100rab25's picture
End of training
7b6f065 verified
{
"best_metric": 0.9558704453441296,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-hotel_images_classifier_v5_10epocs/checkpoint-6940",
"epoch": 9.989204749910039,
"eval_steps": 500,
"global_step": 6940,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"grad_norm": 5.388793468475342,
"learning_rate": 3.602305475504323e-07,
"loss": 2.0168,
"step": 5
},
{
"epoch": 0.01,
"grad_norm": 5.493525981903076,
"learning_rate": 7.204610951008646e-07,
"loss": 1.9925,
"step": 10
},
{
"epoch": 0.02,
"grad_norm": 4.990756034851074,
"learning_rate": 1.080691642651297e-06,
"loss": 1.9873,
"step": 15
},
{
"epoch": 0.03,
"grad_norm": 4.94097900390625,
"learning_rate": 1.4409221902017292e-06,
"loss": 1.9782,
"step": 20
},
{
"epoch": 0.04,
"grad_norm": 5.246731281280518,
"learning_rate": 1.8011527377521614e-06,
"loss": 1.9525,
"step": 25
},
{
"epoch": 0.04,
"grad_norm": 4.839789867401123,
"learning_rate": 2.161383285302594e-06,
"loss": 1.9125,
"step": 30
},
{
"epoch": 0.05,
"grad_norm": 5.326338291168213,
"learning_rate": 2.521613832853026e-06,
"loss": 1.9043,
"step": 35
},
{
"epoch": 0.06,
"grad_norm": 4.707699298858643,
"learning_rate": 2.8818443804034585e-06,
"loss": 1.8724,
"step": 40
},
{
"epoch": 0.06,
"grad_norm": 4.742429256439209,
"learning_rate": 3.2420749279538904e-06,
"loss": 1.8261,
"step": 45
},
{
"epoch": 0.07,
"grad_norm": 7.061827659606934,
"learning_rate": 3.602305475504323e-06,
"loss": 1.783,
"step": 50
},
{
"epoch": 0.08,
"grad_norm": 5.174450397491455,
"learning_rate": 3.962536023054755e-06,
"loss": 1.7679,
"step": 55
},
{
"epoch": 0.09,
"grad_norm": 4.2161641120910645,
"learning_rate": 4.322766570605188e-06,
"loss": 1.7214,
"step": 60
},
{
"epoch": 0.09,
"grad_norm": 6.96381950378418,
"learning_rate": 4.68299711815562e-06,
"loss": 1.6884,
"step": 65
},
{
"epoch": 0.1,
"grad_norm": 5.790141582489014,
"learning_rate": 5.043227665706052e-06,
"loss": 1.5961,
"step": 70
},
{
"epoch": 0.11,
"grad_norm": 5.271084308624268,
"learning_rate": 5.403458213256484e-06,
"loss": 1.5396,
"step": 75
},
{
"epoch": 0.12,
"grad_norm": 5.163028717041016,
"learning_rate": 5.763688760806917e-06,
"loss": 1.4907,
"step": 80
},
{
"epoch": 0.12,
"grad_norm": 4.985466003417969,
"learning_rate": 6.123919308357349e-06,
"loss": 1.3996,
"step": 85
},
{
"epoch": 0.13,
"grad_norm": 5.5068745613098145,
"learning_rate": 6.484149855907781e-06,
"loss": 1.3632,
"step": 90
},
{
"epoch": 0.14,
"grad_norm": 4.985295295715332,
"learning_rate": 6.844380403458213e-06,
"loss": 1.2596,
"step": 95
},
{
"epoch": 0.14,
"grad_norm": 5.087879657745361,
"learning_rate": 7.204610951008646e-06,
"loss": 1.1991,
"step": 100
},
{
"epoch": 0.15,
"grad_norm": 5.104519844055176,
"learning_rate": 7.564841498559078e-06,
"loss": 1.1213,
"step": 105
},
{
"epoch": 0.16,
"grad_norm": 7.174159049987793,
"learning_rate": 7.92507204610951e-06,
"loss": 1.0379,
"step": 110
},
{
"epoch": 0.17,
"grad_norm": 7.0040788650512695,
"learning_rate": 8.285302593659942e-06,
"loss": 0.976,
"step": 115
},
{
"epoch": 0.17,
"grad_norm": 6.505568981170654,
"learning_rate": 8.645533141210376e-06,
"loss": 0.9049,
"step": 120
},
{
"epoch": 0.18,
"grad_norm": 5.980679988861084,
"learning_rate": 9.005763688760807e-06,
"loss": 0.8347,
"step": 125
},
{
"epoch": 0.19,
"grad_norm": 7.585896968841553,
"learning_rate": 9.36599423631124e-06,
"loss": 0.7498,
"step": 130
},
{
"epoch": 0.19,
"grad_norm": 10.884958267211914,
"learning_rate": 9.726224783861671e-06,
"loss": 0.7642,
"step": 135
},
{
"epoch": 0.2,
"grad_norm": 6.63291072845459,
"learning_rate": 1.0086455331412104e-05,
"loss": 0.7189,
"step": 140
},
{
"epoch": 0.21,
"grad_norm": 7.328841209411621,
"learning_rate": 1.0446685878962537e-05,
"loss": 0.6749,
"step": 145
},
{
"epoch": 0.22,
"grad_norm": 6.733003616333008,
"learning_rate": 1.0806916426512968e-05,
"loss": 0.6278,
"step": 150
},
{
"epoch": 0.22,
"grad_norm": 5.942975044250488,
"learning_rate": 1.1167146974063401e-05,
"loss": 0.6223,
"step": 155
},
{
"epoch": 0.23,
"grad_norm": 7.776107311248779,
"learning_rate": 1.1527377521613834e-05,
"loss": 0.6424,
"step": 160
},
{
"epoch": 0.24,
"grad_norm": 9.42983341217041,
"learning_rate": 1.1887608069164267e-05,
"loss": 0.6153,
"step": 165
},
{
"epoch": 0.24,
"grad_norm": 7.377403259277344,
"learning_rate": 1.2247838616714698e-05,
"loss": 0.5434,
"step": 170
},
{
"epoch": 0.25,
"grad_norm": 8.021675109863281,
"learning_rate": 1.2608069164265129e-05,
"loss": 0.5843,
"step": 175
},
{
"epoch": 0.26,
"grad_norm": 9.733675003051758,
"learning_rate": 1.2968299711815562e-05,
"loss": 0.5932,
"step": 180
},
{
"epoch": 0.27,
"grad_norm": 6.97348690032959,
"learning_rate": 1.3328530259365996e-05,
"loss": 0.5906,
"step": 185
},
{
"epoch": 0.27,
"grad_norm": 7.889408111572266,
"learning_rate": 1.3688760806916426e-05,
"loss": 0.5389,
"step": 190
},
{
"epoch": 0.28,
"grad_norm": 8.591092109680176,
"learning_rate": 1.4048991354466859e-05,
"loss": 0.5179,
"step": 195
},
{
"epoch": 0.29,
"grad_norm": 11.209196090698242,
"learning_rate": 1.4409221902017291e-05,
"loss": 0.5233,
"step": 200
},
{
"epoch": 0.3,
"grad_norm": 9.045553207397461,
"learning_rate": 1.4769452449567724e-05,
"loss": 0.5049,
"step": 205
},
{
"epoch": 0.3,
"grad_norm": 6.595673561096191,
"learning_rate": 1.5129682997118155e-05,
"loss": 0.4618,
"step": 210
},
{
"epoch": 0.31,
"grad_norm": 6.817091941833496,
"learning_rate": 1.548991354466859e-05,
"loss": 0.4885,
"step": 215
},
{
"epoch": 0.32,
"grad_norm": 20.116256713867188,
"learning_rate": 1.585014409221902e-05,
"loss": 0.4788,
"step": 220
},
{
"epoch": 0.32,
"grad_norm": 5.532088279724121,
"learning_rate": 1.6210374639769456e-05,
"loss": 0.5049,
"step": 225
},
{
"epoch": 0.33,
"grad_norm": 8.998950004577637,
"learning_rate": 1.6570605187319883e-05,
"loss": 0.469,
"step": 230
},
{
"epoch": 0.34,
"grad_norm": 8.597137451171875,
"learning_rate": 1.6930835734870318e-05,
"loss": 0.5467,
"step": 235
},
{
"epoch": 0.35,
"grad_norm": 6.714593887329102,
"learning_rate": 1.7291066282420752e-05,
"loss": 0.4635,
"step": 240
},
{
"epoch": 0.35,
"grad_norm": 7.317856788635254,
"learning_rate": 1.765129682997118e-05,
"loss": 0.4495,
"step": 245
},
{
"epoch": 0.36,
"grad_norm": 7.8448591232299805,
"learning_rate": 1.8011527377521615e-05,
"loss": 0.4652,
"step": 250
},
{
"epoch": 0.37,
"grad_norm": 7.0624213218688965,
"learning_rate": 1.8371757925072046e-05,
"loss": 0.4267,
"step": 255
},
{
"epoch": 0.37,
"grad_norm": 8.130319595336914,
"learning_rate": 1.873198847262248e-05,
"loss": 0.5095,
"step": 260
},
{
"epoch": 0.38,
"grad_norm": 12.399904251098633,
"learning_rate": 1.909221902017291e-05,
"loss": 0.4315,
"step": 265
},
{
"epoch": 0.39,
"grad_norm": 8.656498908996582,
"learning_rate": 1.9452449567723343e-05,
"loss": 0.442,
"step": 270
},
{
"epoch": 0.4,
"grad_norm": 9.993114471435547,
"learning_rate": 1.9812680115273777e-05,
"loss": 0.4471,
"step": 275
},
{
"epoch": 0.4,
"grad_norm": 8.428305625915527,
"learning_rate": 2.017291066282421e-05,
"loss": 0.4121,
"step": 280
},
{
"epoch": 0.41,
"grad_norm": 11.432868003845215,
"learning_rate": 2.053314121037464e-05,
"loss": 0.5125,
"step": 285
},
{
"epoch": 0.42,
"grad_norm": 6.569263935089111,
"learning_rate": 2.0893371757925074e-05,
"loss": 0.4177,
"step": 290
},
{
"epoch": 0.42,
"grad_norm": 6.448131084442139,
"learning_rate": 2.1253602305475505e-05,
"loss": 0.4183,
"step": 295
},
{
"epoch": 0.43,
"grad_norm": 6.2162628173828125,
"learning_rate": 2.1613832853025936e-05,
"loss": 0.378,
"step": 300
},
{
"epoch": 0.44,
"grad_norm": 15.349369049072266,
"learning_rate": 2.197406340057637e-05,
"loss": 0.3972,
"step": 305
},
{
"epoch": 0.45,
"grad_norm": 11.309712409973145,
"learning_rate": 2.2334293948126802e-05,
"loss": 0.3721,
"step": 310
},
{
"epoch": 0.45,
"grad_norm": 11.087018966674805,
"learning_rate": 2.2694524495677236e-05,
"loss": 0.4345,
"step": 315
},
{
"epoch": 0.46,
"grad_norm": 8.34060001373291,
"learning_rate": 2.3054755043227668e-05,
"loss": 0.4085,
"step": 320
},
{
"epoch": 0.47,
"grad_norm": 8.439334869384766,
"learning_rate": 2.34149855907781e-05,
"loss": 0.3855,
"step": 325
},
{
"epoch": 0.47,
"grad_norm": 7.37805700302124,
"learning_rate": 2.3775216138328533e-05,
"loss": 0.4104,
"step": 330
},
{
"epoch": 0.48,
"grad_norm": 10.912667274475098,
"learning_rate": 2.413544668587896e-05,
"loss": 0.4226,
"step": 335
},
{
"epoch": 0.49,
"grad_norm": 8.15848159790039,
"learning_rate": 2.4495677233429396e-05,
"loss": 0.4045,
"step": 340
},
{
"epoch": 0.5,
"grad_norm": 9.146815299987793,
"learning_rate": 2.485590778097983e-05,
"loss": 0.3754,
"step": 345
},
{
"epoch": 0.5,
"grad_norm": 24.85068130493164,
"learning_rate": 2.5216138328530258e-05,
"loss": 0.3406,
"step": 350
},
{
"epoch": 0.51,
"grad_norm": 9.068620681762695,
"learning_rate": 2.5576368876080692e-05,
"loss": 0.3673,
"step": 355
},
{
"epoch": 0.52,
"grad_norm": 9.011393547058105,
"learning_rate": 2.5936599423631124e-05,
"loss": 0.3766,
"step": 360
},
{
"epoch": 0.53,
"grad_norm": 7.080930233001709,
"learning_rate": 2.6296829971181558e-05,
"loss": 0.4243,
"step": 365
},
{
"epoch": 0.53,
"grad_norm": 8.521806716918945,
"learning_rate": 2.6657060518731993e-05,
"loss": 0.3984,
"step": 370
},
{
"epoch": 0.54,
"grad_norm": 8.133712768554688,
"learning_rate": 2.7017291066282424e-05,
"loss": 0.3957,
"step": 375
},
{
"epoch": 0.55,
"grad_norm": 8.477057456970215,
"learning_rate": 2.737752161383285e-05,
"loss": 0.4256,
"step": 380
},
{
"epoch": 0.55,
"grad_norm": 7.391529083251953,
"learning_rate": 2.7737752161383286e-05,
"loss": 0.3621,
"step": 385
},
{
"epoch": 0.56,
"grad_norm": 7.98112154006958,
"learning_rate": 2.8097982708933717e-05,
"loss": 0.4113,
"step": 390
},
{
"epoch": 0.57,
"grad_norm": 6.7378249168396,
"learning_rate": 2.845821325648415e-05,
"loss": 0.3808,
"step": 395
},
{
"epoch": 0.58,
"grad_norm": 8.315476417541504,
"learning_rate": 2.8818443804034583e-05,
"loss": 0.4359,
"step": 400
},
{
"epoch": 0.58,
"grad_norm": 4.960958480834961,
"learning_rate": 2.9178674351585017e-05,
"loss": 0.3533,
"step": 405
},
{
"epoch": 0.59,
"grad_norm": 7.491067886352539,
"learning_rate": 2.953890489913545e-05,
"loss": 0.3839,
"step": 410
},
{
"epoch": 0.6,
"grad_norm": 8.103771209716797,
"learning_rate": 2.9899135446685883e-05,
"loss": 0.3777,
"step": 415
},
{
"epoch": 0.6,
"grad_norm": 7.958097457885742,
"learning_rate": 3.025936599423631e-05,
"loss": 0.3984,
"step": 420
},
{
"epoch": 0.61,
"grad_norm": 9.805606842041016,
"learning_rate": 3.0619596541786745e-05,
"loss": 0.3952,
"step": 425
},
{
"epoch": 0.62,
"grad_norm": 10.124835014343262,
"learning_rate": 3.097982708933718e-05,
"loss": 0.3892,
"step": 430
},
{
"epoch": 0.63,
"grad_norm": 6.68670654296875,
"learning_rate": 3.134005763688761e-05,
"loss": 0.4251,
"step": 435
},
{
"epoch": 0.63,
"grad_norm": 6.3270440101623535,
"learning_rate": 3.170028818443804e-05,
"loss": 0.4049,
"step": 440
},
{
"epoch": 0.64,
"grad_norm": 5.105160236358643,
"learning_rate": 3.206051873198848e-05,
"loss": 0.3664,
"step": 445
},
{
"epoch": 0.65,
"grad_norm": 6.3673601150512695,
"learning_rate": 3.242074927953891e-05,
"loss": 0.3747,
"step": 450
},
{
"epoch": 0.65,
"grad_norm": 10.308377265930176,
"learning_rate": 3.278097982708934e-05,
"loss": 0.4098,
"step": 455
},
{
"epoch": 0.66,
"grad_norm": 5.274810791015625,
"learning_rate": 3.314121037463977e-05,
"loss": 0.3722,
"step": 460
},
{
"epoch": 0.67,
"grad_norm": 4.596935272216797,
"learning_rate": 3.35014409221902e-05,
"loss": 0.3155,
"step": 465
},
{
"epoch": 0.68,
"grad_norm": 13.104801177978516,
"learning_rate": 3.3861671469740636e-05,
"loss": 0.3442,
"step": 470
},
{
"epoch": 0.68,
"grad_norm": 7.388278484344482,
"learning_rate": 3.422190201729107e-05,
"loss": 0.3528,
"step": 475
},
{
"epoch": 0.69,
"grad_norm": 5.649425029754639,
"learning_rate": 3.4582132564841505e-05,
"loss": 0.3161,
"step": 480
},
{
"epoch": 0.7,
"grad_norm": 5.531047344207764,
"learning_rate": 3.494236311239193e-05,
"loss": 0.3465,
"step": 485
},
{
"epoch": 0.71,
"grad_norm": 5.138969421386719,
"learning_rate": 3.530259365994236e-05,
"loss": 0.3395,
"step": 490
},
{
"epoch": 0.71,
"grad_norm": 12.942193984985352,
"learning_rate": 3.5662824207492795e-05,
"loss": 0.3457,
"step": 495
},
{
"epoch": 0.72,
"grad_norm": 5.338680744171143,
"learning_rate": 3.602305475504323e-05,
"loss": 0.3203,
"step": 500
},
{
"epoch": 0.73,
"grad_norm": 6.598853588104248,
"learning_rate": 3.6383285302593664e-05,
"loss": 0.3691,
"step": 505
},
{
"epoch": 0.73,
"grad_norm": 6.757806777954102,
"learning_rate": 3.674351585014409e-05,
"loss": 0.354,
"step": 510
},
{
"epoch": 0.74,
"grad_norm": 6.43842077255249,
"learning_rate": 3.7103746397694526e-05,
"loss": 0.3492,
"step": 515
},
{
"epoch": 0.75,
"grad_norm": 6.269893169403076,
"learning_rate": 3.746397694524496e-05,
"loss": 0.3861,
"step": 520
},
{
"epoch": 0.76,
"grad_norm": 6.979896545410156,
"learning_rate": 3.782420749279539e-05,
"loss": 0.3423,
"step": 525
},
{
"epoch": 0.76,
"grad_norm": 7.62592077255249,
"learning_rate": 3.818443804034582e-05,
"loss": 0.413,
"step": 530
},
{
"epoch": 0.77,
"grad_norm": 6.566605567932129,
"learning_rate": 3.854466858789626e-05,
"loss": 0.414,
"step": 535
},
{
"epoch": 0.78,
"grad_norm": 5.802229881286621,
