448076e-05,
"loss": 0.9997,
"step": 2485
},
{
"epoch": 2.18,
"learning_rate": 1.2338551986786368e-05,
"loss": 0.9718,
"step": 2490
},
{
"epoch": 2.19,
"learning_rate": 1.2311221852547724e-05,
"loss": 0.9895,
"step": 2495
},
{
"epoch": 2.19,
"learning_rate": 1.2283873468535201e-05,
"loss": 0.993,
"step": 2500
},
{
"epoch": 2.19,
"eval_loss": 1.1193997859954834,
"eval_runtime": 426.4898,
"eval_samples_per_second": 37.905,
"eval_steps_per_second": 1.186,
"step": 2500
},
{
"epoch": 2.19,
"learning_rate": 1.2256507050695978e-05,
"loss": 0.985,
"step": 2505
},
{
"epoch": 2.2,
"learning_rate": 1.2229122815119629e-05,
"loss": 0.9976,
"step": 2510
},
{
"epoch": 2.2,
"learning_rate": 1.2201720978036411e-05,
"loss": 0.9921,
"step": 2515
},
{
"epoch": 2.21,
"learning_rate": 1.2174301755815572e-05,
"loss": 0.9818,
"step": 2520
},
{
"epoch": 2.21,
"learning_rate": 1.2146865364963634e-05,
"loss": 1.0,
"step": 2525
},
{
"epoch": 2.22,
"learning_rate": 1.2119412022122684e-05,
"loss": 1.0082,
"step": 2530
},
{
"epoch": 2.22,
"learning_rate": 1.2091941944068666e-05,
"loss": 0.9906,
"step": 2535
},
{
"epoch": 2.23,
"learning_rate": 1.2064455347709672e-05,
"loss": 0.9886,
"step": 2540
},
{
"epoch": 2.23,
"learning_rate": 1.2036952450084216e-05,
"loss": 0.978,
"step": 2545
},
{
"epoch": 2.23,
"learning_rate": 1.2009433468359539e-05,
"loss": 0.9916,
"step": 2550
},
{
"epoch": 2.24,
"learning_rate": 1.1981898619829879e-05,
"loss": 1.0023,
"step": 2555
},
{
"epoch": 2.24,
"learning_rate": 1.1954348121914758e-05,
"loss": 0.9728,
"step": 2560
},
{
"epoch": 2.25,
"learning_rate": 1.1926782192157276e-05,
"loss": 1.0034,
"step": 2565
},
{
"epoch": 2.25,
"learning_rate": 1.1899201048222377e-05,
"loss": 0.9884,
"step": 2570
},
{
"epoch": 2.26,
"learning_rate": 1.1871604907895149e-05,
"loss": 0.9777,
"step": 2575
},
{
"epoch": 2.26,
"learning_rate": 1.1843993989079081e-05,
"loss": 0.9932,
"step": 2580
},
{
"epoch": 2.26,
"learning_rate": 1.1816368509794365e-05,
"loss": 1.0011,
"step": 2585
},
{
"epoch": 2.27,
"learning_rate": 1.1788728688176163e-05,
"loss": 0.9914,
"step": 2590
},
{
"epoch": 2.27,
"learning_rate": 1.1761074742472882e-05,
"loss": 0.9718,
"step": 2595
},
{
"epoch": 2.28,
"learning_rate": 1.173340689104446e-05,
"loss": 0.9872,
"step": 2600
},
{
"epoch": 2.28,
"eval_loss": 1.1183725595474243,
"eval_runtime": 428.4998,
"eval_samples_per_second": 37.727,
"eval_steps_per_second": 1.181,
"step": 2600
},
{
"epoch": 2.28,
"learning_rate": 1.1705725352360635e-05,
"loss": 0.9991,
"step": 2605
},
{
"epoch": 2.29,
"learning_rate": 1.167803034499922e-05,
"loss": 0.9916,
"step": 2610
},
{
"epoch": 2.29,
"learning_rate": 1.165032208764438e-05,
"loss": 0.988,
"step": 2615
},
{
"epoch": 2.3,
"learning_rate": 1.1622600799084907e-05,
"loss": 0.994,
"step": 2620
},
{
"epoch": 2.3,
"learning_rate": 1.1594866698212484e-05,
"loss": 0.9644,
"step": 2625
},
{
"epoch": 2.3,
"learning_rate": 1.1567120004019967e-05,
"loss": 0.9953,
"step": 2630
},
{
"epoch": 2.31,
"learning_rate": 1.1539360935599646e-05,
"loss": 0.9918,
"step": 2635
},
{
"epoch": 2.31,
"learning_rate": 1.1511589712141527e-05,
"loss": 0.9922,
"step": 2640
},
{
"epoch": 2.32,
"learning_rate": 1.1483806552931585e-05,
"loss": 0.9901,
"step": 2645
},
{
"epoch": 2.32,
"learning_rate": 1.1456011677350052e-05,
"loss": 0.9851,
"step": 2650
},
{
"epoch": 2.33,
"learning_rate": 1.1428205304869662e-05,
"loss": 0.9882,
"step": 2655
},
{
"epoch": 2.33,
"learning_rate": 1.1400387655053942e-05,
"loss": 1.0009,
"step": 2660
},
{
"epoch": 2.33,
"learning_rate": 1.1372558947555457e-05,
"loss": 0.9967,
"step": 2665
},
{
"epoch": 2.34,
"learning_rate": 1.1344719402114093e-05,
"loss": 1.0144,
"step": 2670
},
{
"epoch": 2.34,
"learning_rate": 1.1316869238555312e-05,
"loss": 0.985,
"step": 2675
},
{
"epoch": 2.35,
"learning_rate": 1.1289008676788415e-05,
"loss": 0.9948,
"step": 2680
},
{
"epoch": 2.35,
"learning_rate": 1.1261137936804813e-05,
"loss": 0.9788,
"step": 2685
},
{
"epoch": 2.36,
"learning_rate": 1.123325723867628e-05,
"loss": 0.9884,
"step": 2690
},
{
"epoch": 2.36,
"learning_rate": 1.1205366802553231e-05,
"loss": 0.9982,
"step": 2695
},
{
"epoch": 2.37,
"learning_rate": 1.1177466848662965e-05,
"loss": 0.9849,
"step": 2700
},
{
"epoch": 2.37,
"eval_loss": 1.1171647310256958,
"eval_runtime": 429.8432,
"eval_samples_per_second": 37.609,
"eval_steps_per_second": 1.177,
"step": 2700
},
{
"epoch": 2.37,
"learning_rate": 1.1149557597307936e-05,
"loss": 0.9876,
"step": 2705
},
{
"epoch": 2.37,
"learning_rate": 1.1121639268864014e-05,
"loss": 0.9984,
"step": 2710
},
{
"epoch": 2.38,
"learning_rate": 1.1093712083778748e-05,
"loss": 1.0014,
"step": 2715
},
{
"epoch": 2.38,
"learning_rate": 1.1065776262569605e-05,
"loss": 0.9906,
"step": 2720
},
{
"epoch": 2.39,
"learning_rate": 1.1037832025822266e-05,
"loss": 1.0017,
"step": 2725
},
{
"epoch": 2.39,
"learning_rate": 1.1009879594188844e-05,
"loss": 0.9945,
"step": 2730
},
{
"epoch": 2.4,
"learning_rate": 1.0981919188386171e-05,
"loss": 0.9916,
"step": 2735
},
{
"epoch": 2.4,
"learning_rate": 1.0953951029194039e-05,
"loss": 0.9786,
"step": 2740
},
{
"epoch": 2.4,
"learning_rate": 1.0925975337453462e-05,
"loss": 0.9991,
"step": 2745
},
{
"epoch": 2.41,
"learning_rate": 1.0897992334064937e-05,
"loss": 0.9888,
"step": 2750
},
{
"epoch": 2.41,
"learning_rate": 1.0870002239986689e-05,
"loss": 0.9892,
"step": 2755
},
{
"epoch": 2.42,
"learning_rate": 1.0842005276232932e-05,
"loss": 0.9937,
"step": 2760
},
{
"epoch": 2.42,
"learning_rate": 1.0814001663872133e-05,
"loss": 1.0033,
"step": 2765
},
{
"epoch": 2.43,
"learning_rate": 1.0785991624025249e-05,
"loss": 0.9837,
"step": 2770
},
{
"epoch": 2.43,
"learning_rate": 1.0757975377863989e-05,
"loss": 0.9955,
"step": 2775
},
{
"epoch": 2.44,
"learning_rate": 1.0729953146609076e-05,
"loss": 0.993,
"step": 2780
},
{
"epoch": 2.44,
"learning_rate": 1.0701925151528484e-05,
"loss": 0.9918,
"step": 2785
},
{
"epoch": 2.44,
"learning_rate": 1.06738916139357e-05,
"loss": 0.9777,
"step": 2790
},
{
"epoch": 2.45,
"learning_rate": 1.0645852755187984e-05,
"loss": 0.9869,
"step": 2795
},
{
"epoch": 2.45,
"learning_rate": 1.06178087966846e-05,
"loss": 0.9924,
"step": 2800
},
{
"epoch": 2.45,
"eval_loss": 1.1156463623046875,
"eval_runtime": 434.7239,
"eval_samples_per_second": 37.187,
"eval_steps_per_second": 1.164,
"step": 2800
},
{
"epoch": 2.46,
"learning_rate": 1.0589759959865087e-05,
"loss": 0.9908,
"step": 2805
},
{
"epoch": 2.46,
"learning_rate": 1.0561706466207506e-05,
"loss": 0.9973,
"step": 2810
},
{
"epoch": 2.47,
"learning_rate": 1.0533648537226688e-05,
"loss": 0.9985,
"step": 2815
},
{
"epoch": 2.47,
"learning_rate": 1.0505586394472486e-05,
"loss": 0.99,
"step": 2820
},
{
"epoch": 2.47,
"learning_rate": 1.0477520259528023e-05,
"loss": 0.9794,
"step": 2825
},
{
"epoch": 2.48,
"learning_rate": 1.0449450354007942e-05,
"loss": 0.9982,
"step": 2830
},
{
"epoch": 2.48,
"learning_rate": 1.0421376899556668e-05,
"loss": 1.008,
"step": 2835
},
{
"epoch": 2.49,
"learning_rate": 1.0393300117846648e-05,
"loss": 1.0023,
"step": 2840
},
{
"epoch": 2.49,
"learning_rate": 1.0365220230576592e-05,
"loss": 0.9875,
"step": 2845
},
{
"epoch": 2.5,
"learning_rate": 1.033713745946974e-05,
"loss": 0.9962,
"step": 2850
},
{
"epoch": 2.5,
"learning_rate": 1.0309052026272099e-05,
"loss": 1.0005,
"step": 2855
},
{
"epoch": 2.51,
"learning_rate": 1.0280964152750701e-05,
"loss": 0.9762,
"step": 2860
},
{
"epoch": 2.51,
"learning_rate": 1.0252874060691847e-05,
"loss": 1.011,
"step": 2865
},
{
"epoch": 2.51,
"learning_rate": 1.0224781971899353e-05,
"loss": 0.9907,
"step": 2870
},
{
"epoch": 2.52,
"learning_rate": 1.0196688108192798e-05,
"loss": 0.9759,
"step": 2875
},
{
"epoch": 2.52,
"learning_rate": 1.0168592691405782e-05,
"loss": 0.9907,
"step": 2880
},
{
"epoch": 2.53,
"learning_rate": 1.014049594338417e-05,
"loss": 1.0041,
"step": 2885
},
{
"epoch": 2.53,