"learning_rate": 3.8904899135446685e-05,
"loss": 0.336,
"step": 540
},
{
"epoch": 0.78,
"grad_norm": 6.355484962463379,
"learning_rate": 3.926512968299712e-05,
"loss": 0.3425,
"step": 545
},
{
"epoch": 0.79,
"grad_norm": 6.845495700836182,
"learning_rate": 3.9625360230547554e-05,
"loss": 0.3603,
"step": 550
},
{
"epoch": 0.8,
"grad_norm": 5.654047966003418,
"learning_rate": 3.998559077809799e-05,
"loss": 0.3408,
"step": 555
},
{
"epoch": 0.81,
"grad_norm": 6.1373467445373535,
"learning_rate": 4.034582132564842e-05,
"loss": 0.3588,
"step": 560
},
{
"epoch": 0.81,
"grad_norm": 7.257894039154053,
"learning_rate": 4.0706051873198844e-05,
"loss": 0.3414,
"step": 565
},
{
"epoch": 0.82,
"grad_norm": 9.20009994506836,
"learning_rate": 4.106628242074928e-05,
"loss": 0.3749,
"step": 570
},
{
"epoch": 0.83,
"grad_norm": 8.215997695922852,
"learning_rate": 4.1426512968299713e-05,
"loss": 0.3673,
"step": 575
},
{
"epoch": 0.83,
"grad_norm": 8.917034149169922,
"learning_rate": 4.178674351585015e-05,
"loss": 0.3804,
"step": 580
},
{
"epoch": 0.84,
"grad_norm": 4.663518905639648,
"learning_rate": 4.214697406340058e-05,
"loss": 0.3454,
"step": 585
},
{
"epoch": 0.85,
"grad_norm": 5.577089786529541,
"learning_rate": 4.250720461095101e-05,
"loss": 0.3176,
"step": 590
},
{
"epoch": 0.86,
"grad_norm": 8.387089729309082,
"learning_rate": 4.2867435158501445e-05,
"loss": 0.3112,
"step": 595
},
{
"epoch": 0.86,
"grad_norm": 5.613365650177002,
"learning_rate": 4.322766570605187e-05,
"loss": 0.3491,
"step": 600
},
{
"epoch": 0.87,
"grad_norm": 6.961580753326416,
"learning_rate": 4.358789625360231e-05,
"loss": 0.3005,
"step": 605
},
{
"epoch": 0.88,
"grad_norm": 5.606225967407227,
"learning_rate": 4.394812680115274e-05,
"loss": 0.2919,
"step": 610
},
{
"epoch": 0.89,
"grad_norm": 5.59161376953125,
"learning_rate": 4.430835734870317e-05,
"loss": 0.3206,
"step": 615
},
{
"epoch": 0.89,
"grad_norm": 5.032812595367432,
"learning_rate": 4.4668587896253604e-05,
"loss": 0.3447,
"step": 620
},
{
"epoch": 0.9,
"grad_norm": 6.477130889892578,
"learning_rate": 4.502881844380404e-05,
"loss": 0.2768,
"step": 625
},
{
"epoch": 0.91,
"grad_norm": 6.052380084991455,
"learning_rate": 4.538904899135447e-05,
"loss": 0.3161,
"step": 630
},
{
"epoch": 0.91,
"grad_norm": 6.129565238952637,
"learning_rate": 4.57492795389049e-05,
"loss": 0.2634,
"step": 635
},
{
"epoch": 0.92,
"grad_norm": 7.382906436920166,
"learning_rate": 4.6109510086455335e-05,
"loss": 0.3402,
"step": 640
},
{
"epoch": 0.93,
"grad_norm": 6.6154375076293945,
"learning_rate": 4.646974063400576e-05,
"loss": 0.2461,
"step": 645
},
{
"epoch": 0.94,
"grad_norm": 4.32603645324707,
"learning_rate": 4.68299711815562e-05,
"loss": 0.2847,
"step": 650
},
{
"epoch": 0.94,
"grad_norm": 5.073220252990723,
"learning_rate": 4.719020172910663e-05,
"loss": 0.3306,
"step": 655
},
{
"epoch": 0.95,
"grad_norm": 6.942915916442871,
"learning_rate": 4.7550432276657067e-05,
"loss": 0.3361,
"step": 660
},
{
"epoch": 0.96,
"grad_norm": 5.637122631072998,
"learning_rate": 4.7910662824207494e-05,
"loss": 0.3489,
"step": 665
},
{
"epoch": 0.96,
"grad_norm": 5.205631256103516,
"learning_rate": 4.827089337175792e-05,
"loss": 0.3484,
"step": 670
},
{
"epoch": 0.97,
"grad_norm": 8.950785636901855,
"learning_rate": 4.8631123919308357e-05,
"loss": 0.3296,
"step": 675
},
{
"epoch": 0.98,
"grad_norm": 5.128692150115967,
"learning_rate": 4.899135446685879e-05,
"loss": 0.3583,
"step": 680
},
{
"epoch": 0.99,
"grad_norm": 3.9243364334106445,
"learning_rate": 4.9351585014409226e-05,
"loss": 0.346,
"step": 685
},
{
"epoch": 0.99,
"grad_norm": 6.407800197601318,
"learning_rate": 4.971181556195966e-05,
"loss": 0.3795,
"step": 690
},
{
"epoch": 1.0,
"eval_accuracy": 0.932591093117409,
"eval_loss": 0.19223400950431824,
"eval_runtime": 27.08,
"eval_samples_per_second": 364.844,
"eval_steps_per_second": 11.411,
"step": 694
},
{
"epoch": 1.0,
"grad_norm": 4.62927770614624,
"learning_rate": 4.99919948767211e-05,
"loss": 0.3346,
"step": 695
},
{
"epoch": 1.01,
"grad_norm": 4.929562568664551,
"learning_rate": 4.995196926032661e-05,
"loss": 0.3193,
"step": 700
},
{
"epoch": 1.01,
"grad_norm": 5.248744964599609,
"learning_rate": 4.9911943643932116e-05,
"loss": 0.3038,
"step": 705
},
{
"epoch": 1.02,
"grad_norm": 5.355862140655518,
"learning_rate": 4.9871918027537626e-05,
"loss": 0.3633,
"step": 710
},
{
"epoch": 1.03,
"grad_norm": 6.854907989501953,
"learning_rate": 4.9831892411143136e-05,
"loss": 0.3343,
"step": 715
},
{
"epoch": 1.04,
"grad_norm": 4.099529266357422,
"learning_rate": 4.979186679474864e-05,
"loss": 0.3242,
"step": 720
},
{
"epoch": 1.04,
"grad_norm": 4.553044319152832,
"learning_rate": 4.975184117835415e-05,
"loss": 0.3036,
"step": 725
},
{
"epoch": 1.05,
"grad_norm": 5.2062602043151855,
"learning_rate": 4.971181556195966e-05,
"loss": 0.3119,
"step": 730
},
{
"epoch": 1.06,
"grad_norm": 6.551631450653076,
"learning_rate": 4.9671789945565164e-05,
"loss": 0.3169,
"step": 735
},
{
"epoch": 1.07,
"grad_norm": 4.4297637939453125,
"learning_rate": 4.9631764329170674e-05,
"loss": 0.3358,
"step": 740
},
{
"epoch": 1.07,
"grad_norm": 4.483184814453125,
"learning_rate": 4.9591738712776184e-05,
"loss": 0.2771,
"step": 745
},
{
"epoch": 1.08,
"grad_norm": 4.935523986816406,
"learning_rate": 4.955171309638169e-05,
"loss": 0.2786,
"step": 750
},
{
"epoch": 1.09,
"grad_norm": 8.67397403717041,
"learning_rate": 4.95116874799872e-05,
"loss": 0.302,
"step": 755
},
{
"epoch": 1.09,
"grad_norm": 5.298563480377197,
"learning_rate": 4.94716618635927e-05,
"loss": 0.317,
"step": 760
},
{
"epoch": 1.1,
"grad_norm": 4.321622371673584,
"learning_rate": 4.943163624719821e-05,
"loss": 0.3213,
"step": 765
},
{
"epoch": 1.11,
"grad_norm": 5.290216445922852,
"learning_rate": 4.9391610630803715e-05,
"loss": 0.3334,
"step": 770
},
{
"epoch": 1.12,
"grad_norm": 5.450775146484375,
"learning_rate": 4.9351585014409226e-05,
"loss": 0.319,
"step": 775
},
{
"epoch": 1.12,
"grad_norm": 5.571349620819092,
"learning_rate": 4.9311559398014736e-05,
"loss": 0.27,
"step": 780
},
{
"epoch": 1.13,
"grad_norm": 5.267562389373779,
"learning_rate": 4.927153378162024e-05,
"loss": 0.3218,
"step": 785
},
{
"epoch": 1.14,
"grad_norm": 3.8232898712158203,
"learning_rate": 4.923150816522575e-05,
"loss": 0.2955,
"step": 790
},
{
"epoch": 1.14,
"grad_norm": 5.099688529968262,
"learning_rate": 4.919148254883125e-05,
"loss": 0.3217,
"step": 795
},
{
"epoch": 1.15,
"grad_norm": 3.9266700744628906,
"learning_rate": 4.9151456932436764e-05,
"loss": 0.3177,
"step": 800
},
{
"epoch": 1.16,
"grad_norm": 5.140632629394531,
"learning_rate": 4.911143131604227e-05,
"loss": 0.3068,
"step": 805
},
{
"epoch": 1.17,
"grad_norm": 4.642924785614014,
"learning_rate": 4.907140569964778e-05,
"loss": 0.2502,
"step": 810
},
{
"epoch": 1.17,
"grad_norm": 3.127683401107788,
"learning_rate": 4.903138008325329e-05,
"loss": 0.3102,
"step": 815
},
{
"epoch": 1.18,
"grad_norm": 5.013187408447266,
"learning_rate": 4.899135446685879e-05,
"loss": 0.312,
"step": 820
},
{
"epoch": 1.19,
"grad_norm": 5.893740177154541,
"learning_rate": 4.89513288504643e-05,
"loss": 0.3123,
"step": 825
},
{
"epoch": 1.19,
"grad_norm": 4.198624610900879,
"learning_rate": 4.8911303234069805e-05,
"loss": 0.2843,
"step": 830
},
{
"epoch": 1.2,
"grad_norm": 5.550350666046143,
"learning_rate": 4.8871277617675315e-05,
"loss": 0.3515,
"step": 835
},
{
"epoch": 1.21,
"grad_norm": 4.0659966468811035,
"learning_rate": 4.883125200128082e-05,
"loss": 0.2849,
"step": 840
},
{
"epoch": 1.22,
"grad_norm": 4.937682628631592,
"learning_rate": 4.879122638488633e-05,
"loss": 0.3533,
"step": 845
},
{
"epoch": 1.22,
"grad_norm": 4.22144079208374,
"learning_rate": 4.875120076849184e-05,
"loss": 0.3136,
"step": 850
},
{
"epoch": 1.23,
"grad_norm": 5.999542236328125,
"learning_rate": 4.871117515209734e-05,
"loss": 0.2703,
"step": 855
},
{
"epoch": 1.24,
"grad_norm": 6.5856828689575195,
"learning_rate": 4.867114953570285e-05,
"loss": 0.2894,
"step": 860
},
{
"epoch": 1.25,
"grad_norm": 3.6321206092834473,
"learning_rate": 4.8631123919308357e-05,
"loss": 0.2804,
"step": 865
},
{
"epoch": 1.25,
"grad_norm": 4.104198455810547,
"learning_rate": 4.859109830291387e-05,
"loss": 0.2767,
"step": 870
},
{
"epoch": 1.26,
"grad_norm": 5.859384059906006,
"learning_rate": 4.855107268651937e-05,
"loss": 0.2994,
"step": 875
},
{
"epoch": 1.27,
"grad_norm": 6.007699966430664,
"learning_rate": 4.851104707012488e-05,
"loss": 0.3492,
"step": 880
},
{
"epoch": 1.27,
"grad_norm": 4.246580123901367,
"learning_rate": 4.847102145373039e-05,
"loss": 0.2911,
"step": 885
},
{
"epoch": 1.28,
"grad_norm": 4.882290840148926,
"learning_rate": 4.8430995837335894e-05,
"loss": 0.3154,
"step": 890
},
{
"epoch": 1.29,
"grad_norm": 4.810091018676758,
"learning_rate": 4.8390970220941405e-05,
"loss": 0.2723,
"step": 895
},
{
"epoch": 1.3,
"grad_norm": 5.89586877822876,
"learning_rate": 4.835094460454691e-05,
"loss": 0.3139,
"step": 900
},
{
"epoch": 1.3,
"grad_norm": 4.357601642608643,
"learning_rate": 4.831091898815242e-05,
"loss": 0.2872,
"step": 905
},
{
"epoch": 1.31,
"grad_norm": 4.09877347946167,
"learning_rate": 4.827089337175792e-05,
"loss": 0.2555,
"step": 910
},
{
"epoch": 1.32,
"grad_norm": 4.7353925704956055,
"learning_rate": 4.823086775536343e-05,
"loss": 0.35,
"step": 915
},
{
"epoch": 1.32,
"grad_norm": 5.126601219177246,
"learning_rate": 4.819084213896894e-05,
"loss": 0.2799,
"step": 920
},
{
"epoch": 1.33,
"grad_norm": 5.320415496826172,
"learning_rate": 4.8150816522574446e-05,
"loss": 0.2905,
"step": 925
},
{
"epoch": 1.34,
"grad_norm": 5.826961040496826,
"learning_rate": 4.8110790906179956e-05,
"loss": 0.2916,
"step": 930
},
{
"epoch": 1.35,
"grad_norm": 4.833921432495117,
"learning_rate": 4.807076528978546e-05,
"loss": 0.324,
"step": 935
},
{
"epoch": 1.35,
"grad_norm": 5.386748313903809,
"learning_rate": 4.803073967339097e-05,
"loss": 0.3005,
"step": 940
},
{
"epoch": 1.36,
"grad_norm": 4.9491424560546875,
"learning_rate": 4.799071405699648e-05,
"loss": 0.2901,
"step": 945
},
{
"epoch": 1.37,
"grad_norm": 4.132954120635986,
"learning_rate": 4.7950688440601984e-05,
"loss": 0.278,
"step": 950
},
{
"epoch": 1.37,
"grad_norm": 5.6594319343566895,
"learning_rate": 4.7910662824207494e-05,
"loss": 0.3134,
"step": 955
},
{
"epoch": 1.38,
"grad_norm": 5.512797832489014,
"learning_rate": 4.7870637207813005e-05,
"loss": 0.2781,
"step": 960
},
{
"epoch": 1.39,
"grad_norm": 5.356082916259766,
"learning_rate": 4.7830611591418515e-05,
"loss": 0.2965,
"step": 965
},
{
"epoch": 1.4,
"grad_norm": 4.4097113609313965,
"learning_rate": 4.779058597502402e-05,
"loss": 0.2756,
"step": 970
},
{
"epoch": 1.4,
"grad_norm": 5.423006534576416,
"learning_rate": 4.775056035862953e-05,
"loss": 0.3404,
"step": 975
},
{
"epoch": 1.41,
"grad_norm": 4.108374118804932,
"learning_rate": 4.771053474223503e-05,
"loss": 0.2822,
"step": 980
},
{
"epoch": 1.42,
"grad_norm": 5.638019561767578,
"learning_rate": 4.767050912584054e-05,
"loss": 0.2729,
"step": 985
},
{
"epoch": 1.42,
"grad_norm": 4.037389278411865,
"learning_rate": 4.763048350944605e-05,
"loss": 0.2832,
"step": 990
},
{
"epoch": 1.43,
"grad_norm": 5.815566539764404,
"learning_rate": 4.7590457893051556e-05,
"loss": 0.2859,
"step": 995
},
{
"epoch": 1.44,
"grad_norm": 4.1045732498168945,
"learning_rate": 4.7550432276657067e-05,
"loss": 0.2734,
"step": 1000
},
{
"epoch": 1.45,
"grad_norm": 7.296690464019775,
"learning_rate": 4.751040666026257e-05,
"loss": 0.3211,
"step": 1005
},
{
"epoch": 1.45,
"grad_norm": 8.276504516601562,
"learning_rate": 4.747038104386808e-05,
"loss": 0.2964,
"step": 1010
},
{
"epoch": 1.46,
"grad_norm": 5.901710510253906,
"learning_rate": 4.7430355427473584e-05,
"loss": 0.318,
"step": 1015
},
{
"epoch": 1.47,
"grad_norm": 5.140340328216553,
"learning_rate": 4.7390329811079094e-05,
"loss": 0.3337,
"step": 1020
},
{
"epoch": 1.48,
"grad_norm": 5.367193222045898,
"learning_rate": 4.7350304194684604e-05,
"loss": 0.2844,
"step": 1025
},
{
"epoch": 1.48,
"grad_norm": 5.234673500061035,
"learning_rate": 4.731027857829011e-05,
"loss": 0.2676,
"step": 1030
},
{
"epoch": 1.49,
"grad_norm": 4.406520366668701,
"learning_rate": 4.727025296189562e-05,
"loss": 0.2394,
"step": 1035
},
{
"epoch": 1.5,
"grad_norm": 5.445221900939941,
"learning_rate": 4.723022734550112e-05,
"loss": 0.277,
"step": 1040
},
{
"epoch": 1.5,
"grad_norm": 5.237115383148193,
"learning_rate": 4.719020172910663e-05,
"loss": 0.3117,
"step": 1045
},
{
"epoch": 1.51,
"grad_norm": 4.234028339385986,
"learning_rate": 4.7150176112712136e-05,
"loss": 0.2931,
"step": 1050
},
{
"epoch": 1.52,
"grad_norm": 4.2522430419921875,
"learning_rate": 4.7110150496317646e-05,
"loss": 0.3156,
"step": 1055
},
{
"epoch": 1.53,
"grad_norm": 6.036770820617676,
"learning_rate": 4.7070124879923156e-05,
"loss": 0.2222,
"step": 1060
},
{
"epoch": 1.53,
"grad_norm": 5.12794303894043,
"learning_rate": 4.703009926352866e-05,
"loss": 0.3077,
"step": 1065
},
{
"epoch": 1.54,
"grad_norm": 4.9581756591796875,
"learning_rate": 4.699007364713417e-05,
"loss": 0.2537,
"step": 1070
},
{
"epoch": 1.55,
"grad_norm": 4.344903945922852,
"learning_rate": 4.6950048030739673e-05,
"loss": 0.3009,
"step": 1075
},
{
"epoch": 1.55,
"grad_norm": 3.9201972484588623,
"learning_rate": 4.6910022414345184e-05,
"loss": 0.2605,
"step": 1080
},
{
"epoch": 1.56,
"grad_norm": 4.709476947784424,
"learning_rate": 4.686999679795069e-05,
"loss": 0.3037,
"step": 1085
},
{
"epoch": 1.57,
"grad_norm": 5.150713920593262,
"learning_rate": 4.68299711815562e-05,
"loss": 0.258,
"step": 1090
},
{
"epoch": 1.58,
"grad_norm": 2.914226531982422,
"learning_rate": 4.678994556516171e-05,
"loss": 0.3184,
"step": 1095
},
{
"epoch": 1.58,
"grad_norm": 3.4970335960388184,
"learning_rate": 4.674991994876721e-05,
"loss": 0.2623,
"step": 1100
},
{
"epoch": 1.59,
"grad_norm": 5.648881912231445,
"learning_rate": 4.670989433237272e-05,
"loss": 0.2897,
"step": 1105
},
{
"epoch": 1.6,
"grad_norm": 5.417716026306152,
"learning_rate": 4.6669868715978225e-05,
"loss": 0.2765,
"step": 1110
},
{
"epoch": 1.6,
"grad_norm": 5.414010524749756,
"learning_rate": 4.6629843099583735e-05,
"loss": 0.2596,
"step": 1115
},
{
"epoch": 1.61,
"grad_norm": 6.044466972351074,
"learning_rate": 4.658981748318924e-05,
"loss": 0.2846,
"step": 1120
},
{
"epoch": 1.62,
"grad_norm": 4.162852764129639,
"learning_rate": 4.654979186679475e-05,
"loss": 0.2756,
"step": 1125
},
{
"epoch": 1.63,
"grad_norm": 4.842267036437988,
"learning_rate": 4.650976625040026e-05,
"loss": 0.2836,
"step": 1130
},
{
"epoch": 1.63,
"grad_norm": 5.485855579376221,
"learning_rate": 4.646974063400576e-05,
"loss": 0.3205,
"step": 1135
},
{
"epoch": 1.64,
"grad_norm": 3.7354228496551514,
"learning_rate": 4.642971501761127e-05,
"loss": 0.2127,
"step": 1140
},
{
"epoch": 1.65,
"grad_norm": 5.490901470184326,
"learning_rate": 4.638968940121678e-05,
"loss": 0.3052,
"step": 1145
},
{
"epoch": 1.66,
"grad_norm": 5.564297676086426,
"learning_rate": 4.634966378482229e-05,
"loss": 0.2564,
"step": 1150
},
{
"epoch": 1.66,
"grad_norm": 8.057859420776367,
"learning_rate": 4.630963816842779e-05,
"loss": 0.2581,
"step": 1155
},
{
"epoch": 1.67,
"grad_norm": 7.346789836883545,
"learning_rate": 4.62696125520333e-05,
"loss": 0.2802,
"step": 1160
},
{
"epoch": 1.68,
"grad_norm": 6.667150020599365,
"learning_rate": 4.622958693563881e-05,
"loss": 0.2845,
"step": 1165
},
{
"epoch": 1.68,
"grad_norm": 5.178923606872559,
"learning_rate": 4.6189561319244315e-05,
"loss": 0.2948,
"step": 1170
},
{
"epoch": 1.69,
"grad_norm": 4.015057563781738,
"learning_rate": 4.6149535702849825e-05,
"loss": 0.3022,
"step": 1175
},
{
"epoch": 1.7,
"grad_norm": 3.952420234680176,
"learning_rate": 4.6109510086455335e-05,
"loss": 0.2521,
"step": 1180
},
{
"epoch": 1.71,
"grad_norm": 4.373567581176758,
"learning_rate": 4.606948447006084e-05,
"loss": 0.2505,
"step": 1185
},
{
"epoch": 1.71,
"grad_norm": 3.232614040374756,
"learning_rate": 4.602945885366635e-05,
"loss": 0.2713,
"step": 1190
},
{
"epoch": 1.72,
"grad_norm": 3.8914895057678223,
"learning_rate": 4.598943323727186e-05,
"loss": 0.2547,
"step": 1195
},
{
"epoch": 1.73,
"grad_norm": 5.603899002075195,
"learning_rate": 4.594940762087736e-05,
"loss": 0.282,
"step": 1200
},
{
"epoch": 1.73,
"grad_norm": 3.351109743118286,
"learning_rate": 4.590938200448287e-05,
"loss": 0.2214,
"step": 1205
},
{
"epoch": 1.74,
"grad_norm": 5.184418201446533,
"learning_rate": 4.586935638808838e-05,
"loss": 0.298,
"step": 1210
},
{
"epoch": 1.75,
"grad_norm": 5.310287952423096,
"learning_rate": 4.582933077169389e-05,
"loss": 0.2825,
"step": 1215
},
{
"epoch": 1.76,
"grad_norm": 5.883835792541504,
"learning_rate": 4.57893051552994e-05,
"loss": 0.3257,
"step": 1220
},
{
"epoch": 1.76,
"grad_norm": 3.4902472496032715,
"learning_rate": 4.57492795389049e-05,
"loss": 0.2691,
"step": 1225
},
{
"epoch": 1.77,
"grad_norm": 3.718014717102051,
"learning_rate": 4.570925392251041e-05,
"loss": 0.2918,
"step": 1230
},
{
"epoch": 1.78,
"grad_norm": 5.416603088378906,
"learning_rate": 4.566922830611592e-05,
"loss": 0.2776,
"step": 1235
},
{
"epoch": 1.78,
"grad_norm": 5.364027500152588,
"learning_rate": 4.5629202689721425e-05,
"loss": 0.2804,
"step": 1240
},
{
"epoch": 1.79,
"grad_norm": 5.734582424163818,
"learning_rate": 4.5589177073326935e-05,
"loss": 0.265,
"step": 1245
},
{
"epoch": 1.8,
"grad_norm": 3.5217783451080322,
"learning_rate": 4.554915145693244e-05,
"loss": 0.2499,
"step": 1250
},
{
"epoch": 1.81,
"grad_norm": 4.218166351318359,
"learning_rate": 4.550912584053795e-05,
"loss": 0.292,
"step": 1255
},
{
"epoch": 1.81,
"grad_norm": 4.511210918426514,
"learning_rate": 4.546910022414345e-05,
"loss": 0.2325,
"step": 1260
},
{
"epoch": 1.82,
"grad_norm": 4.017195224761963,
"learning_rate": 4.542907460774896e-05,
"loss": 0.2584,
"step": 1265
},
{
"epoch": 1.83,
"grad_norm": 3.4999618530273438,
"learning_rate": 4.538904899135447e-05,
"loss": 0.3007,
"step": 1270
},
{
"epoch": 1.84,
"grad_norm": 5.2072248458862305,
"learning_rate": 4.5349023374959976e-05,
"loss": 0.2659,
"step": 1275
},
{
"epoch": 1.84,
"grad_norm": 4.940614223480225,
"learning_rate": 4.530899775856549e-05,
"loss": 0.282,
"step": 1280
},
{
"epoch": 1.85,
"grad_norm": 3.722916603088379,
"learning_rate": 4.526897214217099e-05,
"loss": 0.2555,
"step": 1285
},
{
"epoch": 1.86,
"grad_norm": 4.483854293823242,
"learning_rate": 4.52289465257765e-05,
"loss": 0.2594,
"step": 1290
},
{
"epoch": 1.86,
"grad_norm": 4.568425178527832,
"learning_rate": 4.5188920909382004e-05,
"loss": 0.262,
"step": 1295
},
{
"epoch": 1.87,
"grad_norm": 4.76126766204834,
"learning_rate": 4.5148895292987514e-05,
"loss": 0.2645,
"step": 1300
},
{
"epoch": 1.88,
"grad_norm": 4.418682098388672,
"learning_rate": 4.5108869676593025e-05,
"loss": 0.268,
"step": 1305
},
{
"epoch": 1.89,
"grad_norm": 3.255800485610962,
"learning_rate": 4.506884406019853e-05,
"loss": 0.2665,
"step": 1310
},
{
"epoch": 1.89,
"grad_norm": 4.66806697845459,
"learning_rate": 4.502881844380404e-05,
"loss": 0.2965,
"step": 1315
},
{
"epoch": 1.9,
"grad_norm": 5.117862701416016,
"learning_rate": 4.498879282740954e-05,
"loss": 0.2423,
"step": 1320
},
{
"epoch": 1.91,
"grad_norm": 3.500521421432495,
"learning_rate": 4.494876721101505e-05,
"loss": 0.2682,
"step": 1325
},
{
"epoch": 1.91,
"grad_norm": 3.541145086288452,
"learning_rate": 4.4908741594620556e-05,
"loss": 0.2676,
"step": 1330
},
{
"epoch": 1.92,
"grad_norm": 5.114723205566406,
"learning_rate": 4.4868715978226066e-05,
"loss": 0.214,
"step": 1335
},
{
"epoch": 1.93,
"grad_norm": 3.2529759407043457,
"learning_rate": 4.4828690361831576e-05,
"loss": 0.2757,
"step": 1340
},
{
"epoch": 1.94,
"grad_norm": 4.737575531005859,
"learning_rate": 4.478866474543708e-05,
"loss": 0.2541,
"step": 1345
},
{
"epoch": 1.94,
"grad_norm": 3.7248144149780273,
"learning_rate": 4.474863912904259e-05,
"loss": 0.2837,
"step": 1350
},
{
"epoch": 1.95,
"grad_norm": 6.532453536987305,
"learning_rate": 4.4708613512648094e-05,
"loss": 0.2667,
"step": 1355
},
{
"epoch": 1.96,
"grad_norm": 4.162599563598633,
"learning_rate": 4.4668587896253604e-05,
"loss": 0.3142,
"step": 1360
},
{
"epoch": 1.96,
"grad_norm": 4.6284027099609375,
"learning_rate": 4.462856227985911e-05,
"loss": 0.2566,
"step": 1365
},
{
"epoch": 1.97,
"grad_norm": 3.6103484630584717,
"learning_rate": 4.458853666346462e-05,
"loss": 0.2723,
"step": 1370
},
{
"epoch": 1.98,
"grad_norm": 3.301093816757202,
"learning_rate": 4.454851104707013e-05,
"loss": 0.2213,
"step": 1375
},
{
"epoch": 1.99,
"grad_norm": 4.462352752685547,
"learning_rate": 4.450848543067563e-05,
"loss": 0.2655,
"step": 1380
},
{
"epoch": 1.99,
"grad_norm": 3.400181770324707,
"learning_rate": 4.446845981428114e-05,
"loss": 0.261,
"step": 1385
},
{
"epoch": 2.0,
"eval_accuracy": 0.933502024291498,
"eval_loss": 0.18496161699295044,
"eval_runtime": 26.9431,
"eval_samples_per_second": 366.698,
"eval_steps_per_second": 11.469,
"step": 1389
},
{
"epoch": 2.0,
"grad_norm": 4.32210111618042,
"learning_rate": 4.4428434197886645e-05,
"loss": 0.2947,
"step": 1390
},
{
"epoch": 2.01,
"grad_norm": 3.8557567596435547,
"learning_rate": 4.4388408581492156e-05,
"loss": 0.2373,
"step": 1395
},
{
"epoch": 2.02,
"grad_norm": 4.735769748687744,
"learning_rate": 4.434838296509766e-05,
"loss": 0.2476,
"step": 1400
},
{
"epoch": 2.02,
"grad_norm": 4.38801383972168,
"learning_rate": 4.430835734870317e-05,
"loss": 0.2279,
"step": 1405
},
{
"epoch": 2.03,
"grad_norm": 4.940765857696533,
"learning_rate": 4.426833173230868e-05,
"loss": 0.2529,
"step": 1410
},
{
"epoch": 2.04,
"grad_norm": 4.5427045822143555,
"learning_rate": 4.422830611591418e-05,
"loss": 0.2499,
"step": 1415
},
{
"epoch": 2.04,
"grad_norm": 4.766185760498047,
"learning_rate": 4.4188280499519693e-05,
"loss": 0.2052,
"step": 1420
},
{
"epoch": 2.05,
"grad_norm": 4.980799674987793,
"learning_rate": 4.4148254883125204e-05,
"loss": 0.2676,
"step": 1425
},
{
"epoch": 2.06,
"grad_norm": 3.4634597301483154,
"learning_rate": 4.4108229266730714e-05,
"loss": 0.2417,
"step": 1430
},
{
"epoch": 2.07,
"grad_norm": 3.6540205478668213,
"learning_rate": 4.406820365033622e-05,
"loss": 0.2876,
"step": 1435
},
{
"epoch": 2.07,
"grad_norm": 3.4874844551086426,
"learning_rate": 4.402817803394173e-05,
"loss": 0.2635,
"step": 1440
},
{
"epoch": 2.08,
"grad_norm": 4.6576457023620605,
"learning_rate": 4.398815241754724e-05,
"loss": 0.3051,
"step": 1445
},
{
"epoch": 2.09,
"grad_norm": 4.110597610473633,
"learning_rate": 4.394812680115274e-05,
"loss": 0.2595,
"step": 1450
},
{
"epoch": 2.09,
"grad_norm": 4.409259796142578,
"learning_rate": 4.390810118475825e-05,
"loss": 0.2099,
"step": 1455
},
{
"epoch": 2.1,
"grad_norm": 3.2214550971984863,
"learning_rate": 4.3868075568363755e-05,
"loss": 0.2627,
"step": 1460
},
{
"epoch": 2.11,
"grad_norm": 4.47964334487915,
"learning_rate": 4.3828049951969266e-05,
"loss": 0.3545,
"step": 1465
},
{
"epoch": 2.12,
"grad_norm": 4.1163105964660645,
"learning_rate": 4.378802433557477e-05,
"loss": 0.2459,
"step": 1470
},
{
"epoch": 2.12,
"grad_norm": 3.891965866088867,
"learning_rate": 4.374799871918028e-05,
"loss": 0.2431,
"step": 1475
},
{
"epoch": 2.13,
"grad_norm": 3.4245457649230957,
"learning_rate": 4.370797310278579e-05,
"loss": 0.2511,
"step": 1480
},
{
"epoch": 2.14,
"grad_norm": 6.8035807609558105,
"learning_rate": 4.366794748639129e-05,
"loss": 0.2619,
"step": 1485
},
{
"epoch": 2.14,
"grad_norm": 5.102532386779785,
"learning_rate": 4.3627921869996804e-05,
"loss": 0.2707,
"step": 1490
},
{
"epoch": 2.15,
"grad_norm": 3.1594481468200684,
"learning_rate": 4.358789625360231e-05,
"loss": 0.2273,
"step": 1495
},
{
"epoch": 2.16,
"grad_norm": 3.9931840896606445,
"learning_rate": 4.354787063720782e-05,
"loss": 0.2537,
"step": 1500
},
{
"epoch": 2.17,
"grad_norm": 5.2231221199035645,
"learning_rate": 4.350784502081332e-05,
"loss": 0.2431,
"step": 1505
},
{
"epoch": 2.17,
"grad_norm": 4.415401458740234,
"learning_rate": 4.346781940441883e-05,
"loss": 0.2431,
"step": 1510
},
{
"epoch": 2.18,
"grad_norm": 4.261172294616699,
"learning_rate": 4.342779378802434e-05,
"loss": 0.2779,
"step": 1515
},
{
"epoch": 2.19,
"grad_norm": 3.6781246662139893,
"learning_rate": 4.3387768171629845e-05,
"loss": 0.2052,
"step": 1520
},
{
"epoch": 2.2,
"grad_norm": 4.412373065948486,
"learning_rate": 4.3347742555235355e-05,
"loss": 0.2797,
"step": 1525
},
{
"epoch": 2.2,
"grad_norm": 5.768673419952393,
"learning_rate": 4.330771693884086e-05,
"loss": 0.2637,
"step": 1530
},
{
"epoch": 2.21,
"grad_norm": 7.167098045349121,
"learning_rate": 4.326769132244637e-05,
"loss": 0.2486,
"step": 1535
},
{
"epoch": 2.22,
"grad_norm": 3.754503011703491,
"learning_rate": 4.322766570605187e-05,
"loss": 0.2315,
"step": 1540
},
{
"epoch": 2.22,
"grad_norm": 3.3107941150665283,
"learning_rate": 4.318764008965738e-05,
"loss": 0.2175,
"step": 1545
},
{
"epoch": 2.23,
"grad_norm": 3.4948480129241943,
"learning_rate": 4.314761447326289e-05,
"loss": 0.2228,
"step": 1550
},
{
"epoch": 2.24,
"grad_norm": 3.740145444869995,
"learning_rate": 4.31075888568684e-05,
"loss": 0.2475,
"step": 1555
},
{
"epoch": 2.25,
"grad_norm": 5.239324569702148,
"learning_rate": 4.306756324047391e-05,
"loss": 0.2427,
"step": 1560
},
{
"epoch": 2.25,
"grad_norm": 5.795839309692383,
"learning_rate": 4.302753762407941e-05,
"loss": 0.2177,
"step": 1565
},
{
"epoch": 2.26,
"grad_norm": 4.080209255218506,
"learning_rate": 4.298751200768492e-05,
"loss": 0.236,
"step": 1570
},
{
"epoch": 2.27,
"grad_norm": 6.359370708465576,
"learning_rate": 4.2947486391290424e-05,
"loss": 0.2796,
"step": 1575
},
{
"epoch": 2.27,
"grad_norm": 4.7210235595703125,
"learning_rate": 4.2907460774895934e-05,
"loss": 0.2409,
"step": 1580
},
{
"epoch": 2.28,
"grad_norm": 4.770945072174072,
"learning_rate": 4.2867435158501445e-05,
"loss": 0.2774,
"step": 1585
},
{
"epoch": 2.29,
"grad_norm": 5.309744834899902,
"learning_rate": 4.282740954210695e-05,
"loss": 0.2317,
"step": 1590
},
{
"epoch": 2.3,
"grad_norm": 4.85918664932251,
"learning_rate": 4.278738392571246e-05,
"loss": 0.2666,
"step": 1595
},
{
"epoch": 2.3,
"grad_norm": 3.067629337310791,
"learning_rate": 4.274735830931796e-05,
"loss": 0.2635,
"step": 1600
},
{
"epoch": 2.31,
"grad_norm": 3.3474929332733154,
"learning_rate": 4.270733269292347e-05,
"loss": 0.2271,
"step": 1605
},
{
"epoch": 2.32,
"grad_norm": 3.961737871170044,
"learning_rate": 4.2667307076528976e-05,
"loss": 0.2403,
"step": 1610
},
{
"epoch": 2.32,
"grad_norm": 2.509068489074707,