"learning_rate": 1.0112398085984333e-05,
"loss": 1.0098,
"step": 2890
},
{
"epoch": 2.54,
"learning_rate": 1.0084299341071404e-05,
"loss": 0.9962,
"step": 2895
},
{
"epoch": 2.54,
"learning_rate": 1.0056199930517525e-05,
"loss": 0.9971,
"step": 2900
},
{
"epoch": 2.54,
"eval_loss": 1.1144633293151855,
"eval_runtime": 433.0074,
"eval_samples_per_second": 37.334,
"eval_steps_per_second": 1.169,
"step": 2900
},
{
"epoch": 2.54,
"learning_rate": 1.0028100076200088e-05,
"loss": 0.9757,
"step": 2905
},
{
"epoch": 2.55,
"learning_rate": 1e-05,
"loss": 0.9864,
"step": 2910
},
{
"epoch": 2.55,
"learning_rate": 9.971899923799915e-06,
"loss": 0.9928,
"step": 2915
},
{
"epoch": 2.56,
"learning_rate": 9.943800069482478e-06,
"loss": 0.9893,
"step": 2920
},
{
"epoch": 2.56,
"learning_rate": 9.915700658928599e-06,
"loss": 0.9939,
"step": 2925
},
{
"epoch": 2.57,
"learning_rate": 9.887601914015668e-06,
"loss": 0.9845,
"step": 2930
},
{
"epoch": 2.57,
"learning_rate": 9.859504056615832e-06,
"loss": 1.0015,
"step": 2935
},
{
"epoch": 2.58,
"learning_rate": 9.831407308594216e-06,
"loss": 0.9892,
"step": 2940
},
{
"epoch": 2.58,
"learning_rate": 9.803311891807205e-06,
"loss": 0.9891,
"step": 2945
},
{
"epoch": 2.58,
"learning_rate": 9.77521802810065e-06,
"loss": 0.9769,
"step": 2950
},
{
"epoch": 2.59,
"learning_rate": 9.747125939308154e-06,
"loss": 0.9904,
"step": 2955
},
{
"epoch": 2.59,
"learning_rate": 9.719035847249304e-06,
"loss": 0.9794,
"step": 2960
},
{
"epoch": 2.6,
"learning_rate": 9.690947973727905e-06,
"loss": 0.9893,
"step": 2965
},
{
"epoch": 2.6,
"learning_rate": 9.662862540530267e-06,
"loss": 1.0053,
"step": 2970
},
{
"epoch": 2.61,
"learning_rate": 9.634779769423412e-06,
"loss": 0.9977,
"step": 2975
},
{
"epoch": 2.61,
"learning_rate": 9.606699882153357e-06,
"loss": 0.9895,
"step": 2980
},
{
"epoch": 2.61,
"learning_rate": 9.578623100443333e-06,
"loss": 1.0,
"step": 2985
},
{
"epoch": 2.62,
"learning_rate": 9.550549645992061e-06,
"loss": 0.9878,
"step": 2990
},
{
"epoch": 2.62,
"learning_rate": 9.522479740471979e-06,
"loss": 0.9942,
"step": 2995
},
{
"epoch": 2.63,
"learning_rate": 9.494413605527515e-06,
"loss": 0.9786,
"step": 3000
},
{
"epoch": 2.63,
"eval_loss": 1.1129907369613647,
"eval_runtime": 436.9997,
"eval_samples_per_second": 36.993,
"eval_steps_per_second": 1.158,
"step": 3000
},
{
"epoch": 2.63,
"learning_rate": 9.466351462773314e-06,
"loss": 0.9917,
"step": 3005
},
{
"epoch": 2.64,
"learning_rate": 9.438293533792496e-06,
"loss": 0.9784,
"step": 3010
},
{
"epoch": 2.64,
"learning_rate": 9.410240040134918e-06,
"loss": 0.9877,
"step": 3015
},
{
"epoch": 2.65,
"learning_rate": 9.382191203315404e-06,
"loss": 0.983,
"step": 3020
},
{
"epoch": 2.65,
"learning_rate": 9.354147244812021e-06,
"loss": 1.0003,
"step": 3025
},
{
"epoch": 2.65,
"learning_rate": 9.326108386064303e-06,
"loss": 0.9796,
"step": 3030
},
{
"epoch": 2.66,
"learning_rate": 9.298074848471521e-06,
"loss": 0.9753,
"step": 3035
},
{
"epoch": 2.66,
"learning_rate": 9.270046853390924e-06,
"loss": 0.992,
"step": 3040
},
{
"epoch": 2.67,
"learning_rate": 9.242024622136013e-06,
"loss": 0.9919,
"step": 3045
},
{
"epoch": 2.67,
"learning_rate": 9.214008375974751e-06,
"loss": 0.9876,
"step": 3050
},
{
"epoch": 2.68,
"learning_rate": 9.185998336127869e-06,
"loss": 0.9985,
"step": 3055
},
{
"epoch": 2.68,
"learning_rate": 9.157994723767071e-06,
"loss": 0.9925,
"step": 3060
},
{
"epoch": 2.69,
"learning_rate": 9.129997760013316e-06,
"loss": 1.0047,
"step": 3065
},
{
"epoch": 2.69,
"learning_rate": 9.102007665935068e-06,
"loss": 0.9851,
"step": 3070
},
{
"epoch": 2.69,
"learning_rate": 9.07402466254654e-06,
"loss": 0.9819,
"step": 3075
},
{
"epoch": 2.7,
"learning_rate": 9.046048970805964e-06,
"loss": 0.9903,
"step": 3080
},
{
"epoch": 2.7,
"learning_rate": 9.018080811613832e-06,
"loss": 1.0065,
"step": 3085
},
{
"epoch": 2.71,
"learning_rate": 8.99012040581116e-06,
"loss": 0.9884,
"step": 3090
},
{
"epoch": 2.71,
"learning_rate": 8.962167974177736e-06,
"loss": 0.9914,
"step": 3095
},
{
"epoch": 2.72,
"learning_rate": 8.934223737430397e-06,
"loss": 0.9923,
"step": 3100
},
{
"epoch": 2.72,
"eval_loss": 1.1121999025344849,
"eval_runtime": 436.2817,
"eval_samples_per_second": 37.054,
"eval_steps_per_second": 1.16,
"step": 3100
},
{
"epoch": 2.72,
"learning_rate": 8.906287916221259e-06,
"loss": 0.9978,
"step": 3105
},
{
"epoch": 2.72,
"learning_rate": 8.878360731135987e-06,
"loss": 0.9882,
"step": 3110
},
{
"epoch": 2.73,
"learning_rate": 8.850442402692068e-06,
"loss": 0.9883,
"step": 3115
},
{
"epoch": 2.73,
"learning_rate": 8.822533151337038e-06,
"loss": 0.9794,
"step": 3120
},
{
"epoch": 2.74,
"learning_rate": 8.79463319744677e-06,
"loss": 0.9828,
"step": 3125
},
{
"epoch": 2.74,
"learning_rate": 8.766742761323718e-06,
"loss": 0.9974,
"step": 3130
},
{
"epoch": 2.75,
"learning_rate": 8.73886206319519e-06,
"loss": 1.0021,
"step": 3135
},
{
"epoch": 2.75,
"learning_rate": 8.710991323211586e-06,
"loss": 0.9778,
"step": 3140
},
{
"epoch": 2.76,
"learning_rate": 8.683130761444691e-06,
"loss": 0.979,
"step": 3145
},
{
"epoch": 2.76,
"learning_rate": 8.65528059788591e-06,
"loss": 0.9965,
"step": 3150
},
{
"epoch": 2.76,
"learning_rate": 8.627441052444545e-06,
"loss": 0.9993,
"step": 3155
},
{
"epoch": 2.77,
"learning_rate": 8.599612344946063e-06,
"loss": 0.977,
"step": 3160
},
{
"epoch": 2.77,
"learning_rate": 8.571794695130343e-06,
"loss": 0.9768,
"step": 3165
},
{
"epoch": 2.78,
"learning_rate": 8.543988322649954e-06,
"loss": 0.9896,
"step": 3170
},
{
"epoch": 2.78,
"learning_rate": 8.516193447068415e-06,
"loss": 0.9913,
"step": 3175
},
{
"epoch": 2.79,
"learning_rate": 8.488410287858478e-06,
"loss": 0.9929,
"step": 3180
},
{
"epoch": 2.79,
"learning_rate": 8.460639064400354e-06,
"loss": 0.9836,
"step": 3185
},
{
"epoch": 2.79,
"learning_rate": 8.432879995980036e-06,
"loss": 0.9988,
"step": 3190
},
{
"epoch": 2.8,
"learning_rate": 8.405133301787521e-06,
"loss": 0.9934,
"step": 3195
},
{
"epoch": 2.8,
"learning_rate": 8.377399200915098e-06,
"loss": 0.9888,
"step": 3200
},
{
"epoch": 2.8,
"eval_loss": 1.1105643510818481,
"eval_runtime": 437.3273,
"eval_samples_per_second": 36.965,
"eval_steps_per_second": 1.157,
"step": 3200
},
{
"epoch": 2.81,
"learning_rate": 8.349677912355624e-06,
"loss": 0.977,
"step": 3205
},
{
"epoch": 2.81,
"learning_rate": 8.321969655000782e-06,
"loss": 0.987,
"step": 3210
},
{
"epoch": 2.82,
"learning_rate": 8.294274647639369e-06,
"loss": 0.9819,
"step": 3215
},
{
"epoch": 2.82,
"learning_rate": 8.266593108955541e-06,
"loss": 0.9884,
"step": 3220
},
{
"epoch": 2.83,
"learning_rate": 8.238925257527122e-06,
"loss": 0.9865,
"step": 3225
},
{
"epoch": 2.83,
"learning_rate": 8.211271311823839e-06,
"loss": 0.996,
"step": 3230
},
{
"epoch": 2.83,
"learning_rate": 8.183631490205636e-06,
"loss": 0.9826,
"step": 3235
},
{
"epoch": 2.84,
"learning_rate": 8.156006010920919e-06,
"loss": 1.0028,
"step": 3240
},
{
"epoch": 2.84,
"learning_rate": 8.128395092104855e-06,
"loss": 0.9725,
"step": 3245
},
{
"epoch": 2.85,
"learning_rate": 8.100798951777626e-06,
"loss": 1.0054,
"step": 3250
},
{
"epoch": 2.85,
"learning_rate": 8.073217807842727e-06,
"loss": 0.9879,
"step": 3255
},
{
"epoch": 2.86,
"learning_rate": 8.045651878085245e-06,
"loss": 0.998,
"step": 3260
},
{
"epoch": 2.86,
"learning_rate": 8.018101380170123e-06,
"loss": 0.9806,
"step": 3265
},
{
"epoch": 2.86,
"learning_rate": 7.990566531640463e-06,
"loss": 0.9806,
"step": 3270
},
{
"epoch": 2.87,
"learning_rate": 7.963047549915785e-06,
"loss": 0.9914,
"step": 3275
},
{
"epoch": 2.87,
"learning_rate": 7.935544652290331e-06,
"loss": 0.9929,
"step": 3280
},
{
"epoch": 2.88,
"learning_rate": 7.908058055931334e-06,
"loss": 0.9777,
"step": 3285
},
{
"epoch": 2.88,
"learning_rate": 7.88058797787732e-06,
"loss": 0.9831,
"step": 3290
},
{
"epoch": 2.89,
"learning_rate": 7.853134635036371e-06,
"loss": 0.9942,
"step": 3295
},
{