"learning_rate": 4.2627281460134486e-05,
"loss": 0.2283,
"step": 1615
},
{
"epoch": 2.33,
"grad_norm": 5.405132293701172,
"learning_rate": 4.2587255843739996e-05,
"loss": 0.1761,
"step": 1620
},
{
"epoch": 2.34,
"grad_norm": 4.707294940948486,
"learning_rate": 4.25472302273455e-05,
"loss": 0.2561,
"step": 1625
},
{
"epoch": 2.35,
"grad_norm": 4.00047492980957,
"learning_rate": 4.250720461095101e-05,
"loss": 0.2367,
"step": 1630
},
{
"epoch": 2.35,
"grad_norm": 5.079905986785889,
"learning_rate": 4.2467178994556514e-05,
"loss": 0.2878,
"step": 1635
},
{
"epoch": 2.36,
"grad_norm": 5.759792804718018,
"learning_rate": 4.2427153378162024e-05,
"loss": 0.2989,
"step": 1640
},
{
"epoch": 2.37,
"grad_norm": 2.7700154781341553,
"learning_rate": 4.2387127761767534e-05,
"loss": 0.2531,
"step": 1645
},
{
"epoch": 2.37,
"grad_norm": 3.7735087871551514,
"learning_rate": 4.234710214537304e-05,
"loss": 0.2304,
"step": 1650
},
{
"epoch": 2.38,
"grad_norm": 4.496865749359131,
"learning_rate": 4.230707652897855e-05,
"loss": 0.263,
"step": 1655
},
{
"epoch": 2.39,
"grad_norm": 4.752154350280762,
"learning_rate": 4.226705091258406e-05,
"loss": 0.2565,
"step": 1660
},
{
"epoch": 2.4,
"grad_norm": 3.979544162750244,
"learning_rate": 4.222702529618956e-05,
"loss": 0.2166,
"step": 1665
},
{
"epoch": 2.4,
"grad_norm": 3.6765263080596924,
"learning_rate": 4.218699967979507e-05,
"loss": 0.2438,
"step": 1670
},
{
"epoch": 2.41,
"grad_norm": 4.9830322265625,
"learning_rate": 4.214697406340058e-05,
"loss": 0.2551,
"step": 1675
},
{
"epoch": 2.42,
"grad_norm": 3.655550956726074,
"learning_rate": 4.2106948447006086e-05,
"loss": 0.2367,
"step": 1680
},
{
"epoch": 2.43,
"grad_norm": 7.952022075653076,
"learning_rate": 4.2066922830611596e-05,
"loss": 0.2355,
"step": 1685
},
{
"epoch": 2.43,
"grad_norm": 5.706752777099609,
"learning_rate": 4.2026897214217107e-05,
"loss": 0.2127,
"step": 1690
},
{
"epoch": 2.44,
"grad_norm": 5.1583099365234375,
"learning_rate": 4.198687159782261e-05,
"loss": 0.2496,
"step": 1695
},
{
"epoch": 2.45,
"grad_norm": 5.349088191986084,
"learning_rate": 4.194684598142812e-05,
"loss": 0.2593,
"step": 1700
},
{
"epoch": 2.45,
"grad_norm": 4.123900890350342,
"learning_rate": 4.1906820365033624e-05,
"loss": 0.2203,
"step": 1705
},
{
"epoch": 2.46,
"grad_norm": 6.052340030670166,
"learning_rate": 4.1866794748639134e-05,
"loss": 0.2613,
"step": 1710
},
{
"epoch": 2.47,
"grad_norm": 5.388720989227295,
"learning_rate": 4.182676913224464e-05,
"loss": 0.2277,
"step": 1715
},
{
"epoch": 2.48,
"grad_norm": 17.35828971862793,
"learning_rate": 4.178674351585015e-05,
"loss": 0.2445,
"step": 1720
},
{
"epoch": 2.48,
"grad_norm": 5.437941074371338,
"learning_rate": 4.174671789945566e-05,
"loss": 0.226,
"step": 1725
},
{
"epoch": 2.49,
"grad_norm": 4.627963066101074,
"learning_rate": 4.170669228306116e-05,
"loss": 0.2684,
"step": 1730
},
{
"epoch": 2.5,
"grad_norm": 2.598243474960327,
"learning_rate": 4.166666666666667e-05,
"loss": 0.2242,
"step": 1735
},
{
"epoch": 2.5,
"grad_norm": 3.801103353500366,
"learning_rate": 4.1626641050272176e-05,
"loss": 0.2603,
"step": 1740
},
{
"epoch": 2.51,
"grad_norm": 3.749072313308716,
"learning_rate": 4.1586615433877686e-05,
"loss": 0.2382,
"step": 1745
},
{
"epoch": 2.52,
"grad_norm": 5.224049091339111,
"learning_rate": 4.154658981748319e-05,
"loss": 0.2103,
"step": 1750
},
{
"epoch": 2.53,
"grad_norm": 6.435360431671143,
"learning_rate": 4.15065642010887e-05,
"loss": 0.244,
"step": 1755
},
{
"epoch": 2.53,
"grad_norm": 3.173849105834961,
"learning_rate": 4.146653858469421e-05,
"loss": 0.2413,
"step": 1760
},
{
"epoch": 2.54,
"grad_norm": 4.47773551940918,
"learning_rate": 4.1426512968299713e-05,
"loss": 0.2762,
"step": 1765
},
{
"epoch": 2.55,
"grad_norm": 3.3660101890563965,
"learning_rate": 4.1386487351905224e-05,
"loss": 0.1729,
"step": 1770
},
{
"epoch": 2.55,
"grad_norm": 3.466449737548828,
"learning_rate": 4.134646173551073e-05,
"loss": 0.2526,
"step": 1775
},
{
"epoch": 2.56,
"grad_norm": 4.319614887237549,
"learning_rate": 4.130643611911624e-05,
"loss": 0.2421,
"step": 1780
},
{
"epoch": 2.57,
"grad_norm": 3.959930181503296,
"learning_rate": 4.126641050272174e-05,
"loss": 0.252,
"step": 1785
},
{
"epoch": 2.58,
"grad_norm": 5.325150012969971,
"learning_rate": 4.122638488632725e-05,
"loss": 0.2186,
"step": 1790
},
{
"epoch": 2.58,
"grad_norm": 4.520905494689941,
"learning_rate": 4.118635926993276e-05,
"loss": 0.212,
"step": 1795
},
{
"epoch": 2.59,
"grad_norm": 4.795076847076416,
"learning_rate": 4.1146333653538265e-05,
"loss": 0.2496,
"step": 1800
},
{
"epoch": 2.6,
"grad_norm": 4.291173934936523,
"learning_rate": 4.1106308037143775e-05,
"loss": 0.2587,
"step": 1805
},
{
"epoch": 2.61,
"grad_norm": 4.245218753814697,
"learning_rate": 4.106628242074928e-05,
"loss": 0.2725,
"step": 1810
},
{
"epoch": 2.61,
"grad_norm": 3.0920708179473877,
"learning_rate": 4.102625680435479e-05,
"loss": 0.2218,
"step": 1815
},
{
"epoch": 2.62,
"grad_norm": 2.9690959453582764,
"learning_rate": 4.098623118796029e-05,
"loss": 0.251,
"step": 1820
},
{
"epoch": 2.63,
"grad_norm": 4.769535064697266,
"learning_rate": 4.09462055715658e-05,
"loss": 0.2301,
"step": 1825
},
{
"epoch": 2.63,
"grad_norm": 2.9699225425720215,
"learning_rate": 4.090617995517131e-05,
"loss": 0.1879,
"step": 1830
},
{
"epoch": 2.64,
"grad_norm": 5.330379962921143,
"learning_rate": 4.086615433877682e-05,
"loss": 0.2343,
"step": 1835
},
{
"epoch": 2.65,
"grad_norm": 5.215531826019287,
"learning_rate": 4.082612872238233e-05,
"loss": 0.2879,
"step": 1840
},
{
"epoch": 2.66,
"grad_norm": 5.113295555114746,
"learning_rate": 4.078610310598783e-05,
"loss": 0.2139,
"step": 1845
},
{
"epoch": 2.66,
"grad_norm": 3.597294807434082,
"learning_rate": 4.074607748959334e-05,
"loss": 0.2314,
"step": 1850
},
{
"epoch": 2.67,
"grad_norm": 4.293632507324219,
"learning_rate": 4.0706051873198844e-05,
"loss": 0.2201,
"step": 1855
},
{
"epoch": 2.68,
"grad_norm": 5.499499797821045,
"learning_rate": 4.0666026256804355e-05,
"loss": 0.2463,
"step": 1860
},
{
"epoch": 2.68,
"grad_norm": 4.88955020904541,
"learning_rate": 4.0626000640409865e-05,
"loss": 0.1911,
"step": 1865
},
{
"epoch": 2.69,
"grad_norm": 3.8911263942718506,
"learning_rate": 4.058597502401537e-05,
"loss": 0.2021,
"step": 1870
},
{
"epoch": 2.7,
"grad_norm": 4.9449920654296875,
"learning_rate": 4.054594940762088e-05,
"loss": 0.2542,
"step": 1875
},
{
"epoch": 2.71,
"grad_norm": 5.476675033569336,
"learning_rate": 4.050592379122638e-05,
"loss": 0.2391,
"step": 1880
},
{
"epoch": 2.71,
"grad_norm": 4.407501697540283,
"learning_rate": 4.046589817483189e-05,
"loss": 0.2695,
"step": 1885
},
{
"epoch": 2.72,
"grad_norm": 2.3330883979797363,
"learning_rate": 4.04258725584374e-05,
"loss": 0.2438,
"step": 1890
},
{
"epoch": 2.73,
"grad_norm": 5.288588523864746,
"learning_rate": 4.038584694204291e-05,
"loss": 0.2618,
"step": 1895
},
{
"epoch": 2.73,
"grad_norm": 5.602869510650635,
"learning_rate": 4.034582132564842e-05,
"loss": 0.2402,
"step": 1900
},
{
"epoch": 2.74,
"grad_norm": 5.500049591064453,
"learning_rate": 4.030579570925393e-05,
"loss": 0.294,
"step": 1905
},
{
"epoch": 2.75,
"grad_norm": 4.733332633972168,
"learning_rate": 4.026577009285944e-05,
"loss": 0.2065,
"step": 1910
},
{
"epoch": 2.76,
"grad_norm": 4.457444190979004,
"learning_rate": 4.022574447646494e-05,
"loss": 0.2276,
"step": 1915
},
{
"epoch": 2.76,
"grad_norm": 2.527444362640381,
"learning_rate": 4.018571886007045e-05,
"loss": 0.2209,
"step": 1920
},
{
"epoch": 2.77,
"grad_norm": 3.968388795852661,
"learning_rate": 4.0145693243675955e-05,
"loss": 0.2292,
"step": 1925
},
{
"epoch": 2.78,
"grad_norm": 3.083031415939331,
"learning_rate": 4.0105667627281465e-05,
"loss": 0.2072,
"step": 1930
},
{
"epoch": 2.79,
"grad_norm": 6.950841903686523,
"learning_rate": 4.0065642010886975e-05,
"loss": 0.2179,
"step": 1935
},
{
"epoch": 2.79,
"grad_norm": 4.184908390045166,
"learning_rate": 4.002561639449248e-05,
"loss": 0.2541,
"step": 1940
},
{
"epoch": 2.8,
"grad_norm": 3.3590152263641357,
"learning_rate": 3.998559077809799e-05,
"loss": 0.2091,
"step": 1945
},
{
"epoch": 2.81,
"grad_norm": 3.5402350425720215,
"learning_rate": 3.994556516170349e-05,
"loss": 0.2279,
"step": 1950
},
{
"epoch": 2.81,
"grad_norm": 3.7153353691101074,
"learning_rate": 3.9905539545309e-05,
"loss": 0.2814,
"step": 1955
},
{
"epoch": 2.82,
"grad_norm": 4.787295818328857,
"learning_rate": 3.9865513928914506e-05,
"loss": 0.222,
"step": 1960
},
{
"epoch": 2.83,
"grad_norm": 3.247901201248169,
"learning_rate": 3.9825488312520016e-05,
"loss": 0.2274,
"step": 1965
},
{
"epoch": 2.84,
"grad_norm": 3.51587176322937,
"learning_rate": 3.978546269612553e-05,
"loss": 0.2329,
"step": 1970
},
{
"epoch": 2.84,
"grad_norm": 3.9080677032470703,
"learning_rate": 3.974543707973103e-05,
"loss": 0.2533,
"step": 1975
},
{
"epoch": 2.85,
"grad_norm": 5.821875095367432,
"learning_rate": 3.970541146333654e-05,
"loss": 0.2954,
"step": 1980
},
{
"epoch": 2.86,
"grad_norm": 3.9195449352264404,
"learning_rate": 3.9665385846942044e-05,
"loss": 0.2096,
"step": 1985
},
{
"epoch": 2.86,
"grad_norm": 4.23248815536499,
"learning_rate": 3.9625360230547554e-05,
"loss": 0.2369,
"step": 1990
},
{
"epoch": 2.87,
"grad_norm": 5.370137691497803,
"learning_rate": 3.958533461415306e-05,
"loss": 0.2059,
"step": 1995
},
{
"epoch": 2.88,
"grad_norm": 7.493764400482178,
"learning_rate": 3.954530899775857e-05,
"loss": 0.2312,
"step": 2000
},
{
"epoch": 2.89,
"grad_norm": 5.190505504608154,
"learning_rate": 3.950528338136407e-05,
"loss": 0.1946,
"step": 2005
},
{
"epoch": 2.89,
"grad_norm": 3.261181116104126,
"learning_rate": 3.946525776496958e-05,
"loss": 0.2326,
"step": 2010
},
{
"epoch": 2.9,
"grad_norm": 5.753521919250488,
"learning_rate": 3.942523214857509e-05,
"loss": 0.264,
"step": 2015
},
{
"epoch": 2.91,
"grad_norm": 6.831991195678711,
"learning_rate": 3.9385206532180596e-05,
"loss": 0.2595,
"step": 2020
},
{
"epoch": 2.91,
"grad_norm": 3.9617631435394287,
"learning_rate": 3.9345180915786106e-05,
"loss": 0.2754,
"step": 2025
},
{
"epoch": 2.92,
"grad_norm": 4.283369541168213,
"learning_rate": 3.930515529939161e-05,
"loss": 0.2605,
"step": 2030
},
{
"epoch": 2.93,
"grad_norm": 2.7431774139404297,
"learning_rate": 3.926512968299712e-05,
"loss": 0.2512,
"step": 2035
},
{
"epoch": 2.94,
"grad_norm": 5.476824760437012,
"learning_rate": 3.922510406660262e-05,
"loss": 0.2505,
"step": 2040
},
{
"epoch": 2.94,
"grad_norm": 3.6471610069274902,
"learning_rate": 3.9185078450208134e-05,
"loss": 0.2352,
"step": 2045
},
{
"epoch": 2.95,
"grad_norm": 5.501186847686768,
"learning_rate": 3.9145052833813644e-05,
"loss": 0.2589,
"step": 2050
},
{
"epoch": 2.96,
"grad_norm": 5.927517414093018,
"learning_rate": 3.910502721741915e-05,
"loss": 0.2323,
"step": 2055
},
{
"epoch": 2.97,
"grad_norm": 2.320692539215088,
"learning_rate": 3.906500160102466e-05,
"loss": 0.2417,
"step": 2060
},
{
"epoch": 2.97,
"grad_norm": 4.741787910461426,
"learning_rate": 3.902497598463016e-05,
"loss": 0.2565,
"step": 2065
},
{
"epoch": 2.98,
"grad_norm": 4.28122091293335,
"learning_rate": 3.898495036823567e-05,
"loss": 0.2298,
"step": 2070
},
{
"epoch": 2.99,
"grad_norm": 4.997034072875977,
"learning_rate": 3.8944924751841175e-05,
"loss": 0.2254,
"step": 2075
},
{
"epoch": 2.99,
"grad_norm": 3.0995302200317383,
"learning_rate": 3.8904899135446685e-05,
"loss": 0.2187,
"step": 2080
},
{
"epoch": 3.0,
"eval_accuracy": 0.944838056680162,
"eval_loss": 0.15164293348789215,
"eval_runtime": 26.8648,
"eval_samples_per_second": 367.768,
"eval_steps_per_second": 11.502,
"step": 2084
},
{
"epoch": 3.0,
"grad_norm": 3.3711345195770264,
"learning_rate": 3.8864873519052196e-05,
"loss": 0.2469,
"step": 2085
},
{
"epoch": 3.01,
"grad_norm": 3.6812057495117188,
"learning_rate": 3.88248479026577e-05,
"loss": 0.2199,
"step": 2090
},
{
"epoch": 3.02,
"grad_norm": 3.678959846496582,
"learning_rate": 3.878482228626321e-05,
"loss": 0.2292,
"step": 2095
},
{
"epoch": 3.02,
"grad_norm": 4.287693500518799,
"learning_rate": 3.874479666986871e-05,
"loss": 0.2608,
"step": 2100
},
{
"epoch": 3.03,
"grad_norm": 3.0512278079986572,
"learning_rate": 3.870477105347422e-05,
"loss": 0.186,
"step": 2105
},
{
"epoch": 3.04,
"grad_norm": 7.497838020324707,
"learning_rate": 3.8664745437079733e-05,
"loss": 0.2481,
"step": 2110
},
{
"epoch": 3.04,
"grad_norm": 6.538998126983643,
"learning_rate": 3.862471982068524e-05,
"loss": 0.2415,
"step": 2115
},
{
"epoch": 3.05,
"grad_norm": 3.499305486679077,
"learning_rate": 3.858469420429075e-05,
"loss": 0.2072,
"step": 2120
},
{
"epoch": 3.06,
"grad_norm": 5.734482765197754,
"learning_rate": 3.854466858789626e-05,
"loss": 0.1759,
"step": 2125
},
{
"epoch": 3.07,
"grad_norm": 5.562732696533203,
"learning_rate": 3.850464297150176e-05,
"loss": 0.2327,
"step": 2130
},
{
"epoch": 3.07,
"grad_norm": 3.5141441822052,
"learning_rate": 3.846461735510727e-05,
"loss": 0.2404,
"step": 2135
},
{
"epoch": 3.08,
"grad_norm": 4.6779375076293945,
"learning_rate": 3.842459173871278e-05,
"loss": 0.2097,
"step": 2140
},
{
"epoch": 3.09,
"grad_norm": 5.7871994972229,
"learning_rate": 3.8384566122318285e-05,
"loss": 0.1762,
"step": 2145
},
{
"epoch": 3.09,
"grad_norm": 3.155024290084839,
"learning_rate": 3.8344540505923795e-05,
"loss": 0.2111,
"step": 2150
},
{
"epoch": 3.1,
"grad_norm": 6.282127380371094,
"learning_rate": 3.8304514889529306e-05,
"loss": 0.2194,
"step": 2155
},
{
"epoch": 3.11,
"grad_norm": 4.201207160949707,
"learning_rate": 3.826448927313481e-05,
"loss": 0.2647,
"step": 2160
},
{
"epoch": 3.12,
"grad_norm": 4.582483291625977,
"learning_rate": 3.822446365674032e-05,
"loss": 0.2315,
"step": 2165
},
{
"epoch": 3.12,
"grad_norm": 4.078041076660156,
"learning_rate": 3.818443804034582e-05,
"loss": 0.2249,
"step": 2170
},
{
"epoch": 3.13,
"grad_norm": 4.206385135650635,
"learning_rate": 3.814441242395133e-05,
"loss": 0.2542,
"step": 2175
},
{
"epoch": 3.14,
"grad_norm": 4.039742946624756,
"learning_rate": 3.810438680755684e-05,
"loss": 0.2278,
"step": 2180
},
{
"epoch": 3.15,
"grad_norm": 5.705816745758057,
"learning_rate": 3.806436119116235e-05,
"loss": 0.302,
"step": 2185
},
{
"epoch": 3.15,
"grad_norm": 3.5647683143615723,
"learning_rate": 3.802433557476786e-05,
"loss": 0.2076,
"step": 2190
},
{
"epoch": 3.16,
"grad_norm": 3.260532855987549,
"learning_rate": 3.798430995837336e-05,
"loss": 0.2568,
"step": 2195
},
{
"epoch": 3.17,
"grad_norm": 2.460192918777466,
"learning_rate": 3.794428434197887e-05,
"loss": 0.169,
"step": 2200
},
{
"epoch": 3.17,
"grad_norm": 4.057191848754883,
"learning_rate": 3.7904258725584375e-05,
"loss": 0.2131,
"step": 2205
},
{
"epoch": 3.18,
"grad_norm": 3.4373152256011963,
"learning_rate": 3.7864233109189885e-05,
"loss": 0.2099,
"step": 2210
},
{
"epoch": 3.19,
"grad_norm": 4.2746405601501465,
"learning_rate": 3.782420749279539e-05,
"loss": 0.181,
"step": 2215
},
{
"epoch": 3.2,
"grad_norm": 6.235533714294434,
"learning_rate": 3.77841818764009e-05,
"loss": 0.2497,
"step": 2220
},
{
"epoch": 3.2,
"grad_norm": 4.6660871505737305,
"learning_rate": 3.774415626000641e-05,
"loss": 0.2269,
"step": 2225
},
{
"epoch": 3.21,
"grad_norm": 5.937751293182373,
"learning_rate": 3.770413064361191e-05,
"loss": 0.2385,
"step": 2230
},
{
"epoch": 3.22,
"grad_norm": 3.7235584259033203,
"learning_rate": 3.766410502721742e-05,
"loss": 0.1767,
"step": 2235
},
{
"epoch": 3.22,
"grad_norm": 3.4258034229278564,
"learning_rate": 3.7624079410822926e-05,
"loss": 0.2111,
"step": 2240
},
{
"epoch": 3.23,
"grad_norm": 3.5872249603271484,
"learning_rate": 3.758405379442844e-05,
"loss": 0.1859,
"step": 2245
},
{
"epoch": 3.24,
"grad_norm": 10.837464332580566,
"learning_rate": 3.754402817803394e-05,
"loss": 0.2078,
"step": 2250
},
{
"epoch": 3.25,
"grad_norm": 4.330519199371338,
"learning_rate": 3.750400256163945e-05,
"loss": 0.2145,
"step": 2255
},
{
"epoch": 3.25,
"grad_norm": 2.915856122970581,
"learning_rate": 3.746397694524496e-05,
"loss": 0.2332,
"step": 2260
},
{
"epoch": 3.26,
"grad_norm": 4.104254245758057,
"learning_rate": 3.7423951328850464e-05,
"loss": 0.1799,
"step": 2265
},
{
"epoch": 3.27,
"grad_norm": 4.224066257476807,
"learning_rate": 3.7383925712455975e-05,
"loss": 0.2486,
"step": 2270
},
{
"epoch": 3.27,
"grad_norm": 4.464962005615234,
"learning_rate": 3.734390009606148e-05,
"loss": 0.1884,
"step": 2275
},
{
"epoch": 3.28,
"grad_norm": 4.688321590423584,
"learning_rate": 3.730387447966699e-05,
"loss": 0.2268,
"step": 2280
},
{
"epoch": 3.29,
"grad_norm": 4.25119686126709,
"learning_rate": 3.726384886327249e-05,
"loss": 0.2202,
"step": 2285
},
{
"epoch": 3.3,
"grad_norm": 2.777003765106201,
"learning_rate": 3.7223823246878e-05,
"loss": 0.2411,
"step": 2290
},
{
"epoch": 3.3,
"grad_norm": 4.11234188079834,
"learning_rate": 3.718379763048351e-05,
"loss": 0.22,
"step": 2295
},
{
"epoch": 3.31,
"grad_norm": 4.097933769226074,
"learning_rate": 3.7143772014089016e-05,
"loss": 0.2586,
"step": 2300
},
{
"epoch": 3.32,
"grad_norm": 4.043206691741943,
"learning_rate": 3.7103746397694526e-05,
"loss": 0.2472,
"step": 2305
},
{
"epoch": 3.32,
"grad_norm": 5.116567611694336,
"learning_rate": 3.706372078130003e-05,
"loss": 0.235,
"step": 2310
},
{
"epoch": 3.33,
"grad_norm": 3.207937717437744,
"learning_rate": 3.702369516490554e-05,
"loss": 0.2311,
"step": 2315
},
{
"epoch": 3.34,
"grad_norm": 3.5227158069610596,
"learning_rate": 3.6983669548511043e-05,
"loss": 0.2425,
"step": 2320
},
{
"epoch": 3.35,
"grad_norm": 3.8556931018829346,
"learning_rate": 3.6943643932116554e-05,
"loss": 0.2619,
"step": 2325
},
{
"epoch": 3.35,
"grad_norm": 3.575005054473877,
"learning_rate": 3.6903618315722064e-05,
"loss": 0.1952,
"step": 2330
},
{
"epoch": 3.36,
"grad_norm": 3.004302740097046,
"learning_rate": 3.686359269932757e-05,
"loss": 0.1923,
"step": 2335
},
{
"epoch": 3.37,
"grad_norm": 4.4943976402282715,
"learning_rate": 3.682356708293308e-05,
"loss": 0.2607,
"step": 2340
},
{
"epoch": 3.38,
"grad_norm": 4.194296360015869,
"learning_rate": 3.678354146653858e-05,
"loss": 0.2363,
"step": 2345
},
{
"epoch": 3.38,
"grad_norm": 5.696502208709717,
"learning_rate": 3.674351585014409e-05,
"loss": 0.2458,
"step": 2350
},
{
"epoch": 3.39,
"grad_norm": 3.8679604530334473,
"learning_rate": 3.67034902337496e-05,
"loss": 0.2113,
"step": 2355
},
{
"epoch": 3.4,
"grad_norm": 4.158144474029541,
"learning_rate": 3.666346461735511e-05,
"loss": 0.2335,
"step": 2360
},
{
"epoch": 3.4,
"grad_norm": 3.0606186389923096,
"learning_rate": 3.6623439000960616e-05,
"loss": 0.1956,
"step": 2365
},
{
"epoch": 3.41,
"grad_norm": 3.3307764530181885,
"learning_rate": 3.6583413384566126e-05,
"loss": 0.2589,
"step": 2370
},
{
"epoch": 3.42,
"grad_norm": 4.276596546173096,
"learning_rate": 3.6543387768171636e-05,
"loss": 0.2316,
"step": 2375
},
{
"epoch": 3.43,
"grad_norm": 3.0126774311065674,
"learning_rate": 3.650336215177714e-05,
"loss": 0.185,
"step": 2380
},
{
"epoch": 3.43,
"grad_norm": 3.945223331451416,
"learning_rate": 3.646333653538265e-05,
"loss": 0.1646,
"step": 2385
},
{
"epoch": 3.44,
"grad_norm": 3.874509334564209,
"learning_rate": 3.6423310918988154e-05,
"loss": 0.1987,
"step": 2390
},
{
"epoch": 3.45,
"grad_norm": 4.1151018142700195,
"learning_rate": 3.6383285302593664e-05,
"loss": 0.2219,
"step": 2395
},
{
"epoch": 3.45,
"grad_norm": 4.987639427185059,
"learning_rate": 3.6343259686199174e-05,
"loss": 0.1767,
"step": 2400
},
{
"epoch": 3.46,
"grad_norm": 3.1510512828826904,
"learning_rate": 3.630323406980468e-05,
"loss": 0.1942,
"step": 2405
},
{
"epoch": 3.47,
"grad_norm": 4.838677406311035,
"learning_rate": 3.626320845341019e-05,
"loss": 0.2122,
"step": 2410
},
{
"epoch": 3.48,
"grad_norm": 5.568162441253662,
"learning_rate": 3.622318283701569e-05,
"loss": 0.2429,
"step": 2415
},
{
"epoch": 3.48,
"grad_norm": 3.110318183898926,
"learning_rate": 3.61831572206212e-05,
"loss": 0.2571,
"step": 2420
},
{
"epoch": 3.49,
"grad_norm": 3.7364163398742676,
"learning_rate": 3.6143131604226705e-05,
"loss": 0.2329,
"step": 2425
},
{
"epoch": 3.5,
"grad_norm": 4.126183986663818,
"learning_rate": 3.6103105987832216e-05,
"loss": 0.2439,
"step": 2430
},
{
"epoch": 3.5,
"grad_norm": 5.250816345214844,
"learning_rate": 3.6063080371437726e-05,
"loss": 0.1947,
"step": 2435
},
{
"epoch": 3.51,
"grad_norm": 4.749518871307373,
"learning_rate": 3.602305475504323e-05,
"loss": 0.2585,
"step": 2440
},
{
"epoch": 3.52,
"grad_norm": 4.378204345703125,
"learning_rate": 3.598302913864874e-05,
"loss": 0.2036,
"step": 2445
},
{
"epoch": 3.53,
"grad_norm": 4.115077018737793,
"learning_rate": 3.594300352225424e-05,
"loss": 0.1862,
"step": 2450
},
{
"epoch": 3.53,
"grad_norm": 3.890913963317871,
"learning_rate": 3.5902977905859753e-05,
"loss": 0.1771,
"step": 2455
},
{
"epoch": 3.54,
"grad_norm": 4.176694393157959,
"learning_rate": 3.586295228946526e-05,
"loss": 0.1872,
"step": 2460
},
{
"epoch": 3.55,
"grad_norm": 3.7083818912506104,
"learning_rate": 3.582292667307077e-05,
"loss": 0.2303,
"step": 2465
},
{
"epoch": 3.56,
"grad_norm": 5.073873996734619,
"learning_rate": 3.578290105667628e-05,
"loss": 0.2079,
"step": 2470
},
{
"epoch": 3.56,
"grad_norm": 3.9572906494140625,
"learning_rate": 3.574287544028178e-05,
"loss": 0.2173,
"step": 2475
},
{
"epoch": 3.57,
"grad_norm": 3.172136068344116,
"learning_rate": 3.570284982388729e-05,
"loss": 0.2303,
"step": 2480
},
{
"epoch": 3.58,
"grad_norm": 3.644803285598755,
"learning_rate": 3.5662824207492795e-05,
"loss": 0.2467,
"step": 2485
},
{
"epoch": 3.58,
"grad_norm": 3.4908697605133057,
"learning_rate": 3.5622798591098305e-05,
"loss": 0.2053,
"step": 2490
},
{
"epoch": 3.59,
"grad_norm": 3.2498650550842285,
"learning_rate": 3.558277297470381e-05,
"loss": 0.2058,
"step": 2495
},
{
"epoch": 3.6,
"grad_norm": 4.260957717895508,
"learning_rate": 3.554274735830932e-05,
"loss": 0.25,
"step": 2500
},
{
"epoch": 3.61,
"grad_norm": 4.03501558303833,
"learning_rate": 3.550272174191483e-05,
"loss": 0.2376,
"step": 2505
},
{
"epoch": 3.61,
"grad_norm": 3.145692825317383,
"learning_rate": 3.546269612552033e-05,
"loss": 0.219,
"step": 2510
},
{
"epoch": 3.62,
"grad_norm": 3.201892614364624,
"learning_rate": 3.542267050912584e-05,
"loss": 0.2202,
"step": 2515
},
{
"epoch": 3.63,
"grad_norm": 2.6380460262298584,
"learning_rate": 3.5382644892731347e-05,
"loss": 0.206,
"step": 2520
},
{
"epoch": 3.63,
"grad_norm": 3.144564390182495,
"learning_rate": 3.534261927633686e-05,
"loss": 0.2259,
"step": 2525
},
{
"epoch": 3.64,
"grad_norm": 7.223779201507568,
"learning_rate": 3.530259365994236e-05,
"loss": 0.2447,
"step": 2530
},
{
"epoch": 3.65,
"grad_norm": 4.040417194366455,
"learning_rate": 3.526256804354787e-05,
"loss": 0.2032,
"step": 2535
},
{
"epoch": 3.66,
"grad_norm": 3.7327756881713867,
"learning_rate": 3.522254242715338e-05,
"loss": 0.2437,
"step": 2540
},
{
"epoch": 3.66,
"grad_norm": 6.253401279449463,
"learning_rate": 3.5182516810758884e-05,
"loss": 0.2113,
"step": 2545
},
{
"epoch": 3.67,
"grad_norm": 4.586529731750488,
"learning_rate": 3.5142491194364395e-05,
"loss": 0.2213,
"step": 2550
},
{
"epoch": 3.68,
"grad_norm": 4.374157428741455,
"learning_rate": 3.51024655779699e-05,
"loss": 0.2183,
"step": 2555
},
{
"epoch": 3.68,
"grad_norm": 4.311252593994141,
"learning_rate": 3.506243996157541e-05,
"loss": 0.2223,
"step": 2560
},
{
"epoch": 3.69,
"grad_norm": 5.473079204559326,
"learning_rate": 3.502241434518091e-05,
"loss": 0.2318,
"step": 2565
},
{
"epoch": 3.7,
"grad_norm": 3.0324044227600098,
"learning_rate": 3.498238872878642e-05,
"loss": 0.2141,
"step": 2570
},
{
"epoch": 3.71,
"grad_norm": 3.7828001976013184,
"learning_rate": 3.494236311239193e-05,
"loss": 0.2164,
"step": 2575
},
{
"epoch": 3.71,
"grad_norm": 3.766996383666992,
"learning_rate": 3.4902337495997436e-05,
"loss": 0.2597,
"step": 2580
},
{
"epoch": 3.72,
"grad_norm": 5.172212600708008,
"learning_rate": 3.4862311879602946e-05,
"loss": 0.2471,
"step": 2585
},
{
"epoch": 3.73,
"grad_norm": 3.6711273193359375,
"learning_rate": 3.482228626320846e-05,
"loss": 0.2387,
"step": 2590
},
{
"epoch": 3.74,
"grad_norm": 3.535651445388794,
"learning_rate": 3.478226064681396e-05,
"loss": 0.2029,
"step": 2595
},
{
"epoch": 3.74,
"grad_norm": 3.2503421306610107,
"learning_rate": 3.474223503041947e-05,
"loss": 0.2263,
"step": 2600
},
{
"epoch": 3.75,
"grad_norm": 3.7369303703308105,
"learning_rate": 3.470220941402498e-05,
"loss": 0.2374,
"step": 2605
},
{
"epoch": 3.76,
"grad_norm": 3.4637975692749023,
"learning_rate": 3.4662183797630484e-05,
"loss": 0.215,
"step": 2610
},
{
"epoch": 3.76,
"grad_norm": 3.3387277126312256,
"learning_rate": 3.4622158181235995e-05,
"loss": 0.2168,
"step": 2615
},
{
"epoch": 3.77,
"grad_norm": 4.119049072265625,
"learning_rate": 3.4582132564841505e-05,
"loss": 0.2292,
"step": 2620
},
{
"epoch": 3.78,
"grad_norm": 3.3130059242248535,
"learning_rate": 3.454210694844701e-05,
"loss": 0.1932,
"step": 2625
},
{
"epoch": 3.79,
"grad_norm": 4.4937424659729,
"learning_rate": 3.450208133205252e-05,
"loss": 0.1795,
"step": 2630
},
{
"epoch": 3.79,
"grad_norm": 3.4141628742218018,
"learning_rate": 3.446205571565802e-05,
"loss": 0.2059,
"step": 2635
},
{
"epoch": 3.8,
"grad_norm": 3.247016191482544,
"learning_rate": 3.442203009926353e-05,
"loss": 0.2439,
"step": 2640
},
{
"epoch": 3.81,
"grad_norm": 4.921724319458008,
"learning_rate": 3.438200448286904e-05,
"loss": 0.2441,
"step": 2645
},
{
"epoch": 3.81,
"grad_norm": 9.78645133972168,
"learning_rate": 3.4341978866474546e-05,
"loss": 0.2285,
"step": 2650
},
{
"epoch": 3.82,
"grad_norm": 4.2883148193359375,
"learning_rate": 3.4301953250080056e-05,
"loss": 0.1949,
"step": 2655
},
{
"epoch": 3.83,
"grad_norm": 4.0928544998168945,
"learning_rate": 3.426192763368556e-05,
"loss": 0.2025,
"step": 2660
},
{
"epoch": 3.84,
"grad_norm": 3.5520989894866943,
"learning_rate": 3.422190201729107e-05,
"loss": 0.2041,
"step": 2665
},
{
"epoch": 3.84,
"grad_norm": 6.448659896850586,
"learning_rate": 3.4181876400896574e-05,
"loss": 0.2051,
"step": 2670
},
{
"epoch": 3.85,
"grad_norm": 3.6344971656799316,
"learning_rate": 3.4141850784502084e-05,
"loss": 0.198,
"step": 2675
},
{
"epoch": 3.86,
"grad_norm": 3.866029977798462,
"learning_rate": 3.4101825168107594e-05,
"loss": 0.2465,
"step": 2680
},
{
"epoch": 3.86,
"grad_norm": 5.6679840087890625,
"learning_rate": 3.40617995517131e-05,
"loss": 0.2201,
"step": 2685
},
{
"epoch": 3.87,
"grad_norm": 4.224870204925537,
"learning_rate": 3.402177393531861e-05,
"loss": 0.1742,
"step": 2690
},
{
"epoch": 3.88,
"grad_norm": 3.722661256790161,