"epoch": 2.89,
"learning_rate": 7.825698244184432e-06,
"loss": 0.9826,
"step": 3300
},
{
"epoch": 2.89,
"eval_loss": 1.109144926071167,
"eval_runtime": 436.0343,
"eval_samples_per_second": 37.075,
"eval_steps_per_second": 1.16,
"step": 3300
},
{
"epoch": 2.9,
"learning_rate": 7.798279021963594e-06,
"loss": 0.9707,
"step": 3305
},
{
"epoch": 2.9,
"learning_rate": 7.770877184880373e-06,
"loss": 0.9672,
"step": 3310
},
{
"epoch": 2.9,
"learning_rate": 7.743492949304023e-06,
"loss": 0.979,
"step": 3315
},
{
"epoch": 2.91,
"learning_rate": 7.716126531464799e-06,
"loss": 0.9893,
"step": 3320
},
{
"epoch": 2.91,
"learning_rate": 7.688778147452281e-06,
"loss": 0.9905,
"step": 3325
},
{
"epoch": 2.92,
"learning_rate": 7.66144801321363e-06,
"loss": 0.9895,
"step": 3330
},
{
"epoch": 2.92,
"learning_rate": 7.634136344551926e-06,
"loss": 0.9935,
"step": 3335
},
{
"epoch": 2.93,
"learning_rate": 7.606843357124426e-06,
"loss": 0.9884,
"step": 3340
},
{
"epoch": 2.93,
"learning_rate": 7.579569266440877e-06,
"loss": 0.9894,
"step": 3345
},
{
"epoch": 2.93,
"learning_rate": 7.55231428786183e-06,
"loss": 0.9745,
"step": 3350
},
{
"epoch": 2.94,
"learning_rate": 7.525078636596906e-06,
"loss": 0.9903,
"step": 3355
},
{
"epoch": 2.94,
"learning_rate": 7.497862527703127e-06,
"loss": 1.0,
"step": 3360
},
{
"epoch": 2.95,
"learning_rate": 7.470666176083193e-06,
"loss": 0.9934,
"step": 3365
},
{
"epoch": 2.95,
"learning_rate": 7.443489796483812e-06,
"loss": 0.9847,
"step": 3370
},
{
"epoch": 2.96,
"learning_rate": 7.416333603493977e-06,
"loss": 0.9854,
"step": 3375
},
{
"epoch": 2.96,
"learning_rate": 7.389197811543296e-06,
"loss": 0.9914,
"step": 3380
},
{
"epoch": 2.97,
"learning_rate": 7.362082634900276e-06,
"loss": 0.978,
"step": 3385
},
{
"epoch": 2.97,
"learning_rate": 7.334988287670648e-06,
"loss": 0.9901,
"step": 3390
},
{
"epoch": 2.97,
"learning_rate": 7.3079149837956695e-06,
"loss": 0.9698,
"step": 3395
},
{
"epoch": 2.98,
"learning_rate": 7.280862937050435e-06,
"loss": 0.9997,
"step": 3400
},
{
"epoch": 2.98,
"eval_loss": 1.1089974641799927,
"eval_runtime": 437.9541,
"eval_samples_per_second": 36.913,
"eval_steps_per_second": 1.155,
"step": 3400
},
{
"epoch": 2.98,
"learning_rate": 7.2538323610421965e-06,
"loss": 0.9906,
"step": 3405
},
{
"epoch": 2.99,
"learning_rate": 7.226823469208659e-06,
"loss": 0.9808,
"step": 3410
},
{
"epoch": 2.99,
"learning_rate": 7.199836474816313e-06,
"loss": 0.9828,
"step": 3415
},
{
"epoch": 3.0,
"learning_rate": 7.172871590958739e-06,
"loss": 0.9907,
"step": 3420
},
{
"epoch": 3.0,
"learning_rate": 7.145929030554936e-06,
"loss": 0.9672,
"step": 3425
},
{
"epoch": 3.0,
"learning_rate": 7.119009006347625e-06,
"loss": 0.9034,
"step": 3430
},
{
"epoch": 3.01,
"learning_rate": 7.0921117309015855e-06,
"loss": 0.9385,
"step": 3435
},
{
"epoch": 3.01,
"learning_rate": 7.0652374166019625e-06,
"loss": 0.9349,
"step": 3440
},
{
"epoch": 3.02,
"learning_rate": 7.038386275652594e-06,
"loss": 0.9394,
"step": 3445
},
{
"epoch": 3.02,
"learning_rate": 7.011558520074352e-06,
"loss": 0.9382,
"step": 3450
},
{
"epoch": 3.03,
"learning_rate": 6.984754361703437e-06,
"loss": 0.9311,
"step": 3455
},
{
"epoch": 3.03,
"learning_rate": 6.957974012189734e-06,
"loss": 0.9272,
"step": 3460
},
{
"epoch": 3.04,
"learning_rate": 6.931217682995124e-06,
"loss": 0.9243,
"step": 3465
},
{
"epoch": 3.04,
"learning_rate": 6.9044855853918224e-06,
"loss": 0.9306,
"step": 3470
},
{
"epoch": 3.04,
"learning_rate": 6.877777930460701e-06,
"loss": 0.9366,
"step": 3475
},
{
"epoch": 3.05,
"learning_rate": 6.85109492908964e-06,
"loss": 0.9339,
"step": 3480
},
{
"epoch": 3.05,
"learning_rate": 6.8244367919718445e-06,
"loss": 0.9285,
"step": 3485
},
{
"epoch": 3.06,
"learning_rate": 6.797803729604183e-06,
"loss": 0.9249,
"step": 3490
},
{
"epoch": 3.06,
"learning_rate": 6.771195952285541e-06,
"loss": 0.9211,
"step": 3495
},
{
"epoch": 3.07,
"learning_rate": 6.744613670115137e-06,
"loss": 0.9267,
"step": 3500
},
{
"epoch": 3.07,
"eval_loss": 1.121911883354187,
"eval_runtime": 435.9813,
"eval_samples_per_second": 37.08,
"eval_steps_per_second": 1.161,
"step": 3500
},
{
"epoch": 3.07,
"learning_rate": 6.718057092990889e-06,
"loss": 0.9107,
"step": 3505
},
{
"epoch": 3.07,
"learning_rate": 6.691526430607729e-06,
"loss": 0.9276,
"step": 3510
},
{
"epoch": 3.08,
"learning_rate": 6.665021892455978e-06,
"loss": 0.9386,
"step": 3515
},
{
"epoch": 3.08,
"learning_rate": 6.638543687819663e-06,
"loss": 0.9384,
"step": 3520
},
{
"epoch": 3.09,
"learning_rate": 6.612092025774885e-06,
"loss": 0.935,
"step": 3525
},
{
"epoch": 3.09,
"learning_rate": 6.585667115188157e-06,
"loss": 0.9239,
"step": 3530
},
{
"epoch": 3.1,
"learning_rate": 6.559269164714753e-06,
"loss": 0.9363,
"step": 3535
},
{
"epoch": 3.1,
"learning_rate": 6.532898382797082e-06,
"loss": 0.9234,
"step": 3540
},
{
"epoch": 3.11,
"learning_rate": 6.5065549776630076e-06,
"loss": 0.9343,
"step": 3545
},
{
"epoch": 3.11,
"learning_rate": 6.480239157324235e-06,
"loss": 0.9264,
"step": 3550
},
{
"epoch": 3.11,
"learning_rate": 6.453951129574644e-06,
"loss": 0.9271,
"step": 3555
},
{
"epoch": 3.12,
"learning_rate": 6.427691101988673e-06,
"loss": 0.9241,
"step": 3560
},
{
"epoch": 3.12,
"learning_rate": 6.401459281919656e-06,
"loss": 0.9253,
"step": 3565
},
{
"epoch": 3.13,
"learning_rate": 6.375255876498202e-06,
"loss": 0.9377,
"step": 3570
},
{
"epoch": 3.13,
"learning_rate": 6.3490810926305495e-06,
"loss": 0.923,
"step": 3575
},
{
"epoch": 3.14,
"learning_rate": 6.322935136996935e-06,
"loss": 0.9169,
"step": 3580
},
{
"epoch": 3.14,
"learning_rate": 6.296818216049973e-06,
"loss": 0.9246,
"step": 3585
},
{
"epoch": 3.14,
"learning_rate": 6.270730536013e-06,
"loss": 0.9254,
"step": 3590
},
{
"epoch": 3.15,
"learning_rate": 6.244672302878479e-06,
"loss": 0.9399,
"step": 3595
},
{
"epoch": 3.15,
"learning_rate": 6.218643722406342e-06,
"loss": 0.9465,
"step": 3600
},
{
"epoch": 3.15,
"eval_loss": 1.1225427389144897,
"eval_runtime": 436.2413,
"eval_samples_per_second": 37.057,
"eval_steps_per_second": 1.16,
"step": 3600
},
{
"epoch": 3.16,
"learning_rate": 6.192645000122386e-06,
"loss": 0.9498,
"step": 3605
},
{
"epoch": 3.16,
"learning_rate": 6.1666763413166355e-06,
"loss": 0.9249,
"step": 3610
},
{
"epoch": 3.17,
"learning_rate": 6.140737951041742e-06,
"loss": 0.9548,
"step": 3615
},
{
"epoch": 3.17,
"learning_rate": 6.114830034111336e-06,
"loss": 0.9007,
"step": 3620
},
{
"epoch": 3.18,
"learning_rate": 6.088952795098442e-06,
"loss": 0.9335,
"step": 3625
},
{
"epoch": 3.18,
"learning_rate": 6.063106438333832e-06,
"loss": 0.9424,
"step": 3630
},
{
"epoch": 3.18,
"learning_rate": 6.0372911679044334e-06,
"loss": 0.9382,
"step": 3635
},
{
"epoch": 3.19,
"learning_rate": 6.011507187651715e-06,
"loss": 0.9337,
"step": 3640
},
{
"epoch": 3.19,
"learning_rate": 5.985754701170066e-06,
"loss": 0.9377,
"step": 3645
},
{
"epoch": 3.2,
"learning_rate": 5.960033911805202e-06,
"loss": 0.9472,
"step": 3650
},
{
"epoch": 3.2,
"learning_rate": 5.9343450226525475e-06,
"loss": 0.9342,
"step": 3655
},
{
"epoch": 3.21,
"learning_rate": 5.908688236555644e-06,
"loss": 0.9387,
"step": 3660
},
{
"epoch": 3.21,
"learning_rate": 5.883063756104535e-06,
"loss": 0.936,
"step": 3665
},
{
"epoch": 3.22,
"learning_rate": 5.857471783634181e-06,
"loss": 0.9264,
"step": 3670
},
{
"epoch": 3.22,
"learning_rate": 5.831912521222849e-06,
"loss": 0.9346,
"step": 3675
},
{
"epoch": 3.22,
"learning_rate": 5.806386170690519e-06,
"loss": 0.9337,
"step": 3680
},
{
"epoch": 3.23,
"learning_rate": 5.780892933597309e-06,
"loss": 0.9182,
"step": 3685
},
{
"epoch": 3.23,
"learning_rate": 5.755433011241851e-06,
"loss": 0.9251,
"step": 3690
},
{
"epoch": 3.24,
"learning_rate": 5.730006604659726e-06,
"loss": 0.9358,
"step": 3695
},
{
"epoch": 3.24,
"learning_rate": 5.704613914621865e-06,
"loss": 0.9255,
"step": 3700
},
{
"epoch": 3.24,
"eval_loss": 1.1221187114715576,
"eval_runtime": 435.9069,
"eval_samples_per_second": 37.086,
"eval_steps_per_second": 1.161,
"step": 3700
},
{
"epoch": 3.25,
"learning_rate": 5.679255141632977e-06,
"loss": 0.9204,
"step": 3705
},
{
"epoch": 3.25,
"learning_rate": 5.653930485929944e-06,
"loss": 0.9399,
"step": 3710
},
{
"epoch": 3.25,
"learning_rate": 5.6286401474802675e-06,
"loss": 0.9266,
"step": 3715
},
{
"epoch": 3.26,
"learning_rate": 5.603384325980461e-06,
"loss": 0.9183,
"step": 3720
},
{
"epoch": 3.26,
"learning_rate": 5.578163220854489e-06,
"loss": 0.9397,
"step": 3725
},
{
"epoch": 3.27,
"learning_rate": 5.5529770312522e-06,
"loss": 0.9407,
"step": 3730
},
{
"epoch": 3.27,
"learning_rate": 5.52782595604773e-06,
"loss": 0.9378,
"step": 3735
},
{
"epoch": 3.28,
"learning_rate": 5.502710193837959e-06,
"loss": 0.9291,
"step": 3740
},
{
"epoch": 3.28,
"learning_rate": 5.477629942940923e-06,
"loss": 0.9293,
"step": 3745
},
{
"epoch": 3.29,
"learning_rate": 5.452585401394254e-06,
"loss": 0.9311,
"step": 3750
},
{
"epoch": 3.29,
"learning_rate": 5.427576766953615e-06,
"loss": 0.9228,
"step": 3755
},
{
"epoch": 3.29,
"learning_rate": 5.402604237091156e-06,
"loss": 0.9243,
"step": 3760
},
{
"epoch": 3.3,
"learning_rate": 5.377668008993921e-06,
"loss": 0.9388,
"step": 3765
},
{
"epoch": 3.3,
"learning_rate": 5.352768279562315e-06,
"loss": 0.9354,
"step": 3770
},
{
"epoch": 3.31,
"learning_rate": 5.327905245408556e-06,
"loss": 0.9219,
"step": 3775
},
{
"epoch": 3.31,
"learning_rate": 5.303079102855089e-06,
"loss": 0.9392,
"step": 3780
},
{
"epoch": 3.32,
"learning_rate": 5.278290047933077e-06,
"loss": 0.9303,
"step": 3785
},
{
"epoch": 3.32,
"learning_rate": 5.2535382763808185e-06,
"loss": 0.9294,
"step": 3790
},
{
"epoch": 3.32,
"learning_rate": 5.228823983642228e-06,
"loss": 0.9239,
"step": 3795
},
{
"epoch": 3.33,
"learning_rate": 5.204147364865275e-06,
"loss": 0.9532,
"step": 3800
},
{
"epoch": 3.33,
"eval_loss": 1.12142813205719,
"eval_runtime": 437.0064,
"eval_samples_per_second": 36.993,
"eval_steps_per_second": 1.158,
"step": 3800
},
{
"epoch": 3.33,
"learning_rate": 5.179508614900448e-06,
"loss": 0.9383,
"step": 3805
},
{
"epoch": 3.34,
"learning_rate": 5.154907928299217e-06,
"loss": 0.9507,
"step": 3810
},
{
"epoch": 3.34,
"learning_rate": 5.1303454993125075e-06,
"loss": 0.9423,
"step": 3815
},
{
"epoch": 3.35,
"learning_rate": 5.105821521889147e-06,
"loss": 0.9433,
"step": 3820
},
{
"epoch": 3.35,
"learning_rate": 5.081336189674342e-06,
"loss": 0.9223,
"step": 3825
},
{
"epoch": 3.36,
"learning_rate": 5.056889696008162e-06,
"loss": 0.9367,
"step": 3830
},
{
"epoch": 3.36,
"learning_rate": 5.032482233923984e-06,
"loss": 0.9276,
"step": 3835
},
{
"epoch": 3.36,
"learning_rate": 5.008113996147006e-06,
"loss": 0.9436,
"step": 3840
},
{
"epoch": 3.37,
"learning_rate": 4.983785175092684e-06,
"loss": 0.9505,
"step": 3845
},
{
"epoch": 3.37,
"learning_rate": 4.959495962865249e-06,
"loss": 0.9427,
"step": 3850
},
{
"epoch": 3.38,
"learning_rate": 4.93524655125617e-06,
"loss": 0.9331,
"step": 3855
},
{
"epoch": 3.38,
"learning_rate": 4.911037131742637e-06,
"loss": 0.9269,
"step": 3860
},
{
"epoch": 3.39,
"learning_rate": 4.886867895486065e-06,
"loss": 0.9226,
"step": 3865
},
{
"epoch": 3.39,
"learning_rate": 4.862739033330566e-06,
"loss": 0.9334,
"step": 3870
},
{
"epoch": 3.39,
"learning_rate": 4.8386507358014664e-06,
"loss": 0.9188,
"step": 3875
},
{
"epoch": 3.4,
"learning_rate": 4.814603193103773e-06,
"loss": 0.9248,
"step": 3880
},
{
"epoch": 3.4,
"learning_rate": 4.790596595120699e-06,
"loss": 0.9235,
"step": 3885
},
{
"epoch": 3.41,
"learning_rate": 4.766631131412136e-06,
"loss": 0.9387,
"step": 3890
},
{
"epoch": 3.41,
"learning_rate": 4.742706991213191e-06,
"loss": 0.9332,
"step": 3895
},
{
"epoch": 3.42,
"learning_rate": 4.718824363432663e-06,
"loss": 0.9372,
"step": 3900
},
{
"epoch": 3.42,
"eval_loss": 1.1215200424194336,
"eval_runtime": 437.1913,
"eval_samples_per_second": 36.977,
"eval_steps_per_second": 1.157,
"step": 3900
},
{
"epoch": 3.42,
"learning_rate": 4.694983436651559e-06,
"loss": 0.9355,
"step": 3905
},
{
"epoch": 3.43,
"learning_rate": 4.671184399121619e-06,
"loss": 0.9223,
"step": 3910
},
{
"epoch": 3.43,
"learning_rate": 4.647427438763806e-06,
"loss": 0.9315,
"step": 3915
},
{
"epoch": 3.43,
"learning_rate": 4.623712743166851e-06,
"loss": 0.9329,
"step": 3920
},
{
"epoch": 3.44,
"learning_rate": 4.60004049958574e-06,
"loss": 0.9284,
"step": 3925
},
{
"epoch": 3.44,
"learning_rate": 4.576410894940265e-06,
"loss": 0.9469,
"step": 3930
},
{
"epoch": 3.45,
"learning_rate": 4.552824115813522e-06,
"loss": 0.9308,
"step": 3935
},
{
"epoch": 3.45,
"learning_rate": 4.52928034845046e-06,
"loss": 0.9299,
"step": 3940
},
{
"epoch": 3.46,
"learning_rate": 4.5057797787563905e-06,
"loss": 0.932,
"step": 3945
},
{
"epoch": 3.46,
"learning_rate": 4.482322592295541e-06,
"loss": 0.9312,
"step": 3950
},
{
"epoch": 3.46,
"learning_rate": 4.458908974289568e-06,
"loss": 0.9338,
"step": 3955
},
{
"epoch": 3.47,
"learning_rate": 4.4355391096161085e-06,
"loss": 0.9232,
"step": 3960
},
{
"epoch": 3.47,
"learning_rate": 4.412213182807315e-06,
"loss": 0.9365,
"step": 3965
},
{
"epoch": 3.48,
"learning_rate": 4.3889313780484e-06,
"loss": 0.9303,
"step": 3970
},
{
"epoch": 3.48,
"learning_rate": 4.365693879176186e-06,
"loss": 0.9432,
"step": 3975
},
{
"epoch": 3.49,
"learning_rate": 4.342500869677638e-06,
"loss": 0.9348,
"step": 3980
},
{
"epoch": 3.49,
"learning_rate": 4.319352532688444e-06,
"loss": 0.9435,
"step": 3985
},
{
"epoch": 3.5,
"learning_rate": 4.296249050991529e-06,
"loss": 0.9437,
"step": 3990
},
{
"epoch": 3.5,
"learning_rate": 4.273190607015652e-06,
"loss": 0.9241,
"step": 3995
},
{
"epoch": 3.5,
"learning_rate": 4.250177382833937e-06,
"loss": 0.9206,
"step": 4000
},
{
"epoch": 3.5,
"eval_loss": 1.1212979555130005,
"eval_runtime": 437.0246,
"eval_samples_per_second": 36.991,
"eval_steps_per_second": 1.158,
"step": 4000
},
{
"epoch": 3.51,
"learning_rate": 4.227209560162443e-06,
"loss": 0.941,
"step": 4005
},
{
"epoch": 3.51,
"learning_rate": 4.204287320358744e-06,
"loss": 0.9412,
"step": 4010
},
{
"epoch": 3.52,
"learning_rate": 4.181410844420473e-06,
"loss": 0.9333,
"step": 4015
},
{
"epoch": 3.52,
"learning_rate": 4.158580312983908e-06,
"loss": 0.9199,
"step": 4020
},
{
"epoch": 3.53,
"learning_rate": 4.135795906322539e-06,
"loss": 0.9588,
"step": 4025
},
{
"epoch": 3.53,
"learning_rate": 4.113057804345656e-06,
"loss": 0.9237,
"step": 4030
},
{
"epoch": 3.53,
"learning_rate": 4.090366186596907e-06,
"loss": 0.9307,
"step": 4035
},
{
"epoch": 3.54,
"learning_rate": 4.067721232252909e-06,
"loss": 0.9377,
"step": 4040
},
{
"epoch": 3.54,
"learning_rate": 4.0451231201217996e-06,
"loss": 0.9283,
"step": 4045
},
{
"epoch": 3.55,
"learning_rate": 4.022572028641861e-06,
"loss": 0.9409,
"step": 4050
},
{
"epoch": 3.55,
"learning_rate": 4.00006813588008e-06,
"loss": 0.9178,
"step": 4055
},
{
"epoch": 3.56,
"learning_rate": 3.977611619530755e-06,
"loss": 0.9433,
"step": 4060
},
{
"epoch": 3.56,
"learning_rate": 3.955202656914102e-06,
"loss": 0.9512,
"step": 4065
},
{
"epoch": 3.57,
"learning_rate": 3.932841424974838e-06,
"loss": 0.9408,
"step": 4070
},
{
"epoch": 3.57,
"learning_rate": 3.910528100280793e-06,
"loss": 0.9341,
"step": 4075
},
{
"epoch": 3.57,
"learning_rate": 3.888262859021508e-06,
"loss": 0.9376,
"step": 4080
},
{
"epoch": 3.58,
"learning_rate": 3.866045877006863e-06,
"loss": 0.9355,
"step": 4085
},
{
"epoch": 3.58,
"learning_rate": 3.843877329665662e-06,
"loss": 0.9307,
"step": 4090
},
{
"epoch": 3.59,
"learning_rate": 3.821757392044271e-06,
"loss": 0.9243,
"step": 4095
},
{
"epoch": 3.59,
"learning_rate": 3.7996862388052192e-06,
"loss": 0.9394,
"step": 4100
},
{
"epoch": 3.59,
"eval_loss": 1.1206693649291992,
"eval_runtime": 436.5414,
"eval_samples_per_second": 37.032,
"eval_steps_per_second": 1.159,
"step": 4100
},
{
"epoch": 3.6,