"learning_rate": 3.398174831892411e-05,
"loss": 0.2421,
"step": 2695
},
{
"epoch": 3.89,
"grad_norm": 3.9871938228607178,
"learning_rate": 3.394172270252962e-05,
"loss": 0.2048,
"step": 2700
},
{
"epoch": 3.89,
"grad_norm": 4.0545806884765625,
"learning_rate": 3.3901697086135125e-05,
"loss": 0.1936,
"step": 2705
},
{
"epoch": 3.9,
"grad_norm": 3.1876699924468994,
"learning_rate": 3.3861671469740636e-05,
"loss": 0.2055,
"step": 2710
},
{
"epoch": 3.91,
"grad_norm": 4.426385879516602,
"learning_rate": 3.3821645853346146e-05,
"loss": 0.2047,
"step": 2715
},
{
"epoch": 3.92,
"grad_norm": 3.2227375507354736,
"learning_rate": 3.378162023695165e-05,
"loss": 0.2183,
"step": 2720
},
{
"epoch": 3.92,
"grad_norm": 4.088392734527588,
"learning_rate": 3.374159462055716e-05,
"loss": 0.2094,
"step": 2725
},
{
"epoch": 3.93,
"grad_norm": 5.6704182624816895,
"learning_rate": 3.370156900416266e-05,
"loss": 0.2646,
"step": 2730
},
{
"epoch": 3.94,
"grad_norm": 4.020977973937988,
"learning_rate": 3.3661543387768174e-05,
"loss": 0.2152,
"step": 2735
},
{
"epoch": 3.94,
"grad_norm": 3.417588710784912,
"learning_rate": 3.362151777137368e-05,
"loss": 0.2011,
"step": 2740
},
{
"epoch": 3.95,
"grad_norm": 4.737537384033203,
"learning_rate": 3.358149215497919e-05,
"loss": 0.2468,
"step": 2745
},
{
"epoch": 3.96,
"grad_norm": 5.6779351234436035,
"learning_rate": 3.35414665385847e-05,
"loss": 0.2208,
"step": 2750
},
{
"epoch": 3.97,
"grad_norm": 3.3792531490325928,
"learning_rate": 3.35014409221902e-05,
"loss": 0.2221,
"step": 2755
},
{
"epoch": 3.97,
"grad_norm": 2.7218449115753174,
"learning_rate": 3.346141530579571e-05,
"loss": 0.2376,
"step": 2760
},
{
"epoch": 3.98,
"grad_norm": 5.76757287979126,
"learning_rate": 3.3421389689401215e-05,
"loss": 0.2505,
"step": 2765
},
{
"epoch": 3.99,
"grad_norm": 4.965346813201904,
"learning_rate": 3.3381364073006725e-05,
"loss": 0.2339,
"step": 2770
},
{
"epoch": 3.99,
"grad_norm": 3.6965980529785156,
"learning_rate": 3.334133845661223e-05,
"loss": 0.1491,
"step": 2775
},
{
"epoch": 4.0,
"eval_accuracy": 0.9518218623481781,
"eval_loss": 0.13597902655601501,
"eval_runtime": 26.7799,
"eval_samples_per_second": 368.933,
"eval_steps_per_second": 11.538,
"step": 2779
},
{
"epoch": 4.0,
"grad_norm": 4.749961853027344,
"learning_rate": 3.330131284021774e-05,
"loss": 0.2327,
"step": 2780
},
{
"epoch": 4.01,
"grad_norm": 2.841573476791382,
"learning_rate": 3.326128722382325e-05,
"loss": 0.2243,
"step": 2785
},
{
"epoch": 4.02,
"grad_norm": 5.107615947723389,
"learning_rate": 3.322126160742875e-05,
"loss": 0.2101,
"step": 2790
},
{
"epoch": 4.02,
"grad_norm": 5.680942535400391,
"learning_rate": 3.318123599103426e-05,
"loss": 0.1936,
"step": 2795
},
{
"epoch": 4.03,
"grad_norm": 3.2696499824523926,
"learning_rate": 3.314121037463977e-05,
"loss": 0.2145,
"step": 2800
},
{
"epoch": 4.04,
"grad_norm": 3.674295425415039,
"learning_rate": 3.310118475824528e-05,
"loss": 0.2228,
"step": 2805
},
{
"epoch": 4.04,
"grad_norm": 3.9477553367614746,
"learning_rate": 3.306115914185078e-05,
"loss": 0.2029,
"step": 2810
},
{
"epoch": 4.05,
"grad_norm": 2.8797388076782227,
"learning_rate": 3.302113352545629e-05,
"loss": 0.1741,
"step": 2815
},
{
"epoch": 4.06,
"grad_norm": 3.3909058570861816,
"learning_rate": 3.29811079090618e-05,
"loss": 0.2246,
"step": 2820
},
{
"epoch": 4.07,
"grad_norm": 5.92777156829834,
"learning_rate": 3.2941082292667305e-05,
"loss": 0.2162,
"step": 2825
},
{
"epoch": 4.07,
"grad_norm": 9.48619556427002,
"learning_rate": 3.2901056676272815e-05,
"loss": 0.1948,
"step": 2830
},
{
"epoch": 4.08,
"grad_norm": 4.641541004180908,
"learning_rate": 3.2861031059878325e-05,
"loss": 0.2002,
"step": 2835
},
{
"epoch": 4.09,
"grad_norm": 3.901582956314087,
"learning_rate": 3.2821005443483835e-05,
"loss": 0.2303,
"step": 2840
},
{
"epoch": 4.09,
"grad_norm": 3.974966526031494,
"learning_rate": 3.278097982708934e-05,
"loss": 0.17,
"step": 2845
},
{
"epoch": 4.1,
"grad_norm": 4.02310037612915,
"learning_rate": 3.274095421069485e-05,
"loss": 0.1959,
"step": 2850
},
{
"epoch": 4.11,
"grad_norm": 2.948073148727417,
"learning_rate": 3.270092859430036e-05,
"loss": 0.1819,
"step": 2855
},
{
"epoch": 4.12,
"grad_norm": 4.0745391845703125,
"learning_rate": 3.266090297790586e-05,
"loss": 0.2008,
"step": 2860
},
{
"epoch": 4.12,
"grad_norm": 4.957521438598633,
"learning_rate": 3.262087736151137e-05,
"loss": 0.2019,
"step": 2865
},
{
"epoch": 4.13,
"grad_norm": 3.015536069869995,
"learning_rate": 3.258085174511688e-05,
"loss": 0.2166,
"step": 2870
},
{
"epoch": 4.14,
"grad_norm": 4.570614337921143,
"learning_rate": 3.254082612872239e-05,
"loss": 0.2058,
"step": 2875
},
{
"epoch": 4.15,
"grad_norm": 4.153733253479004,
"learning_rate": 3.250080051232789e-05,
"loss": 0.2092,
"step": 2880
},
{
"epoch": 4.15,
"grad_norm": 5.32897424697876,
"learning_rate": 3.24607748959334e-05,
"loss": 0.2379,
"step": 2885
},
{
"epoch": 4.16,
"grad_norm": 4.037844181060791,
"learning_rate": 3.242074927953891e-05,
"loss": 0.2191,
"step": 2890
},
{
"epoch": 4.17,
"grad_norm": 4.094261646270752,
"learning_rate": 3.2380723663144415e-05,
"loss": 0.1733,
"step": 2895
},
{
"epoch": 4.17,
"grad_norm": 5.6198015213012695,
"learning_rate": 3.2340698046749925e-05,
"loss": 0.2437,
"step": 2900
},
{
"epoch": 4.18,
"grad_norm": 2.6774141788482666,
"learning_rate": 3.230067243035543e-05,
"loss": 0.1806,
"step": 2905
},
{
"epoch": 4.19,
"grad_norm": 4.591428279876709,
"learning_rate": 3.226064681396094e-05,
"loss": 0.1828,
"step": 2910
},
{
"epoch": 4.2,
"grad_norm": 3.5464799404144287,
"learning_rate": 3.222062119756644e-05,
"loss": 0.2177,
"step": 2915
},
{
"epoch": 4.2,
"grad_norm": 3.6952743530273438,
"learning_rate": 3.218059558117195e-05,
"loss": 0.1963,
"step": 2920
},
{
"epoch": 4.21,
"grad_norm": 2.857217788696289,
"learning_rate": 3.214056996477746e-05,
"loss": 0.1828,
"step": 2925
},
{
"epoch": 4.22,
"grad_norm": 3.064542293548584,
"learning_rate": 3.2100544348382966e-05,
"loss": 0.2227,
"step": 2930
},
{
"epoch": 4.22,
"grad_norm": 5.486659526824951,
"learning_rate": 3.206051873198848e-05,
"loss": 0.2019,
"step": 2935
},
{
"epoch": 4.23,
"grad_norm": 3.4268643856048584,
"learning_rate": 3.202049311559398e-05,
"loss": 0.169,
"step": 2940
},
{
"epoch": 4.24,
"grad_norm": 4.867826461791992,
"learning_rate": 3.198046749919949e-05,
"loss": 0.2222,
"step": 2945
},
{
"epoch": 4.25,
"grad_norm": 1.942135214805603,
"learning_rate": 3.1940441882804994e-05,
"loss": 0.2087,
"step": 2950
},
{
"epoch": 4.25,
"grad_norm": 4.17222261428833,
"learning_rate": 3.1900416266410504e-05,
"loss": 0.1833,
"step": 2955
},
{
"epoch": 4.26,
"grad_norm": 3.8982694149017334,
"learning_rate": 3.1860390650016015e-05,
"loss": 0.1877,
"step": 2960
},
{
"epoch": 4.27,
"grad_norm": 3.6078577041625977,
"learning_rate": 3.182036503362152e-05,
"loss": 0.1587,
"step": 2965
},
{
"epoch": 4.27,
"grad_norm": 4.184298992156982,
"learning_rate": 3.178033941722703e-05,
"loss": 0.1787,
"step": 2970
},
{
"epoch": 4.28,
"grad_norm": 4.788322925567627,
"learning_rate": 3.174031380083253e-05,
"loss": 0.2037,
"step": 2975
},
{
"epoch": 4.29,
"grad_norm": 4.06505012512207,
"learning_rate": 3.170028818443804e-05,
"loss": 0.2102,
"step": 2980
},
{
"epoch": 4.3,
"grad_norm": 4.116428375244141,
"learning_rate": 3.1660262568043546e-05,
"loss": 0.1717,
"step": 2985
},
{
"epoch": 4.3,
"grad_norm": 2.654528856277466,
"learning_rate": 3.1620236951649056e-05,
"loss": 0.1519,
"step": 2990
},
{
"epoch": 4.31,
"grad_norm": 2.9073519706726074,
"learning_rate": 3.1580211335254566e-05,
"loss": 0.2359,
"step": 2995
},
{
"epoch": 4.32,
"grad_norm": 4.120742321014404,
"learning_rate": 3.154018571886007e-05,
"loss": 0.2294,
"step": 3000
},
{
"epoch": 4.33,
"grad_norm": 3.727126121520996,
"learning_rate": 3.150016010246558e-05,
"loss": 0.186,
"step": 3005
},
{
"epoch": 4.33,
"grad_norm": 3.9322612285614014,
"learning_rate": 3.1460134486071084e-05,
"loss": 0.2311,
"step": 3010
},
{
"epoch": 4.34,
"grad_norm": 3.876603841781616,
"learning_rate": 3.1420108869676594e-05,
"loss": 0.2384,
"step": 3015
},
{
"epoch": 4.35,
"grad_norm": 3.5300958156585693,
"learning_rate": 3.13800832532821e-05,
"loss": 0.2141,
"step": 3020
},
{
"epoch": 4.35,
"grad_norm": 3.2177224159240723,
"learning_rate": 3.134005763688761e-05,
"loss": 0.2157,
"step": 3025
},
{
"epoch": 4.36,
"grad_norm": 5.607058048248291,
"learning_rate": 3.130003202049312e-05,
"loss": 0.2121,
"step": 3030
},
{
"epoch": 4.37,
"grad_norm": 4.676594257354736,
"learning_rate": 3.126000640409862e-05,
"loss": 0.2647,
"step": 3035
},
{
"epoch": 4.38,
"grad_norm": 4.193298816680908,
"learning_rate": 3.121998078770413e-05,
"loss": 0.1926,
"step": 3040
},
{
"epoch": 4.38,
"grad_norm": 4.498405933380127,
"learning_rate": 3.1179955171309635e-05,
"loss": 0.2174,
"step": 3045
},
{
"epoch": 4.39,
"grad_norm": 2.3342103958129883,
"learning_rate": 3.1139929554915145e-05,
"loss": 0.1952,
"step": 3050
},
{
"epoch": 4.4,
"grad_norm": 3.535201072692871,
"learning_rate": 3.1099903938520656e-05,
"loss": 0.2383,
"step": 3055
},
{
"epoch": 4.4,
"grad_norm": 3.458115816116333,
"learning_rate": 3.105987832212616e-05,
"loss": 0.251,
"step": 3060
},
{
"epoch": 4.41,
"grad_norm": 5.184045314788818,
"learning_rate": 3.101985270573167e-05,
"loss": 0.2367,
"step": 3065
},
{
"epoch": 4.42,
"grad_norm": 5.617215156555176,
"learning_rate": 3.097982708933718e-05,
"loss": 0.2113,
"step": 3070
},
{
"epoch": 4.43,
"grad_norm": 2.9581499099731445,
"learning_rate": 3.093980147294268e-05,
"loss": 0.1746,
"step": 3075
},
{
"epoch": 4.43,
"grad_norm": 4.895233631134033,
"learning_rate": 3.0899775856548194e-05,
"loss": 0.2228,
"step": 3080
},
{
"epoch": 4.44,
"grad_norm": 6.0638346672058105,
"learning_rate": 3.0859750240153704e-05,
"loss": 0.2143,
"step": 3085
},
{
"epoch": 4.45,
"grad_norm": 3.597208023071289,
"learning_rate": 3.081972462375921e-05,
"loss": 0.1849,
"step": 3090
},
{
"epoch": 4.45,
"grad_norm": 4.183586597442627,
"learning_rate": 3.077969900736472e-05,
"loss": 0.1877,
"step": 3095
},
{
"epoch": 4.46,
"grad_norm": 4.0433855056762695,
"learning_rate": 3.073967339097023e-05,
"loss": 0.1655,
"step": 3100
},
{
"epoch": 4.47,
"grad_norm": 4.732285499572754,
"learning_rate": 3.069964777457573e-05,
"loss": 0.2016,
"step": 3105
},
{
"epoch": 4.48,
"grad_norm": 3.617846727371216,
"learning_rate": 3.065962215818124e-05,
"loss": 0.197,
"step": 3110
},
{
"epoch": 4.48,
"grad_norm": 4.5917134284973145,
"learning_rate": 3.0619596541786745e-05,
"loss": 0.1746,
"step": 3115
},
{
"epoch": 4.49,
"grad_norm": 5.0926032066345215,
"learning_rate": 3.0579570925392256e-05,
"loss": 0.1932,
"step": 3120
},
{
"epoch": 4.5,
"grad_norm": 3.462219715118408,
"learning_rate": 3.053954530899776e-05,
"loss": 0.1851,
"step": 3125
},
{
"epoch": 4.51,
"grad_norm": 4.013971328735352,
"learning_rate": 3.049951969260327e-05,
"loss": 0.1873,
"step": 3130
},
{
"epoch": 4.51,
"grad_norm": 4.1465373039245605,
"learning_rate": 3.0459494076208776e-05,
"loss": 0.2215,
"step": 3135
},
{
"epoch": 4.52,
"grad_norm": 3.554340124130249,
"learning_rate": 3.0419468459814283e-05,
"loss": 0.2169,
"step": 3140
},
{
"epoch": 4.53,
"grad_norm": 5.059957981109619,
"learning_rate": 3.0379442843419793e-05,
"loss": 0.2123,
"step": 3145
},
{
"epoch": 4.53,
"grad_norm": 3.727722406387329,
"learning_rate": 3.0339417227025297e-05,
"loss": 0.1717,
"step": 3150
},
{
"epoch": 4.54,
"grad_norm": 3.030172348022461,
"learning_rate": 3.0299391610630807e-05,
"loss": 0.1813,
"step": 3155
},
{
"epoch": 4.55,
"grad_norm": 3.366147994995117,
"learning_rate": 3.025936599423631e-05,
"loss": 0.1761,
"step": 3160
},
{
"epoch": 4.56,
"grad_norm": 3.159599542617798,
"learning_rate": 3.021934037784182e-05,
"loss": 0.2178,
"step": 3165
},
{
"epoch": 4.56,
"grad_norm": 4.538158416748047,
"learning_rate": 3.017931476144733e-05,
"loss": 0.2123,
"step": 3170
},
{
"epoch": 4.57,
"grad_norm": 4.336025714874268,
"learning_rate": 3.0139289145052835e-05,
"loss": 0.2083,
"step": 3175
},
{
"epoch": 4.58,
"grad_norm": 8.234651565551758,
"learning_rate": 3.0099263528658345e-05,
"loss": 0.1929,
"step": 3180
},
{
"epoch": 4.58,
"grad_norm": 3.061892509460449,
"learning_rate": 3.005923791226385e-05,
"loss": 0.1805,
"step": 3185
},
{
"epoch": 4.59,
"grad_norm": 3.444242238998413,
"learning_rate": 3.001921229586936e-05,
"loss": 0.1897,
"step": 3190
},
{
"epoch": 4.6,
"grad_norm": 3.281337261199951,
"learning_rate": 2.9979186679474862e-05,
"loss": 0.1676,
"step": 3195
},
{
"epoch": 4.61,
"grad_norm": 4.178293228149414,
"learning_rate": 2.9939161063080373e-05,
"loss": 0.2371,
"step": 3200
},
{
"epoch": 4.61,
"grad_norm": 3.2570972442626953,
"learning_rate": 2.9899135446685883e-05,
"loss": 0.2106,
"step": 3205
},
{
"epoch": 4.62,
"grad_norm": 4.102814197540283,
"learning_rate": 2.9859109830291387e-05,
"loss": 0.2123,
"step": 3210
},
{
"epoch": 4.63,
"grad_norm": 4.337564945220947,
"learning_rate": 2.9819084213896897e-05,
"loss": 0.2158,
"step": 3215
},
{
"epoch": 4.63,
"grad_norm": 4.369962692260742,
"learning_rate": 2.97790585975024e-05,
"loss": 0.1974,
"step": 3220
},
{
"epoch": 4.64,
"grad_norm": 3.994830369949341,
"learning_rate": 2.973903298110791e-05,
"loss": 0.1798,
"step": 3225
},
{
"epoch": 4.65,
"grad_norm": 5.044273376464844,
"learning_rate": 2.9699007364713418e-05,
"loss": 0.2527,
"step": 3230
},
{
"epoch": 4.66,
"grad_norm": 4.7880682945251465,
"learning_rate": 2.9658981748318924e-05,
"loss": 0.2461,
"step": 3235
},
{
"epoch": 4.66,
"grad_norm": 3.5522334575653076,
"learning_rate": 2.9618956131924435e-05,
"loss": 0.1891,
"step": 3240
},
{
"epoch": 4.67,
"grad_norm": 3.3652679920196533,
"learning_rate": 2.957893051552994e-05,
"loss": 0.2123,
"step": 3245
},
{
"epoch": 4.68,
"grad_norm": 3.255449056625366,
"learning_rate": 2.953890489913545e-05,
"loss": 0.204,
"step": 3250
},
{
"epoch": 4.69,
"grad_norm": 3.2218666076660156,
"learning_rate": 2.9498879282740955e-05,
"loss": 0.1947,
"step": 3255
},
{
"epoch": 4.69,
"grad_norm": 4.256184101104736,
"learning_rate": 2.9458853666346466e-05,
"loss": 0.1948,
"step": 3260
},
{
"epoch": 4.7,
"grad_norm": 3.0359115600585938,
"learning_rate": 2.941882804995197e-05,
"loss": 0.2015,
"step": 3265
},
{
"epoch": 4.71,
"grad_norm": 3.3159563541412354,
"learning_rate": 2.937880243355748e-05,
"loss": 0.223,
"step": 3270
},
{
"epoch": 4.71,
"grad_norm": 4.972184658050537,
"learning_rate": 2.933877681716299e-05,
"loss": 0.2057,
"step": 3275
},
{
"epoch": 4.72,
"grad_norm": 3.500704526901245,
"learning_rate": 2.9298751200768493e-05,
"loss": 0.1895,
"step": 3280
},
{
"epoch": 4.73,
"grad_norm": 2.18817400932312,
"learning_rate": 2.9258725584374004e-05,
"loss": 0.2213,
"step": 3285
},
{
"epoch": 4.74,
"grad_norm": 3.3040566444396973,
"learning_rate": 2.9218699967979507e-05,
"loss": 0.1929,
"step": 3290
},
{
"epoch": 4.74,
"grad_norm": 3.868074417114258,
"learning_rate": 2.9178674351585017e-05,
"loss": 0.2007,
"step": 3295
},
{
"epoch": 4.75,
"grad_norm": 2.987790584564209,
"learning_rate": 2.913864873519052e-05,
"loss": 0.1843,
"step": 3300
},
{
"epoch": 4.76,
"grad_norm": 3.395069122314453,
"learning_rate": 2.909862311879603e-05,
"loss": 0.2006,
"step": 3305
},
{
"epoch": 4.76,
"grad_norm": 5.1089606285095215,
"learning_rate": 2.905859750240154e-05,
"loss": 0.1996,
"step": 3310
},
{
"epoch": 4.77,
"grad_norm": 3.1184518337249756,
"learning_rate": 2.9018571886007045e-05,
"loss": 0.1927,
"step": 3315
},
{
"epoch": 4.78,
"grad_norm": 3.2215800285339355,
"learning_rate": 2.8978546269612555e-05,
"loss": 0.2127,
"step": 3320
},
{
"epoch": 4.79,
"grad_norm": 3.3580570220947266,
"learning_rate": 2.893852065321806e-05,
"loss": 0.2274,
"step": 3325
},
{
"epoch": 4.79,
"grad_norm": 4.6123366355896,
"learning_rate": 2.889849503682357e-05,
"loss": 0.1864,
"step": 3330
},
{
"epoch": 4.8,
"grad_norm": 3.454392910003662,
"learning_rate": 2.8858469420429073e-05,
"loss": 0.1665,
"step": 3335
},
{
"epoch": 4.81,
"grad_norm": 3.018871784210205,
"learning_rate": 2.8818443804034583e-05,
"loss": 0.1969,
"step": 3340
},
{
"epoch": 4.81,
"grad_norm": 3.789463520050049,
"learning_rate": 2.8778418187640093e-05,
"loss": 0.2156,
"step": 3345
},
{
"epoch": 4.82,
"grad_norm": 2.578465223312378,
"learning_rate": 2.8738392571245597e-05,
"loss": 0.1985,
"step": 3350
},
{
"epoch": 4.83,
"grad_norm": 5.120762825012207,
"learning_rate": 2.8698366954851107e-05,
"loss": 0.1897,
"step": 3355
},
{
"epoch": 4.84,
"grad_norm": 5.5928425788879395,
"learning_rate": 2.8658341338456614e-05,
"loss": 0.2231,
"step": 3360
},
{
"epoch": 4.84,
"grad_norm": 3.3155384063720703,
"learning_rate": 2.861831572206212e-05,
"loss": 0.1672,
"step": 3365
},
{
"epoch": 4.85,
"grad_norm": 2.947312116622925,
"learning_rate": 2.8578290105667628e-05,
"loss": 0.1473,
"step": 3370
},
{
"epoch": 4.86,
"grad_norm": 5.2089314460754395,
"learning_rate": 2.8538264489273138e-05,
"loss": 0.2265,
"step": 3375
},
{
"epoch": 4.87,
"grad_norm": 3.4539883136749268,
"learning_rate": 2.8498238872878645e-05,
"loss": 0.1995,
"step": 3380
},
{
"epoch": 4.87,
"grad_norm": 4.001387596130371,
"learning_rate": 2.845821325648415e-05,
"loss": 0.1999,
"step": 3385
},
{
"epoch": 4.88,
"grad_norm": 3.7366297245025635,
"learning_rate": 2.8418187640089662e-05,
"loss": 0.2098,
"step": 3390
},
{
"epoch": 4.89,
"grad_norm": 3.679403066635132,
"learning_rate": 2.8378162023695165e-05,
"loss": 0.1831,
"step": 3395
},
{
"epoch": 4.89,
"grad_norm": 3.4778425693511963,
"learning_rate": 2.8338136407300676e-05,
"loss": 0.1985,
"step": 3400
},
{
"epoch": 4.9,
"grad_norm": 3.987553834915161,
"learning_rate": 2.829811079090618e-05,
"loss": 0.2251,
"step": 3405
},
{
"epoch": 4.91,
"grad_norm": 4.322044372558594,
"learning_rate": 2.825808517451169e-05,
"loss": 0.2069,
"step": 3410
},
{
"epoch": 4.92,
"grad_norm": 5.449688911437988,
"learning_rate": 2.8218059558117193e-05,
"loss": 0.1794,
"step": 3415
},
{
"epoch": 4.92,
"grad_norm": 3.6124987602233887,
"learning_rate": 2.8178033941722703e-05,
"loss": 0.2339,
"step": 3420
},
{
"epoch": 4.93,
"grad_norm": 3.3568663597106934,
"learning_rate": 2.8138008325328214e-05,
"loss": 0.1907,
"step": 3425
},
{
"epoch": 4.94,
"grad_norm": 4.301671028137207,
"learning_rate": 2.8097982708933717e-05,
"loss": 0.2468,
"step": 3430
},
{
"epoch": 4.94,
"grad_norm": 2.871671676635742,
"learning_rate": 2.8057957092539227e-05,
"loss": 0.1939,
"step": 3435
},
{
"epoch": 4.95,
"grad_norm": 3.9504876136779785,
"learning_rate": 2.801793147614473e-05,
"loss": 0.2247,
"step": 3440
},
{
"epoch": 4.96,
"grad_norm": 3.9967737197875977,
"learning_rate": 2.797790585975024e-05,
"loss": 0.2198,
"step": 3445
},
{
"epoch": 4.97,
"grad_norm": 3.5244204998016357,
"learning_rate": 2.7937880243355745e-05,
"loss": 0.2026,
"step": 3450
},
{
"epoch": 4.97,
"grad_norm": 3.6014513969421387,
"learning_rate": 2.7897854626961255e-05,
"loss": 0.2131,
"step": 3455
},
{
"epoch": 4.98,
"grad_norm": 4.934159278869629,
"learning_rate": 2.7857829010566765e-05,
"loss": 0.2254,
"step": 3460
},
{
"epoch": 4.99,
"grad_norm": 3.0008413791656494,
"learning_rate": 2.781780339417227e-05,
"loss": 0.2359,
"step": 3465
},
{
"epoch": 4.99,
"grad_norm": 4.055464267730713,
"learning_rate": 2.777777777777778e-05,
"loss": 0.2038,
"step": 3470
},
{
"epoch": 5.0,
"eval_accuracy": 0.951417004048583,
"eval_loss": 0.1312309056520462,
"eval_runtime": 26.8652,
"eval_samples_per_second": 367.761,
"eval_steps_per_second": 11.502,
"step": 3473
},
{
"epoch": 5.0,
"grad_norm": 3.999800205230713,
"learning_rate": 2.7737752161383286e-05,
"loss": 0.2119,
"step": 3475
},
{
"epoch": 5.01,
"grad_norm": 4.014290809631348,
"learning_rate": 2.7697726544988796e-05,
"loss": 0.1888,
"step": 3480
},
{
"epoch": 5.02,
"grad_norm": 3.3272271156311035,
"learning_rate": 2.76577009285943e-05,
"loss": 0.2187,
"step": 3485
},
{
"epoch": 5.02,
"grad_norm": 5.279153347015381,
"learning_rate": 2.761767531219981e-05,
"loss": 0.1644,
"step": 3490
},
{
"epoch": 5.03,
"grad_norm": 2.8314247131347656,
"learning_rate": 2.757764969580532e-05,
"loss": 0.1695,
"step": 3495
},
{
"epoch": 5.04,
"grad_norm": 3.674100637435913,
"learning_rate": 2.7537624079410824e-05,
"loss": 0.1971,
"step": 3500
},
{
"epoch": 5.04,
"grad_norm": 4.037543296813965,
"learning_rate": 2.7497598463016334e-05,
"loss": 0.145,
"step": 3505
},
{
"epoch": 5.05,
"grad_norm": 2.6359853744506836,
"learning_rate": 2.7457572846621838e-05,
"loss": 0.1859,
"step": 3510
},
{
"epoch": 5.06,
"grad_norm": 3.589770555496216,
"learning_rate": 2.7417547230227348e-05,
"loss": 0.2093,
"step": 3515
},
{
"epoch": 5.07,
"grad_norm": 2.143383741378784,
"learning_rate": 2.737752161383285e-05,
"loss": 0.1568,
"step": 3520
},
{
"epoch": 5.07,
"grad_norm": 2.837106227874756,
"learning_rate": 2.7337495997438362e-05,
"loss": 0.1999,
"step": 3525
},
{
"epoch": 5.08,
"grad_norm": 3.294853448867798,
"learning_rate": 2.7297470381043872e-05,
"loss": 0.1655,
"step": 3530
},
{
"epoch": 5.09,
"grad_norm": 4.975580215454102,
"learning_rate": 2.7257444764649376e-05,
"loss": 0.1859,
"step": 3535
},
{
"epoch": 5.1,
"grad_norm": 2.522923231124878,
"learning_rate": 2.7217419148254886e-05,
"loss": 0.1736,
"step": 3540
},
{
"epoch": 5.1,
"grad_norm": 4.344380855560303,
"learning_rate": 2.717739353186039e-05,
"loss": 0.1921,
"step": 3545
},
{
"epoch": 5.11,
"grad_norm": 2.803359270095825,
"learning_rate": 2.71373679154659e-05,
"loss": 0.199,
"step": 3550
},
{
"epoch": 5.12,
"grad_norm": 3.1321277618408203,
"learning_rate": 2.7097342299071403e-05,
"loss": 0.2111,
"step": 3555
},
{
"epoch": 5.12,
"grad_norm": 4.107348442077637,
"learning_rate": 2.7057316682676913e-05,
"loss": 0.1625,
"step": 3560
},
{
"epoch": 5.13,
"grad_norm": 3.226501703262329,
"learning_rate": 2.7017291066282424e-05,
"loss": 0.1757,
"step": 3565
},
{
"epoch": 5.14,
"grad_norm": 3.1203489303588867,
"learning_rate": 2.6977265449887927e-05,
"loss": 0.1898,
"step": 3570
},
{
"epoch": 5.15,
"grad_norm": 4.505608081817627,
"learning_rate": 2.6937239833493438e-05,
"loss": 0.2059,
"step": 3575
},
{
"epoch": 5.15,
"grad_norm": 3.804442882537842,
"learning_rate": 2.6897214217098944e-05,
"loss": 0.2217,
"step": 3580
},
{
"epoch": 5.16,
"grad_norm": 3.088012218475342,
"learning_rate": 2.685718860070445e-05,
"loss": 0.1662,
"step": 3585
},
{
"epoch": 5.17,
"grad_norm": 2.7884981632232666,
"learning_rate": 2.6817162984309958e-05,
"loss": 0.2105,
"step": 3590
},
{
"epoch": 5.17,
"grad_norm": 3.278127908706665,
"learning_rate": 2.677713736791547e-05,
"loss": 0.1687,
"step": 3595
},
{
"epoch": 5.18,
"grad_norm": 7.960609436035156,
"learning_rate": 2.6737111751520975e-05,
"loss": 0.1793,
"step": 3600
},
{
"epoch": 5.19,
"grad_norm": 3.4804646968841553,
"learning_rate": 2.6697086135126482e-05,
"loss": 0.175,
"step": 3605
},
{
"epoch": 5.2,
"grad_norm": 3.5590946674346924,
"learning_rate": 2.6657060518731993e-05,
"loss": 0.1993,
"step": 3610
},
{
"epoch": 5.2,
"grad_norm": 3.4853434562683105,
"learning_rate": 2.6617034902337496e-05,
"loss": 0.1666,
"step": 3615
},
{
"epoch": 5.21,
"grad_norm": 4.569425582885742,
"learning_rate": 2.6577009285943006e-05,
"loss": 0.1932,
"step": 3620
},
{
"epoch": 5.22,
"grad_norm": 3.9071567058563232,
"learning_rate": 2.653698366954851e-05,
"loss": 0.1899,
"step": 3625
},
{
"epoch": 5.22,
"grad_norm": 2.170576333999634,
"learning_rate": 2.649695805315402e-05,
"loss": 0.1427,
"step": 3630
},
{
"epoch": 5.23,
"grad_norm": 7.018930912017822,
"learning_rate": 2.645693243675953e-05,
"loss": 0.2429,
"step": 3635
},
{
"epoch": 5.24,
"grad_norm": 4.312514305114746,
"learning_rate": 2.6416906820365034e-05,
"loss": 0.1731,
"step": 3640
},
{
"epoch": 5.25,
"grad_norm": 3.58902907371521,
"learning_rate": 2.6376881203970544e-05,
"loss": 0.1826,
"step": 3645
},
{
"epoch": 5.25,
"grad_norm": 4.274717807769775,
"learning_rate": 2.6336855587576048e-05,
"loss": 0.1643,
"step": 3650
},
{
"epoch": 5.26,
"grad_norm": 5.12242317199707,
"learning_rate": 2.6296829971181558e-05,
"loss": 0.2335,
"step": 3655
},
{
"epoch": 5.27,
"grad_norm": 3.3458309173583984,
"learning_rate": 2.625680435478706e-05,
"loss": 0.1515,
"step": 3660
},
{
"epoch": 5.28,
"grad_norm": 3.759378433227539,
"learning_rate": 2.6216778738392572e-05,
"loss": 0.1848,
"step": 3665
},
{
"epoch": 5.28,
"grad_norm": 2.559863805770874,
"learning_rate": 2.6176753121998082e-05,
"loss": 0.1769,
"step": 3670
},
{
"epoch": 5.29,
"grad_norm": 2.732105016708374,
"learning_rate": 2.6136727505603586e-05,
"loss": 0.1529,
"step": 3675
},
{
"epoch": 5.3,
"grad_norm": 2.628984212875366,
"learning_rate": 2.6096701889209096e-05,
"loss": 0.1971,
"step": 3680
},
{
"epoch": 5.3,
"grad_norm": 4.4776716232299805,
"learning_rate": 2.60566762728146e-05,
"loss": 0.2038,
"step": 3685
},
{
"epoch": 5.31,
"grad_norm": 5.578450679779053,
"learning_rate": 2.601665065642011e-05,
"loss": 0.1579,
"step": 3690
},
{
"epoch": 5.32,
"grad_norm": 5.5884480476379395,
"learning_rate": 2.5976625040025617e-05,
"loss": 0.1939,
"step": 3695
},
{
"epoch": 5.33,
"grad_norm": 5.272316932678223,
"learning_rate": 2.5936599423631124e-05,
"loss": 0.1689,
"step": 3700
},
{
"epoch": 5.33,
"grad_norm": 3.2164971828460693,
"learning_rate": 2.5896573807236634e-05,
"loss": 0.1998,
"step": 3705
},
{
"epoch": 5.34,
"grad_norm": 2.0555613040924072,
"learning_rate": 2.585654819084214e-05,
"loss": 0.1782,
"step": 3710
},
{
"epoch": 5.35,
"grad_norm": 4.028440475463867,
"learning_rate": 2.5816522574447648e-05,
"loss": 0.2054,
"step": 3715
},
{
"epoch": 5.35,
"grad_norm": 2.9159467220306396,
"learning_rate": 2.5776496958053155e-05,
"loss": 0.1815,
"step": 3720
},
{
"epoch": 5.36,
"grad_norm": 4.195735454559326,
"learning_rate": 2.5736471341658665e-05,
"loss": 0.2067,
"step": 3725
},
{
"epoch": 5.37,
"grad_norm": 3.9031412601470947,
"learning_rate": 2.5696445725264168e-05,
"loss": 0.1815,
"step": 3730
},
{
"epoch": 5.38,
"grad_norm": 4.9545087814331055,
"learning_rate": 2.565642010886968e-05,
"loss": 0.1777,
"step": 3735
},
{
"epoch": 5.38,
"grad_norm": 3.0578715801239014,
"learning_rate": 2.561639449247519e-05,
"loss": 0.1842,
"step": 3740
},
{
"epoch": 5.39,
"grad_norm": 4.023972034454346,
"learning_rate": 2.5576368876080692e-05,
"loss": 0.1804,
"step": 3745
},
{
"epoch": 5.4,
"grad_norm": 4.069141387939453,
"learning_rate": 2.5536343259686203e-05,
"loss": 0.1774,
"step": 3750
},
{
"epoch": 5.4,
"grad_norm": 5.315836429595947,
"learning_rate": 2.5496317643291706e-05,
"loss": 0.2041,
"step": 3755
},
{
"epoch": 5.41,
"grad_norm": 2.586327075958252,
"learning_rate": 2.5456292026897216e-05,
"loss": 0.236,
"step": 3760
},
{
"epoch": 5.42,