"learning_rate": 3.777664044225825e-06,
"loss": 0.9185,
"step": 4105
},
{
"epoch": 3.6,
"learning_rate": 3.7556909821968335e-06,
"loss": 0.9235,
"step": 4110
},
{
"epoch": 3.6,
"learning_rate": 3.7337672262210155e-06,
"loss": 0.9342,
"step": 4115
},
{
"epoch": 3.61,
"learning_rate": 3.7118929494118284e-06,
"loss": 0.9214,
"step": 4120
},
{
"epoch": 3.61,
"learning_rate": 3.690068324492023e-06,
"loss": 0.9318,
"step": 4125
},
{
"epoch": 3.62,
"learning_rate": 3.6682935237922956e-06,
"loss": 0.9383,
"step": 4130
},
{
"epoch": 3.62,
"learning_rate": 3.6465687192499176e-06,
"loss": 0.9211,
"step": 4135
},
{
"epoch": 3.63,
"learning_rate": 3.6248940824073955e-06,
"loss": 0.9305,
"step": 4140
},
{
"epoch": 3.63,
"learning_rate": 3.6032697844110896e-06,
"loss": 0.9306,
"step": 4145
},
{
"epoch": 3.64,
"learning_rate": 3.5816959960098817e-06,
"loss": 0.9406,
"step": 4150
},
{
"epoch": 3.64,
"learning_rate": 3.560172887553827e-06,
"loss": 0.9171,
"step": 4155
},
{
"epoch": 3.64,
"learning_rate": 3.538700628992794e-06,
"loss": 0.9166,
"step": 4160
},
{
"epoch": 3.65,
"learning_rate": 3.5172793898751446e-06,
"loss": 0.9133,
"step": 4165
},
{
"epoch": 3.65,
"learning_rate": 3.4959093393463716e-06,
"loss": 0.9347,
"step": 4170
},
{
"epoch": 3.66,
"learning_rate": 3.474590646147781e-06,
"loss": 0.9311,
"step": 4175
},
{
"epoch": 3.66,
"learning_rate": 3.453323478615146e-06,
"loss": 0.9229,
"step": 4180
},
{
"epoch": 3.67,
"learning_rate": 3.4321080046773993e-06,
"loss": 0.9229,
"step": 4185
},
{
"epoch": 3.67,
"learning_rate": 3.410944391855279e-06,
"loss": 0.9242,
"step": 4190
},
{
"epoch": 3.67,
"learning_rate": 3.3898328072600216e-06,
"loss": 0.9246,
"step": 4195
},
{
"epoch": 3.68,
"learning_rate": 3.3687734175920505e-06,
"loss": 0.9367,
"step": 4200
},
{
"epoch": 3.68,
"eval_loss": 1.1195218563079834,
"eval_runtime": 435.7564,
"eval_samples_per_second": 37.099,
"eval_steps_per_second": 1.161,
"step": 4200
},
{
"epoch": 3.68,
"learning_rate": 3.347766389139635e-06,
"loss": 0.9297,
"step": 4205
},
{
"epoch": 3.69,
"learning_rate": 3.326811887777607e-06,
"loss": 0.9328,
"step": 4210
},
{
"epoch": 3.69,
"learning_rate": 3.3059100789660215e-06,
"loss": 0.9167,
"step": 4215
},
{
"epoch": 3.7,
"learning_rate": 3.28506112774888e-06,
"loss": 0.9326,
"step": 4220
},
{
"epoch": 3.7,
"learning_rate": 3.2642651987527984e-06,
"loss": 0.9318,
"step": 4225
},
{
"epoch": 3.71,
"learning_rate": 3.243522456185728e-06,
"loss": 0.9245,
"step": 4230
},
{
"epoch": 3.71,
"learning_rate": 3.2228330638356463e-06,
"loss": 0.9287,
"step": 4235
},
{
"epoch": 3.71,
"learning_rate": 3.202197185069279e-06,
"loss": 0.9302,
"step": 4240
},
{
"epoch": 3.72,
"learning_rate": 3.181614982830792e-06,
"loss": 0.9123,
"step": 4245
},
{
"epoch": 3.72,
"learning_rate": 3.161086619640509e-06,
"loss": 0.9379,
"step": 4250
},
{
"epoch": 3.73,
"learning_rate": 3.1406122575936447e-06,
"loss": 0.9426,
"step": 4255
},
{
"epoch": 3.73,
"learning_rate": 3.1201920583590005e-06,
"loss": 0.9218,
"step": 4260
},
{
"epoch": 3.74,
"learning_rate": 3.09982618317771e-06,
"loss": 0.9317,
"step": 4265
},
{
"epoch": 3.74,
"learning_rate": 3.079514792861943e-06,
"loss": 0.9196,
"step": 4270
},
{
"epoch": 3.75,
"learning_rate": 3.0592580477936606e-06,
"loss": 0.9331,
"step": 4275
},
{
"epoch": 3.75,
"learning_rate": 3.0390561079233294e-06,
"loss": 0.9217,
"step": 4280
},
{
"epoch": 3.75,
"learning_rate": 3.018909132768666e-06,
"loss": 0.9298,
"step": 4285
},
{
"epoch": 3.76,
"learning_rate": 2.9988172814133774e-06,
"loss": 0.9336,
"step": 4290
},
{
"epoch": 3.76,
"learning_rate": 2.978780712505902e-06,
"loss": 0.9317,
"step": 4295
},
{
"epoch": 3.77,
"learning_rate": 2.9587995842581673e-06,
"loss": 0.9245,
"step": 4300
},
{
"epoch": 3.77,
"eval_loss": 1.1190528869628906,
"eval_runtime": 435.7619,
"eval_samples_per_second": 37.098,
"eval_steps_per_second": 1.161,
"step": 4300
},
{
"epoch": 3.77,
"learning_rate": 2.938874054444323e-06,
"loss": 0.9312,
"step": 4305
},
{
"epoch": 3.78,
"learning_rate": 2.9190042803995135e-06,
"loss": 0.9459,
"step": 4310
},
{
"epoch": 3.78,
"learning_rate": 2.899190419018616e-06,
"loss": 0.9273,
"step": 4315
},
{
"epoch": 3.78,
"learning_rate": 2.8794326267550243e-06,
"loss": 0.9402,
"step": 4320
},
{
"epoch": 3.79,
"learning_rate": 2.859731059619393e-06,
"loss": 0.9412,
"step": 4325
},
{
"epoch": 3.79,
"learning_rate": 2.84008587317842e-06,
"loss": 0.9343,
"step": 4330
},
{
"epoch": 3.8,
"learning_rate": 2.8204972225536066e-06,
"loss": 0.9448,
"step": 4335
},
{
"epoch": 3.8,
"learning_rate": 2.8009652624200436e-06,
"loss": 0.9148,
"step": 4340
},
{
"epoch": 3.81,
"learning_rate": 2.7814901470051804e-06,
"loss": 0.933,
"step": 4345
},
{
"epoch": 3.81,
"learning_rate": 2.76207203008761e-06,
"loss": 0.9194,
"step": 4350
},
{
"epoch": 3.82,
"learning_rate": 2.7427110649958676e-06,
"loss": 0.9241,
"step": 4355
},
{
"epoch": 3.82,
"learning_rate": 2.7234074046071933e-06,
"loss": 0.9527,
"step": 4360
},
{
"epoch": 3.82,
"learning_rate": 2.7041612013463535e-06,
"loss": 0.9216,
"step": 4365
},
{
"epoch": 3.83,
"learning_rate": 2.6849726071844117e-06,
"loss": 0.9263,
"step": 4370
},
{
"epoch": 3.83,
"learning_rate": 2.6658417736375543e-06,
"loss": 0.9165,
"step": 4375
},
{
"epoch": 3.84,
"learning_rate": 2.6467688517658685e-06,
"loss": 0.9246,
"step": 4380
},
{
"epoch": 3.84,
"learning_rate": 2.6277539921721663e-06,
"loss": 0.9396,
"step": 4385
},
{
"epoch": 3.85,
"learning_rate": 2.6087973450007953e-06,
"loss": 0.9313,
"step": 4390
},
{
"epoch": 3.85,
"learning_rate": 2.5898990599364416e-06,
"loss": 0.9285,
"step": 4395
},
{
"epoch": 3.85,
"learning_rate": 2.5710592862029594e-06,
"loss": 0.9386,
"step": 4400
},
{
"epoch": 3.85,
"eval_loss": 1.1187002658843994,
"eval_runtime": 435.5382,
"eval_samples_per_second": 37.117,
"eval_steps_per_second": 1.162,
"step": 4400
},
{
"epoch": 3.86,
"learning_rate": 2.5522781725621814e-06,
"loss": 0.9307,
"step": 4405
},
{
"epoch": 3.86,
"learning_rate": 2.5335558673127647e-06,
"loss": 0.9365,
"step": 4410
},
{
"epoch": 3.87,
"learning_rate": 2.514892518288988e-06,
"loss": 0.9354,
"step": 4415
},
{
"epoch": 3.87,
"learning_rate": 2.4962882728596215e-06,
"loss": 0.9458,
"step": 4420
},
{
"epoch": 3.88,
"learning_rate": 2.4777432779267254e-06,
"loss": 0.9212,
"step": 4425
},
{
"epoch": 3.88,
"learning_rate": 2.459257679924524e-06,
"loss": 0.926,
"step": 4430
},
{
"epoch": 3.89,
"learning_rate": 2.440831624818224e-06,
"loss": 0.9283,
"step": 4435
},
{
"epoch": 3.89,
"learning_rate": 2.422465258102875e-06,
"loss": 0.9265,
"step": 4440
},
{
"epoch": 3.89,
"learning_rate": 2.4041587248022215e-06,
"loss": 0.9423,
"step": 4445
},
{
"epoch": 3.9,
"learning_rate": 2.385912169467547e-06,
"loss": 0.9322,
"step": 4450
},
{
"epoch": 3.9,
"learning_rate": 2.367725736176545e-06,
"loss": 0.9316,
"step": 4455
},
{
"epoch": 3.91,
"learning_rate": 2.3495995685321705e-06,
"loss": 0.9347,
"step": 4460
},
{
"epoch": 3.91,
"learning_rate": 2.331533809661518e-06,
"loss": 0.9347,
"step": 4465
},
{
"epoch": 3.92,
"learning_rate": 2.3135286022146785e-06,
"loss": 0.938,
"step": 4470
},
{
"epoch": 3.92,
"learning_rate": 2.295584088363627e-06,
"loss": 0.9361,
"step": 4475
},
{
"epoch": 3.92,
"learning_rate": 2.2777004098010834e-06,
"loss": 0.9288,
"step": 4480
},
{
"epoch": 3.93,
"learning_rate": 2.2598777077394063e-06,
"loss": 0.9234,
"step": 4485
},
{
"epoch": 3.93,
"learning_rate": 2.242116122909478e-06,
"loss": 0.936,
"step": 4490
},
{
"epoch": 3.94,
"learning_rate": 2.2244157955595835e-06,
"loss": 0.941,
"step": 4495
},
{
"epoch": 3.94,
"learning_rate": 2.206776865454312e-06,
"loss": 0.9209,
"step": 4500
},
{
"epoch": 3.94,
"eval_loss": 1.118669867515564,
"eval_runtime": 435.446,
"eval_samples_per_second": 37.125,
"eval_steps_per_second": 1.162,
"step": 4500
},
{
"epoch": 3.95,
"learning_rate": 2.1891994718734477e-06,
"loss": 0.949,
"step": 4505
},
{
"epoch": 3.95,
"learning_rate": 2.171683753610878e-06,
"loss": 0.9396,
"step": 4510
},
{
"epoch": 3.96,
"learning_rate": 2.154229848973486e-06,
"loss": 0.9165,
"step": 4515
},
{
"epoch": 3.96,
"learning_rate": 2.1368378957800716e-06,
"loss": 0.9358,
"step": 4520
},
{
"epoch": 3.96,
"learning_rate": 2.1195080313602514e-06,
"loss": 0.9135,
"step": 4525
},
{
"epoch": 3.97,
"learning_rate": 2.102240392553375e-06,
"loss": 0.9425,
"step": 4530
},
{
"epoch": 3.97,
"learning_rate": 2.08503511570746e-06,
"loss": 0.9174,
"step": 4535
},
{
"epoch": 3.98,
"learning_rate": 2.0678923366780934e-06,
"loss": 0.9472,
"step": 4540
},
{
"epoch": 3.98,
"learning_rate": 2.0508121908273783e-06,
"loss": 0.9342,
"step": 4545
},
{
"epoch": 3.99,
"learning_rate": 2.03379481302285e-06,
"loss": 0.9472,
"step": 4550
},
{
"epoch": 3.99,
"learning_rate": 2.01684033763642e-06,
"loss": 0.9377,
"step": 4555
},
{
"epoch": 3.99,
"learning_rate": 1.999948898543308e-06,
"loss": 0.9325,
"step": 4560
},
{
"epoch": 4.0,
"learning_rate": 1.983120629120998e-06,
"loss": 0.9244,
"step": 4565
},
{
"epoch": 4.0,
"learning_rate": 1.9663556622481695e-06,
"loss": 0.8944,
"step": 4570
},
{
"epoch": 4.01,
"learning_rate": 1.9496541303036533e-06,
"loss": 0.906,
"step": 4575
},
{
"epoch": 4.01,
"learning_rate": 1.9330161651653968e-06,
"loss": 0.8895,
"step": 4580
},
{
"epoch": 4.02,
"learning_rate": 1.9164418982094036e-06,
"loss": 0.9024,
"step": 4585
},
{
"epoch": 4.02,
"learning_rate": 1.899931460308716e-06,
"loss": 0.9009,
"step": 4590
},
{
"epoch": 4.03,
"learning_rate": 1.883484981832362e-06,
"loss": 0.895,
"step": 4595
},
{
"epoch": 4.03,
"learning_rate": 1.8671025926443464e-06,
"loss": 0.9028,
"step": 4600
},
{
"epoch": 4.03,
"eval_loss": 1.126094102859497,
"eval_runtime": 435.6841,
"eval_samples_per_second": 37.105,
"eval_steps_per_second": 1.161,
"step": 4600
},
{
"epoch": 4.03,
"learning_rate": 1.8507844221026061e-06,
"loss": 0.9128,
"step": 4605
},
{
"epoch": 4.04,
"learning_rate": 1.834530599058001e-06,
"loss": 0.902,
"step": 4610
},
{
"epoch": 4.04,
"learning_rate": 1.8183412518532907e-06,
"loss": 0.9013,
"step": 4615
},
{
"epoch": 4.05,
"learning_rate": 1.8022165083221298e-06,
"loss": 0.9072,
"step": 4620
},
{
"epoch": 4.05,
"learning_rate": 1.7861564957880451e-06,
"loss": 0.9039,
"step": 4625
},
{
"epoch": 4.06,
"learning_rate": 1.7701613410634367e-06,
"loss": 0.9058,
"step": 4630
},
{
"epoch": 4.06,
"learning_rate": 1.754231170448586e-06,
"loss": 0.9289,
"step": 4635
},
{
"epoch": 4.06,
"learning_rate": 1.738366109730638e-06,
"loss": 0.8887,
"step": 4640
},
{
"epoch": 4.07,
"learning_rate": 1.7225662841826308e-06,
"loss": 0.8917,
"step": 4645
},
{
"epoch": 4.07,
"learning_rate": 1.7068318185624844e-06,
"loss": 0.8864,
"step": 4650
},
{
"epoch": 4.08,
"learning_rate": 1.691162837112038e-06,
"loss": 0.9031,
"step": 4655
},
{
"epoch": 4.08,
"learning_rate": 1.6755594635560468e-06,
"loss": 0.8963,
"step": 4660
},
{
"epoch": 4.09,
"learning_rate": 1.660021821101222e-06,
"loss": 0.8957,
"step": 4665
},
{
"epoch": 4.09,
"learning_rate": 1.6445500324352493e-06,
"loss": 0.8945,
"step": 4670
},
{
"epoch": 4.1,
"learning_rate": 1.62914421972582e-06,
"loss": 0.9148,
"step": 4675
},
{
"epoch": 4.1,
"learning_rate": 1.613804504619676e-06,
"loss": 0.8986,
"step": 4680
},
{
"epoch": 4.1,
"learning_rate": 1.598531008241635e-06,
"loss": 0.9071,
"step": 4685
},
{
"epoch": 4.11,
"learning_rate": 1.5833238511936467e-06,
"loss": 0.8929,
"step": 4690
},
{
"epoch": 4.11,
"learning_rate": 1.5681831535538283e-06,
"loss": 0.897,
"step": 4695
},
{
"epoch": 4.12,
"learning_rate": 1.5531090348755307e-06,
"loss": 0.9087,
"step": 4700
},
{
"epoch": 4.12,
"eval_loss": 1.1278347969055176,
"eval_runtime": 435.9828,
"eval_samples_per_second": 37.079,
"eval_steps_per_second": 1.161,
"step": 4700
},
{
"epoch": 4.12,
"learning_rate": 1.5381016141863804e-06,
"loss": 0.9185,
"step": 4705
},
{
"epoch": 4.13,
"learning_rate": 1.5231610099873505e-06,
"loss": 0.9038,
"step": 4710
},
{
"epoch": 4.13,
"learning_rate": 1.5082873402518172e-06,
"loss": 0.907,
"step": 4715
},
{
"epoch": 4.13,
"learning_rate": 1.4934807224246338e-06,
"loss": 0.9096,
"step": 4720
},
{
"epoch": 4.14,
"learning_rate": 1.4787412734211992e-06,
"loss": 0.8883,
"step": 4725
},
{
"epoch": 4.14,
"learning_rate": 1.4640691096265358e-06,
"loss": 0.9188,
"step": 4730
},
{
"epoch": 4.15,
"learning_rate": 1.4494643468943782e-06,
"loss": 0.9065,
"step": 4735
},
{
"epoch": 4.15,
"learning_rate": 1.4349271005462395e-06,
"loss": 0.9072,
"step": 4740
},
{
"epoch": 4.16,
"learning_rate": 1.4204574853705278e-06,
"loss": 0.9056,
"step": 4745
},
{
"epoch": 4.16,
"learning_rate": 1.40605561562161e-06,
"loss": 0.9205,
"step": 4750
},
{
"epoch": 4.17,
"learning_rate": 1.3917216050189353e-06,
"loss": 0.9022,
"step": 4755
},
{
"epoch": 4.17,
"learning_rate": 1.377455566746121e-06,
"loss": 0.9068,
"step": 4760
},
{
"epoch": 4.17,
"learning_rate": 1.3632576134500653e-06,
"loss": 0.8971,
"step": 4765
},
{
"epoch": 4.18,
"learning_rate": 1.3491278572400558e-06,
"loss": 0.8975,
"step": 4770
},
{
"epoch": 4.18,
"learning_rate": 1.3350664096868837e-06,
"loss": 0.907,
"step": 4775
},
{
"epoch": 4.19,
"learning_rate": 1.3210733818219701e-06,
"loss": 0.8957,
"step": 4780
},
{
"epoch": 4.19,
"learning_rate": 1.3071488841364776e-06,
"loss": 0.8918,
"step": 4785
},
{
"epoch": 4.2,
"learning_rate": 1.2932930265804478e-06,
"loss": 0.9037,
"step": 4790
},
{
"epoch": 4.2,
"learning_rate": 1.279505918561923e-06,
"loss": 0.9081,
"step": 4795
},
{
"epoch": 4.2,
"learning_rate": 1.2657876689460979e-06,
"loss": 0.9114,
"step": 4800
},
{
"epoch": 4.2,
"eval_loss": 1.1276744604110718,
"eval_runtime": 435.7272,
"eval_samples_per_second": 37.101,
"eval_steps_per_second": 1.161,
"step": 4800
},
{
"epoch": 4.21,
"learning_rate": 1.252138386054441e-06,
"loss": 0.8939,
"step": 4805
},
{
"epoch": 4.21,
"learning_rate": 1.2385581776638512e-06,
"loss": 0.8811,
"step": 4810
},
{
"epoch": 4.22,
"learning_rate": 1.225047151005807e-06,
"loss": 0.9179,
"step": 4815
},
{
"epoch": 4.22,
"learning_rate": 1.2116054127655163e-06,
"loss": 0.9085,
"step": 4820
},
{
"epoch": 4.23,
"learning_rate": 1.1982330690810717e-06,
"loss": 0.8873,
"step": 4825
},
{
"epoch": 4.23,
"learning_rate": 1.1849302255426154e-06,
"loss": 0.895,
"step": 4830
},
{
"epoch": 4.24,
"learning_rate": 1.171696987191513e-06,
"loss": 0.8799,
"step": 4835
},
{
"epoch": 4.24,
"learning_rate": 1.158533458519504e-06,
"loss": 0.8997,
"step": 4840
},
{
"epoch": 4.24,
"learning_rate": 1.1454397434679022e-06,
"loss": 0.9045,
"step": 4845
},
{
"epoch": 4.25,
"learning_rate": 1.1324159454267503e-06,
"loss": 0.9076,
"step": 4850
},
{
"epoch": 4.25,
"learning_rate": 1.1194621672340266e-06,
"loss": 0.8929,
"step": 4855
},
{
"epoch": 4.26,
"learning_rate": 1.1065785111748117e-06,
"loss": 0.9057,
"step": 4860
},
{
"epoch": 4.26,
"learning_rate": 1.0937650789804944e-06,
"loss": 0.8989,
"step": 4865
},
{
"epoch": 4.27,
"learning_rate": 1.081021971827968e-06,
"loss": 0.9081,
"step": 4870
},
{
"epoch": 4.27,
"learning_rate": 1.068349290338825e-06,
"loss": 0.8956,
"step": 4875
},
{
"epoch": 4.28,
"learning_rate": 1.055747134578563e-06,
"loss": 0.9012,
"step": 4880
},
{
"epoch": 4.28,
"learning_rate": 1.0432156040558018e-06,
"loss": 0.8985,
"step": 4885
},
{
"epoch": 4.28,
"learning_rate": 1.0307547977214916e-06,
"loss": 0.9195,
"step": 4890
},
{
"epoch": 4.29,
"learning_rate": 1.0183648139681324e-06,
"loss": 0.9079,
"step": 4895
},
{
"epoch": 4.29,
"learning_rate": 1.006045750629001e-06,
"loss": 0.8854,
"step": 4900
},
{
"epoch": 4.29,
"eval_loss": 1.1279754638671875,
"eval_runtime": 435.9621,
"eval_samples_per_second": 37.081,
"eval_steps_per_second": 1.161,
"step": 4900
},
{
"epoch": 4.3,
"learning_rate": 9.937977049773717e-07,
"loss": 0.9037,
"step": 4905
},
{
"epoch": 4.3,
"learning_rate": 9.81620773725751e-07,
"loss": 0.8908,
"step": 4910
},
{
"epoch": 4.31,
"learning_rate": 9.695150530251207e-07,
"loss": 0.9205,
"step": 4915
},
{
"epoch": 4.31,
"learning_rate": 9.574806384641666e-07,
"loss": 0.8954,
"step": 4920
},
{
"epoch": 4.31,
"learning_rate": 9.455176250685338e-07,
"loss": 0.9131,
"step": 4925
},
{
"epoch": 4.32,
"learning_rate": 9.336261073000707e-07,
"loss": 0.9113,
"step": 4930
},
{
"epoch": 4.32,
"learning_rate": 9.218061790560862e-07,
"loss": 0.8923,
"step": 4935
},
{
"epoch": 4.33,
"learning_rate": 9.100579336686033e-07,
"loss": 0.8957,
"step": 4940
},
{
"epoch": 4.33,
"learning_rate": 8.983814639036314e-07,
"loss": 0.8905,
"step": 4945
},
{
"epoch": 4.34,
"learning_rate": 8.867768619604255e-07,
"loss": 0.9079,
"step": 4950
},
{
"epoch": 4.34,
"learning_rate": 8.752442194707577e-07,
"loss": 0.8856,
"step": 4955
},
{
"epoch": 4.35,
"learning_rate": 8.637836274982048e-07,
"loss": 0.905,
"step": 4960
},
{
"epoch": 4.35,
"learning_rate": 8.523951765374117e-07,
"loss": 0.8866,
"step": 4965
},
{
"epoch": 4.35,
"learning_rate": 8.410789565133948e-07,
"loss": 0.9056,
"step": 4970
},
{
"epoch": 4.36,
"learning_rate": 8.298350567808178e-07,
"loss": 0.8848,
"step": 4975
},
{
"epoch": 4.36,
"learning_rate": 8.186635661232966e-07,
"loss": 0.914,
"step": 4980
},
{
"epoch": 4.37,
"learning_rate": 8.075645727526893e-07,
"loss": 0.9114,
"step": 4985
},
{
"epoch": 4.37,
"learning_rate": 7.965381643084069e-07,
"loss": 0.9009,
"step": 4990
},
{
"epoch": 4.38,
"learning_rate": 7.85584427856716e-07,
"loss": 0.9231,
"step": 4995
},
{
"epoch": 4.38,
"learning_rate": 7.74703449890053e-07,
"loss": 0.902,
"step": 5000
},
{
"epoch": 4.38,
"eval_loss": 1.127801537513733,
"eval_runtime": 435.7089,
"eval_samples_per_second": 37.103,
"eval_steps_per_second": 1.161,
"step": 5000
},
{
"epoch": 4.38,
"learning_rate": 7.638953163263485e-07,
"loss": 0.8891,
"step": 5005
},
{
"epoch": 4.39,
"learning_rate": 7.531601125083321e-07,
"loss": 0.9126,
"step": 5010
},
{
"epoch": 4.39,
"learning_rate": 7.424979232028784e-07,
"loss": 0.9055,
"step": 5015
},
{
"epoch": 4.4,
"learning_rate": 7.319088326003187e-07,
"loss": 0.8927,
"step": 5020
},
{
"epoch": 4.4,
"learning_rate": 7.213929243137952e-07,
"loss": 0.897,
"step": 5025
},
{
"epoch": 4.41,
"learning_rate": 7.109502813785818e-07,
"loss": 0.9028,
"step": 5030
},
{
"epoch": 4.41,
"learning_rate": 7.005809862514412e-07,
"loss": 0.8991,
"step": 5035
},
{
"epoch": 4.42,
"learning_rate": 6.902851208099704e-07,
"loss": 0.8951,
"step": 5040
},
{
"epoch": 4.42,
"learning_rate": 6.800627663519566e-07,
"loss": 0.8852,
"step": 5045
},
{
"epoch": 4.42,
"learning_rate": 6.699140035947271e-07,
"loss": 0.8994,
"step": 5050
},
{
"epoch": 4.43,
"learning_rate": 6.598389126745209e-07,
"loss": 0.9086,
"step": 5055
},
{
"epoch": 4.43,
"learning_rate": 6.498375731458529e-07,
"loss": 0.9007,
"step": 5060
},
{
"epoch": 4.44,
"learning_rate": 6.399100639808819e-07,
"loss": 0.8997,
"step": 5065
},
{
"epoch": 4.44,
"learning_rate": 6.300564635687978e-07,
"loss": 0.8955,
"step": 5070
},
{
"epoch": 4.45,
"learning_rate": 6.202768497151845e-07,
"loss": 0.8908,
"step": 5075
},
{
"epoch": 4.45,
"learning_rate": 6.10571299641427e-07,
"loss": 0.8981,
"step": 5080
},
{
"epoch": 4.45,
"learning_rate": 6.009398899840835e-07,
"loss": 0.9026,
"step": 5085
},
{
"epoch": 4.46,
"learning_rate": 5.913826967942926e-07,
"loss": 0.9035,
"step": 5090
},
{
"epoch": 4.46,
"learning_rate": 5.81899795537163e-07,
"loss": 0.9109,
"step": 5095
},
{
"epoch": 4.47,
"learning_rate": 5.724912610911848e-07,
"loss": 0.9038,
"step": 5100
},
{
"epoch": 4.47,
"eval_loss": 1.1280251741409302,
"eval_runtime": 436.2069,
"eval_samples_per_second": 37.06,
"eval_steps_per_second": 1.16,
"step": 5100
},
{
"epoch": 4.47,
"learning_rate": 5.6315716774764e-07,
"loss": 0.905,
"step": 5105
},
{
"epoch": 4.48,
"learning_rate": 5.538975892100041e-07,
"loss": 0.8944,
"step": 5110
},
{
"epoch": 4.48,
"learning_rate": 5.44712598593381e-07,
"loss": 0.9034,
"step": 5115
},
{
"epoch": 4.49,
"learning_rate": 5.35602268423906e-07,
"loss": 0.9053,
"step": 5120
},
{
"epoch": 4.49,
"learning_rate": 5.265666706381944e-07,
"loss": 0.9201,
"step": 5125
},
{
"epoch": 4.49,
"learning_rate": 5.176058765827541e-07,
"loss": 0.9142,
"step": 5130
},
{
"epoch": 4.5,
"learning_rate": 5.087199570134371e-07,
"loss": 0.9092,
"step": 5135
},
{
"epoch": 4.5,
"learning_rate": 4.9990898209487e-07,
"loss": 0.9046,
"step": 5140
},
{
"epoch": 4.51,
"learning_rate": 4.911730213999078e-07,
"loss": 0.8922,
"step": 5145
},
{
"epoch": 4.51,
"learning_rate": 4.825121439090785e-07,
"loss": 0.8799,
"step": 5150
},
{
"epoch": 4.52,
"learning_rate": 4.7392641801004336e-07,
"loss": 0.9117,
"step": 5155
},
{
"epoch": 4.52,
"learning_rate": 4.654159114970547e-07,
"loss": 0.8987,
"step": 5160
},
{
"epoch": 4.52,
"learning_rate": 4.5698069157042e-07,
"loss": 0.902,
"step": 5165
},
{
"epoch": 4.53,
"learning_rate": 4.4862082483597247e-07,
"loss": 0.9031,
"step": 5170
},
{
"epoch": 4.53,
"learning_rate": 4.4033637730454103e-07,
"loss": 0.9031,
"step": 5175
},
{
"epoch": 4.54,
"learning_rate": 4.321274143914411e-07,
"loss": 0.8961,
"step": 5180
},
{
"epoch": 4.54,
"learning_rate": 4.2399400091594154e-07,
"loss": 0.9032,
"step": 5185
},
{
"epoch": 4.55,
"learning_rate": 4.159362011007606e-07,
"loss": 0.9152,
"step": 5190
},
{
"epoch": 4.55,
"learning_rate": 4.079540785715674e-07,
"loss": 0.9044,
"step": 5195
},
{
"epoch": 4.56,
"learning_rate": 4.000476963564637e-07,
"loss": 0.8935,
"step": 5200
},
{
"epoch": 4.56,
"eval_loss": 1.1279747486114502,
"eval_runtime": 435.5117,
"eval_samples_per_second": 37.12,
"eval_steps_per_second": 1.162,
"step": 5200
},
{
"epoch": 4.56,
"learning_rate": 3.9221711688549493e-07,
"loss": 0.8852,
"step": 5205
},
{
"epoch": 4.56,
"learning_rate": 3.844624019901566e-07,
"loss": 0.9019,
"step": 5210
},
{
"epoch": 4.57,
"learning_rate": 3.76783612902909e-07,
"loss": 0.8971,
"step": 5215
},
{
"epoch": 4.57,
"learning_rate": 3.6918081025668296e-07,
"loss": 0.8902,
"step": 5220
},
{
"epoch": 4.58,
"learning_rate": 3.616540540844171e-07,
"loss": 0.9043,
"step": 5225
},
{
"epoch": 4.58,
"learning_rate": 3.54203403818566e-07,
"loss": 0.8943,
"step": 5230
},
{
"epoch": 4.59,
"learning_rate": 3.4682891829064923e-07,
"loss": 0.9106,
"step": 5235
},
{
"epoch": 4.59,
"learning_rate": 3.395306557307698e-07,
"loss": 0.8853,
"step": 5240
},
{
"epoch": 4.59,
"learning_rate": 3.3230867376716324e-07,
"loss": 0.8912,
"step": 5245
},
{
"epoch": 4.6,
"learning_rate": 3.2516302942574794e-07,
"loss": 0.9133,
"step": 5250
},
{
"epoch": 4.6,
"learning_rate": 3.180937791296601e-07,
"loss": 0.9023,
"step": 5255
},
{
"epoch": 4.61,
"learning_rate": 3.1110097869882063e-07,
"loss": 0.9151,
"step": 5260
},
{
"epoch": 4.61,
"learning_rate": 3.041846833494888e-07,
"loss": 0.9023,
"step": 5265
},
{
"epoch": 4.62,
"learning_rate": 2.973449476938306e-07,
"loss": 0.9021,
"step": 5270
},
{
"epoch": 4.62,
"learning_rate": 2.905818257394799e-07,
"loss": 0.9142,
"step": 5275
},
{
"epoch": 4.63,
"learning_rate": 2.838953708891201e-07,
"loss": 0.8908,
"step": 5280
},
{
"epoch": 4.63,
"learning_rate": 2.772856359400589e-07,
"loss": 0.9061,
"step": 5285
},
{
"epoch": 4.63,
"learning_rate": 2.7075267308380726e-07,
"loss": 0.8914,
"step": 5290
},
{
"epoch": 4.64,
"learning_rate": 2.64296533905678e-07,
"loss": 0.9053,
"step": 5295
},
{
"epoch": 4.64,
"learning_rate": 2.579172693843679e-07,
"loss": 0.9053,
"step": 5300
},
{
"epoch": 4.64,
"eval_loss": 1.1279793977737427,
"eval_runtime": 436.5436,
"eval_samples_per_second": 37.032,
"eval_steps_per_second": 1.159,
"step": 5300
},
{
"epoch": 4.65,
"learning_rate": 2.5161492989156046e-07,
"loss": 0.8985,
"step": 5305
},
{
"epoch": 4.65,
"learning_rate": 2.45389565191525e-07,
"loss": 0.9,
"step": 5310
},
{
"epoch": 4.66,
"learning_rate": 2.392412244407294e-07,
"loss": 0.8978,
"step": 5315
},
{
"epoch": 4.66,
"learning_rate": 2.3316995618744343e-07,
"loss": 0.9026,
"step": 5320
},
{
"epoch": 4.66,
"learning_rate": 2.271758083713671e-07,
"loss": 0.8969,
"step": 5325
},
{
"epoch": 4.67,
"learning_rate": 2.2125882832323865e-07,
"loss": 0.9007,
"step": 5330
},
{
"epoch": 4.67,
"learning_rate": 2.1541906276446922e-07,
"loss": 0.9003,
"step": 5335
},
{
"epoch": 4.68,
"learning_rate": 2.0965655780677441e-07,
"loss": 0.8952,
"step": 5340
},
{
"epoch": 4.68,
"learning_rate": 2.0397135895180443e-07,
"loss": 0.9143,
"step": 5345
},
{
"epoch": 4.69,
"learning_rate": 1.983635110907922e-07,
"loss": 0.8936,
"step": 5350
},
{
"epoch": 4.69,
"learning_rate": 1.9283305850419153e-07,
"loss": 0.9068,
"step": 5355
},
{
"epoch": 4.7,
"learning_rate": 1.8738004486133166e-07,
"loss": 0.9059,
"step": 5360
},
{
"epoch": 4.7,
"learning_rate": 1.8200451322007208e-07,
"loss": 0.9156,
"step": 5365
},
{
"epoch": 4.7,
"learning_rate": 1.76706506026465e-07,
"loss": 0.8897,
"step": 5370
},
{
"epoch": 4.71,
"learning_rate": 1.7148606511441123e-07,
"loss": 0.8995,
"step": 5375
},
{
"epoch": 4.71,
"learning_rate": 1.6634323170533928e-07,
"loss": 0.8976,
"step": 5380
},
{
"epoch": 4.72,
"learning_rate": 1.6127804640787781e-07,
"loss": 0.8973,
"step": 5385
},
{
"epoch": 4.72,
"learning_rate": 1.562905492175326e-07,
"loss": 0.9179,
"step": 5390
},
{
"epoch": 4.73,
"learning_rate": 1.513807795163702e-07,
"loss": 0.9091,
"step": 5395
},
{
"epoch": 4.73,
"learning_rate": 1.4654877607271024e-07,
"loss": 0.9091,
"step": 5400
},
{
"epoch": 4.73,
"eval_loss": 1.1277990341186523,
"eval_runtime": 435.7368,
"eval_samples_per_second": 37.1,
"eval_steps_per_second": 1.161,
"step": 5400
},
{
"epoch": 4.73,
"learning_rate": 1.417945770408191e-07,
"loss": 0.8984,
"step": 5405
},
{
"epoch": 4.74,
"learning_rate": 1.371182199606036e-07,
"loss": 0.9025,
"step": 5410
},
{
"epoch": 4.74,
"learning_rate": 1.3251974175731986e-07,
"loss": 0.913,
"step": 5415
},
{
"epoch": 4.75,
"learning_rate": 1.2799917874128154e-07,
"loss": 0.9018,
"step": 5420
},
{
"epoch": 4.75,
"learning_rate": 1.2355656660757e-07,
"loss": 0.8923,
"step": 5425
},
{
"epoch": 4.76,
"learning_rate": 1.1919194043575333e-07,
"loss": 0.9171,
"step": 5430
},
{
"epoch": 4.76,
"learning_rate": 1.1490533468961118e-07,
"loss": 0.8981,
"step": 5435
},
{
"epoch": 4.77,
"learning_rate": 1.1069678321686372e-07,
"loss": 0.8963,
"step": 5440
},
{
"epoch": 4.77,
"learning_rate": 1.0656631924889749e-07,
"loss": 0.9035,
"step": 5445
},
{
"epoch": 4.77,
"learning_rate": 1.0251397540051334e-07,
"loss": 0.903,
"step": 5450
},
{
"epoch": 4.78,
"learning_rate": 9.853978366965667e-08,
"loss": 0.9133,
"step": 5455
},
{
"epoch": 4.78,
"learning_rate": 9.464377543717651e-08,
"loss": 0.8988,
"step": 5460
},
{
"epoch": 4.79,
"learning_rate": 9.08259814665724e-08,
"loss": 0.9008,
"step": 5465
},
{
"epoch": 4.79,
"learning_rate": 8.7086431903749e-08,
"loss": 0.8861,
"step": 5470
},
{
"epoch": 4.8,
"learning_rate": 8.342515627678182e-08,
"loss": 0.9158,
"step": 5475
},
{
"epoch": 4.8,
"learning_rate": 7.984218349568307e-08,
"loss": 0.9057,
"step": 5480
},
{
"epoch": 4.81,
"learning_rate": 7.63375418521739e-08,
"loss": 0.8892,
"step": 5485
},
{
"epoch": 4.81,
"learning_rate": 7.291125901946027e-08,
"loss": 0.9238,
"step": 5490
},
{
"epoch": 4.81,
"learning_rate": 6.956336205201419e-08,
"loss": 0.8901,
"step": 5495
},
{
"epoch": 4.82,
"learning_rate": 6.62938773853583e-08,
"loss": 0.8968,
"step": 5500
},
{
"epoch": 4.82,
"eval_loss": 1.127881646156311,
"eval_runtime": 437.416,
"eval_samples_per_second": 36.958,
"eval_steps_per_second": 1.157,
"step": 5500
},
{
"epoch": 4.82,
"learning_rate": 6.310283083586499e-08,
"loss": 0.9202,
"step": 5505
},
{
"epoch": 4.83,
"learning_rate": 5.999024760054095e-08,
"loss": 0.9119,
"step": 5510
},
{
"epoch": 4.83,
"learning_rate": 5.695615225683737e-08,
"loss": 0.9056,
"step": 5515
},
{
"epoch": 4.84,
"learning_rate": 5.400056876245119e-08,
"loss": 0.8842,
"step": 5520
},
{
"epoch": 4.84,
"learning_rate": 5.1123520455137465e-08,
"loss": 0.8926,
"step": 5525
},
{
"epoch": 4.84,
"learning_rate": 4.832503005252731e-08,
"loss": 0.9064,
"step": 5530
},
{
"epoch": 4.85,
"learning_rate": 4.5605119651941366e-08,
"loss": 0.9129,
"step": 5535
},
{
"epoch": 4.85,
"learning_rate": 4.296381073022549e-08,
"loss": 0.906,
"step": 5540
},
{
"epoch": 4.86,
"learning_rate": 4.040112414357311e-08,
"loss": 0.8982,
"step": 5545
},
{
"epoch": 4.86,
"learning_rate": 3.791708012736206e-08,
"loss": 0.9011,
"step": 5550
},
{
"epoch": 4.87,
"learning_rate": 3.551169829599799e-08,
"loss": 0.8882,
"step": 5555
},
{
"epoch": 4.87,
"learning_rate": 3.318499764275784e-08,
"loss": 0.8883,
"step": 5560
},
{
"epoch": 4.88,
"learning_rate": 3.0936996539636663e-08,
"loss": 0.8906,
"step": 5565
},
{
"epoch": 4.88,
"learning_rate": 2.876771273720991e-08,
"loss": 0.903,
"step": 5570
},
{
"epoch": 4.88,
"learning_rate": 2.667716336448356e-08,
"loss": 0.9019,
"step": 5575
},
{
"epoch": 4.89,
"learning_rate": 2.4665364928767586e-08,
"loss": 0.8916,
"step": 5580
},
{
"epoch": 4.89,
"learning_rate": 2.273233331554159e-08,
"loss": 0.8997,
"step": 5585
},
{
"epoch": 4.9,
"learning_rate": 2.087808378832934e-08,
"loss": 0.9029,
"step": 5590
},
{
"epoch": 4.9,
"learning_rate": 1.9102630988577788e-08,
"loss": 0.885,
"step": 5595
},
{
"epoch": 4.91,
"learning_rate": 1.7405988935544906e-08,
"loss": 0.9196,
"step": 5600
},
{
"epoch": 4.91,
"eval_loss": 1.1278820037841797,
"eval_runtime": 436.3538,
"eval_samples_per_second": 37.048,
"eval_steps_per_second": 1.16,
"step": 5600
},
{
"epoch": 4.91,
"learning_rate": 1.578817102618424e-08,
"loss": 0.8877,
"step": 5605
},
{
"epoch": 4.91,
"learning_rate": 1.424919003504166e-08,
"loss": 0.9056,
"step": 5610
},
{
"epoch": 4.92,
"learning_rate": 1.2789058114156538e-08,
"loss": 0.8966,
"step": 5615
},
{
"epoch": 4.92,
"learning_rate": 1.140778679296073e-08,
"loss": 0.9023,
"step": 5620
},
{
"epoch": 4.93,
"learning_rate": 1.0105386978191967e-08,
"loss": 0.9105,
"step": 5625
},
{
"epoch": 4.93,
"learning_rate": 8.88186895380505e-09,
"loss": 0.9008,
"step": 5630
},
{
"epoch": 4.94,
"learning_rate": 7.737242380891907e-09,
"loss": 0.9024,
"step": 5635
},
{
"epoch": 4.94,
"learning_rate": 6.671516297606095e-09,
"loss": 0.8923,
"step": 5640
},
{
"epoch": 4.95,
"learning_rate": 5.684699119089532e-09,
"loss": 0.905,
"step": 5645
},
{
"epoch": 4.95,
"learning_rate": 4.776798637405877e-09,
"loss": 0.8933,
"step": 5650
},
{
"epoch": 4.95,
"learning_rate": 3.9478220214816956e-09,
"loss": 0.8988,
"step": 5655
},
{
"epoch": 4.96,
"learning_rate": 3.1977758170476105e-09,
"loss": 0.8974,
"step": 5660
},
{
"epoch": 4.96,
"learning_rate": 2.5266659465883468e-09,
"loss": 0.8958,
"step": 5665
},
{
"epoch": 4.97,
"learning_rate": 1.9344977092927707e-09,
"loss": 0.8963,
"step": 5670
},
{
"epoch": 4.97,
"learning_rate": 1.4212757810150301e-09,
"loss": 0.8992,
"step": 5675
},
{
"epoch": 4.98,
"learning_rate": 9.870042142368085e-10,
"loss": 0.9161,
"step": 5680
},
{
"epoch": 4.98,
"learning_rate": 6.316864380351284e-10,
"loss": 0.9143,
"step": 5685
},
{
"epoch": 4.98,
"learning_rate": 3.553252580557054e-10,
"loss": 0.8801,
"step": 5690
},
{
"epoch": 4.99,
"learning_rate": 1.57922856489634e-10,
"loss": 0.9072,
"step": 5695
},
{
"epoch": 4.99,
"learning_rate": 3.948079205895461e-11,
"loss": 0.9129,
"step": 5700
},
{
"epoch": 4.99,
"eval_loss": 1.1278934478759766,
"eval_runtime": 435.9648,
"eval_samples_per_second": 37.081,
"eval_steps_per_second": 1.161,
"step": 5700
},
{
"epoch": 5.0,
"learning_rate": 0.0,
"loss": 0.8974,
"step": 5705
},
{
"epoch": 5.0,
"step": 5705,
"total_flos": 5.71030084361021e+19,
"train_loss": 1.0224046702974203,
"train_runtime": 86934.362,
"train_samples_per_second": 8.4,
"train_steps_per_second": 0.066
}
],
"logging_steps": 5,
"max_steps": 5705,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 100,
"total_flos": 5.71030084361021e+19,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}