"grad_norm": 4.58768367767334,
"learning_rate": 2.541626641050272e-05,
"loss": 0.1555,
"step": 3765
},
{
"epoch": 5.43,
"grad_norm": 3.5733981132507324,
"learning_rate": 2.537624079410823e-05,
"loss": 0.1699,
"step": 3770
},
{
"epoch": 5.43,
"grad_norm": 3.546565055847168,
"learning_rate": 2.533621517771374e-05,
"loss": 0.1761,
"step": 3775
},
{
"epoch": 5.44,
"grad_norm": 4.250943183898926,
"learning_rate": 2.5296189561319244e-05,
"loss": 0.2104,
"step": 3780
},
{
"epoch": 5.45,
"grad_norm": 3.210358142852783,
"learning_rate": 2.5256163944924754e-05,
"loss": 0.1998,
"step": 3785
},
{
"epoch": 5.46,
"grad_norm": 4.336333274841309,
"learning_rate": 2.5216138328530258e-05,
"loss": 0.2217,
"step": 3790
},
{
"epoch": 5.46,
"grad_norm": 3.3414406776428223,
"learning_rate": 2.5176112712135768e-05,
"loss": 0.1878,
"step": 3795
},
{
"epoch": 5.47,
"grad_norm": 3.6740903854370117,
"learning_rate": 2.513608709574127e-05,
"loss": 0.1686,
"step": 3800
},
{
"epoch": 5.48,
"grad_norm": 2.5579757690429688,
"learning_rate": 2.5096061479346782e-05,
"loss": 0.1725,
"step": 3805
},
{
"epoch": 5.48,
"grad_norm": 4.028023719787598,
"learning_rate": 2.5056035862952292e-05,
"loss": 0.2001,
"step": 3810
},
{
"epoch": 5.49,
"grad_norm": 3.197957992553711,
"learning_rate": 2.5016010246557796e-05,
"loss": 0.1868,
"step": 3815
},
{
"epoch": 5.5,
"grad_norm": 2.4677882194519043,
"learning_rate": 2.4975984630163306e-05,
"loss": 0.2052,
"step": 3820
},
{
"epoch": 5.51,
"grad_norm": 3.9607093334198,
"learning_rate": 2.4935959013768813e-05,
"loss": 0.2223,
"step": 3825
},
{
"epoch": 5.51,
"grad_norm": 3.7235755920410156,
"learning_rate": 2.489593339737432e-05,
"loss": 0.173,
"step": 3830
},
{
"epoch": 5.52,
"grad_norm": 4.570876598358154,
"learning_rate": 2.485590778097983e-05,
"loss": 0.195,
"step": 3835
},
{
"epoch": 5.53,
"grad_norm": 4.305147171020508,
"learning_rate": 2.4815882164585337e-05,
"loss": 0.1984,
"step": 3840
},
{
"epoch": 5.53,
"grad_norm": 3.5951242446899414,
"learning_rate": 2.4775856548190844e-05,
"loss": 0.1915,
"step": 3845
},
{
"epoch": 5.54,
"grad_norm": 4.068524360656738,
"learning_rate": 2.473583093179635e-05,
"loss": 0.2063,
"step": 3850
},
{
"epoch": 5.55,
"grad_norm": 4.295510292053223,
"learning_rate": 2.4695805315401858e-05,
"loss": 0.213,
"step": 3855
},
{
"epoch": 5.56,
"grad_norm": 4.255641937255859,
"learning_rate": 2.4655779699007368e-05,
"loss": 0.2068,
"step": 3860
},
{
"epoch": 5.56,
"grad_norm": 3.168423652648926,
"learning_rate": 2.4615754082612875e-05,
"loss": 0.2092,
"step": 3865
},
{
"epoch": 5.57,
"grad_norm": 2.8869216442108154,
"learning_rate": 2.4575728466218382e-05,
"loss": 0.1763,
"step": 3870
},
{
"epoch": 5.58,
"grad_norm": 3.74802565574646,
"learning_rate": 2.453570284982389e-05,
"loss": 0.1539,
"step": 3875
},
{
"epoch": 5.58,
"grad_norm": 4.2037034034729,
"learning_rate": 2.4495677233429396e-05,
"loss": 0.1764,
"step": 3880
},
{
"epoch": 5.59,
"grad_norm": 6.206282615661621,
"learning_rate": 2.4455651617034902e-05,
"loss": 0.2138,
"step": 3885
},
{
"epoch": 5.6,
"grad_norm": 2.233813524246216,
"learning_rate": 2.441562600064041e-05,
"loss": 0.1478,
"step": 3890
},
{
"epoch": 5.61,
"grad_norm": 5.018454074859619,
"learning_rate": 2.437560038424592e-05,
"loss": 0.166,
"step": 3895
},
{
"epoch": 5.61,
"grad_norm": 4.372547626495361,
"learning_rate": 2.4335574767851427e-05,
"loss": 0.1964,
"step": 3900
},
{
"epoch": 5.62,
"grad_norm": 5.0093488693237305,
"learning_rate": 2.4295549151456933e-05,
"loss": 0.208,
"step": 3905
},
{
"epoch": 5.63,
"grad_norm": 4.177137851715088,
"learning_rate": 2.425552353506244e-05,
"loss": 0.2036,
"step": 3910
},
{
"epoch": 5.64,
"grad_norm": 3.151520013809204,
"learning_rate": 2.4215497918667947e-05,
"loss": 0.1867,
"step": 3915
},
{
"epoch": 5.64,
"grad_norm": 2.9322078227996826,
"learning_rate": 2.4175472302273454e-05,
"loss": 0.1692,
"step": 3920
},
{
"epoch": 5.65,
"grad_norm": 3.5823426246643066,
"learning_rate": 2.413544668587896e-05,
"loss": 0.2218,
"step": 3925
},
{
"epoch": 5.66,
"grad_norm": 3.663548707962036,
"learning_rate": 2.409542106948447e-05,
"loss": 0.1654,
"step": 3930
},
{
"epoch": 5.66,
"grad_norm": 3.8358802795410156,
"learning_rate": 2.4055395453089978e-05,
"loss": 0.161,
"step": 3935
},
{
"epoch": 5.67,
"grad_norm": 3.567988157272339,
"learning_rate": 2.4015369836695485e-05,
"loss": 0.1757,
"step": 3940
},
{
"epoch": 5.68,
"grad_norm": 4.0143046379089355,
"learning_rate": 2.3975344220300992e-05,
"loss": 0.1715,
"step": 3945
},
{
"epoch": 5.69,
"grad_norm": 6.651194095611572,
"learning_rate": 2.3935318603906502e-05,
"loss": 0.191,
"step": 3950
},
{
"epoch": 5.69,
"grad_norm": 3.7898061275482178,
"learning_rate": 2.389529298751201e-05,
"loss": 0.2112,
"step": 3955
},
{
"epoch": 5.7,
"grad_norm": 2.6467232704162598,
"learning_rate": 2.3855267371117516e-05,
"loss": 0.1807,
"step": 3960
},
{
"epoch": 5.71,
"grad_norm": 3.334956407546997,
"learning_rate": 2.3815241754723026e-05,
"loss": 0.1995,
"step": 3965
},
{
"epoch": 5.71,
"grad_norm": 4.663839817047119,
"learning_rate": 2.3775216138328533e-05,
"loss": 0.1653,
"step": 3970
},
{
"epoch": 5.72,
"grad_norm": 3.8154709339141846,
"learning_rate": 2.373519052193404e-05,
"loss": 0.1893,
"step": 3975
},
{
"epoch": 5.73,
"grad_norm": 3.9679391384124756,
"learning_rate": 2.3695164905539547e-05,
"loss": 0.1995,
"step": 3980
},
{
"epoch": 5.74,
"grad_norm": 2.465738296508789,
"learning_rate": 2.3655139289145054e-05,
"loss": 0.1571,
"step": 3985
},
{
"epoch": 5.74,
"grad_norm": 3.7358932495117188,
"learning_rate": 2.361511367275056e-05,
"loss": 0.1911,
"step": 3990
},
{
"epoch": 5.75,
"grad_norm": 2.5874032974243164,
"learning_rate": 2.3575088056356068e-05,
"loss": 0.1535,
"step": 3995
},
{
"epoch": 5.76,
"grad_norm": 4.26051139831543,
"learning_rate": 2.3535062439961578e-05,
"loss": 0.1562,
"step": 4000
},
{
"epoch": 5.76,
"grad_norm": 3.7166473865509033,
"learning_rate": 2.3495036823567085e-05,
"loss": 0.1605,
"step": 4005
},
{
"epoch": 5.77,
"grad_norm": 2.865448474884033,
"learning_rate": 2.3455011207172592e-05,
"loss": 0.1836,
"step": 4010
},
{
"epoch": 5.78,
"grad_norm": 2.9740395545959473,
"learning_rate": 2.34149855907781e-05,
"loss": 0.1539,
"step": 4015
},
{
"epoch": 5.79,
"grad_norm": 2.9424655437469482,
"learning_rate": 2.3374959974383606e-05,
"loss": 0.1671,
"step": 4020
},
{
"epoch": 5.79,
"grad_norm": 2.8022515773773193,
"learning_rate": 2.3334934357989113e-05,
"loss": 0.1841,
"step": 4025
},
{
"epoch": 5.8,
"grad_norm": 3.5465009212493896,
"learning_rate": 2.329490874159462e-05,
"loss": 0.1789,
"step": 4030
},
{
"epoch": 5.81,
"grad_norm": 3.6574504375457764,
"learning_rate": 2.325488312520013e-05,
"loss": 0.2014,
"step": 4035
},
{
"epoch": 5.82,
"grad_norm": 3.3290579319000244,
"learning_rate": 2.3214857508805637e-05,
"loss": 0.1704,
"step": 4040
},
{
"epoch": 5.82,
"grad_norm": 3.483360767364502,
"learning_rate": 2.3174831892411144e-05,
"loss": 0.2339,
"step": 4045
},
{
"epoch": 5.83,
"grad_norm": 4.405252933502197,
"learning_rate": 2.313480627601665e-05,
"loss": 0.1826,
"step": 4050
},
{
"epoch": 5.84,
"grad_norm": 3.58305025100708,
"learning_rate": 2.3094780659622157e-05,
"loss": 0.1709,
"step": 4055
},
{
"epoch": 5.84,
"grad_norm": 3.658742904663086,
"learning_rate": 2.3054755043227668e-05,
"loss": 0.1889,
"step": 4060
},
{
"epoch": 5.85,
"grad_norm": 3.7958412170410156,
"learning_rate": 2.3014729426833175e-05,
"loss": 0.1402,
"step": 4065
},
{
"epoch": 5.86,
"grad_norm": 3.8586859703063965,
"learning_rate": 2.297470381043868e-05,
"loss": 0.1979,
"step": 4070
},
{
"epoch": 5.87,
"grad_norm": 3.769714117050171,
"learning_rate": 2.293467819404419e-05,
"loss": 0.2036,
"step": 4075
},
{
"epoch": 5.87,
"grad_norm": 4.616032600402832,
"learning_rate": 2.28946525776497e-05,
"loss": 0.2101,
"step": 4080
},
{
"epoch": 5.88,
"grad_norm": 4.014847755432129,
"learning_rate": 2.2854626961255205e-05,
"loss": 0.2255,
"step": 4085
},
{
"epoch": 5.89,
"grad_norm": 4.2372002601623535,
"learning_rate": 2.2814601344860712e-05,
"loss": 0.1891,
"step": 4090
},
{
"epoch": 5.89,
"grad_norm": 5.569272518157959,
"learning_rate": 2.277457572846622e-05,
"loss": 0.2282,
"step": 4095
},
{
"epoch": 5.9,
"grad_norm": 4.124386310577393,
"learning_rate": 2.2734550112071726e-05,
"loss": 0.2061,
"step": 4100
},
{
"epoch": 5.91,
"grad_norm": 3.636678457260132,
"learning_rate": 2.2694524495677236e-05,
"loss": 0.1935,
"step": 4105
},
{
"epoch": 5.92,
"grad_norm": 5.7811808586120605,
"learning_rate": 2.2654498879282743e-05,
"loss": 0.1831,
"step": 4110
},
{
"epoch": 5.92,
"grad_norm": 3.2394773960113525,
"learning_rate": 2.261447326288825e-05,
"loss": 0.1831,
"step": 4115
},
{
"epoch": 5.93,
"grad_norm": 3.8104443550109863,
"learning_rate": 2.2574447646493757e-05,
"loss": 0.2237,
"step": 4120
},
{
"epoch": 5.94,
"grad_norm": 3.458548069000244,
"learning_rate": 2.2534422030099264e-05,
"loss": 0.1785,
"step": 4125
},
{
"epoch": 5.94,
"grad_norm": 2.2402632236480713,
"learning_rate": 2.249439641370477e-05,
"loss": 0.1534,
"step": 4130
},
{
"epoch": 5.95,
"grad_norm": 4.9622392654418945,
"learning_rate": 2.2454370797310278e-05,
"loss": 0.2295,
"step": 4135
},
{
"epoch": 5.96,
"grad_norm": 2.6501126289367676,
"learning_rate": 2.2414345180915788e-05,
"loss": 0.1917,
"step": 4140
},
{
"epoch": 5.97,
"grad_norm": 2.041633367538452,
"learning_rate": 2.2374319564521295e-05,
"loss": 0.174,
"step": 4145
},
{
"epoch": 5.97,
"grad_norm": 3.905135154724121,
"learning_rate": 2.2334293948126802e-05,
"loss": 0.2369,
"step": 4150
},
{
"epoch": 5.98,
"grad_norm": 2.685920238494873,
"learning_rate": 2.229426833173231e-05,
"loss": 0.2065,
"step": 4155
},
{
"epoch": 5.99,
"grad_norm": 3.8590340614318848,
"learning_rate": 2.2254242715337816e-05,
"loss": 0.1832,
"step": 4160
},
{
"epoch": 5.99,
"grad_norm": 3.912101984024048,
"learning_rate": 2.2214217098943323e-05,
"loss": 0.1793,
"step": 4165
},
{
"epoch": 6.0,
"eval_accuracy": 0.9522267206477733,
"eval_loss": 0.12901896238327026,
"eval_runtime": 26.8108,
"eval_samples_per_second": 368.508,
"eval_steps_per_second": 11.525,
"step": 4168
},
{
"epoch": 6.0,
"grad_norm": 3.0803754329681396,
"learning_rate": 2.217419148254883e-05,
"loss": 0.1826,
"step": 4170
},
{
"epoch": 6.01,
"grad_norm": 3.364600896835327,
"learning_rate": 2.213416586615434e-05,
"loss": 0.187,
"step": 4175
},
{
"epoch": 6.02,
"grad_norm": 3.815800428390503,
"learning_rate": 2.2094140249759847e-05,
"loss": 0.1836,
"step": 4180
},
{
"epoch": 6.02,
"grad_norm": 2.7681992053985596,
"learning_rate": 2.2054114633365357e-05,
"loss": 0.154,
"step": 4185
},
{
"epoch": 6.03,
"grad_norm": 3.0132570266723633,
"learning_rate": 2.2014089016970864e-05,
"loss": 0.1915,
"step": 4190
},
{
"epoch": 6.04,
"grad_norm": 5.308042526245117,
"learning_rate": 2.197406340057637e-05,
"loss": 0.1596,
"step": 4195
},
{
"epoch": 6.05,
"grad_norm": 4.3803863525390625,
"learning_rate": 2.1934037784181878e-05,
"loss": 0.2031,
"step": 4200
},
{
"epoch": 6.05,
"grad_norm": 3.398024082183838,
"learning_rate": 2.1894012167787385e-05,
"loss": 0.1732,
"step": 4205
},
{
"epoch": 6.06,
"grad_norm": 2.77665114402771,
"learning_rate": 2.1853986551392895e-05,
"loss": 0.1978,
"step": 4210
},
{
"epoch": 6.07,
"grad_norm": 6.4980974197387695,
"learning_rate": 2.1813960934998402e-05,
"loss": 0.1928,
"step": 4215
},
{
"epoch": 6.07,
"grad_norm": 5.6872100830078125,
"learning_rate": 2.177393531860391e-05,
"loss": 0.2128,
"step": 4220
},
{
"epoch": 6.08,
"grad_norm": 3.19380784034729,
"learning_rate": 2.1733909702209416e-05,
"loss": 0.1975,
"step": 4225
},
{
"epoch": 6.09,
"grad_norm": 2.491637945175171,
"learning_rate": 2.1693884085814922e-05,
"loss": 0.1662,
"step": 4230
},
{
"epoch": 6.1,
"grad_norm": 4.6479716300964355,
"learning_rate": 2.165385846942043e-05,
"loss": 0.1598,
"step": 4235
},
{
"epoch": 6.1,
"grad_norm": 3.320585250854492,
"learning_rate": 2.1613832853025936e-05,
"loss": 0.1476,
"step": 4240
},
{
"epoch": 6.11,
"grad_norm": 4.611423015594482,
"learning_rate": 2.1573807236631447e-05,
"loss": 0.1897,
"step": 4245
},
{
"epoch": 6.12,
"grad_norm": 3.57256817817688,
"learning_rate": 2.1533781620236953e-05,
"loss": 0.1562,
"step": 4250
},
{
"epoch": 6.12,
"grad_norm": 5.594003200531006,
"learning_rate": 2.149375600384246e-05,
"loss": 0.2314,
"step": 4255
},
{
"epoch": 6.13,
"grad_norm": 4.315159797668457,
"learning_rate": 2.1453730387447967e-05,
"loss": 0.1761,
"step": 4260
},
{
"epoch": 6.14,
"grad_norm": 4.169220924377441,
"learning_rate": 2.1413704771053474e-05,
"loss": 0.188,
"step": 4265
},
{
"epoch": 6.15,
"grad_norm": 7.287806510925293,
"learning_rate": 2.137367915465898e-05,
"loss": 0.2059,
"step": 4270
},
{
"epoch": 6.15,
"grad_norm": 3.8299801349639893,
"learning_rate": 2.1333653538264488e-05,
"loss": 0.1499,
"step": 4275
},
{
"epoch": 6.16,
"grad_norm": 3.4622209072113037,
"learning_rate": 2.1293627921869998e-05,
"loss": 0.1913,
"step": 4280
},
{
"epoch": 6.17,
"grad_norm": 4.407230377197266,
"learning_rate": 2.1253602305475505e-05,
"loss": 0.1673,
"step": 4285
},
{
"epoch": 6.17,
"grad_norm": 3.8075199127197266,
"learning_rate": 2.1213576689081012e-05,
"loss": 0.1341,
"step": 4290
},
{
"epoch": 6.18,
"grad_norm": 2.8441085815429688,
"learning_rate": 2.117355107268652e-05,
"loss": 0.1937,
"step": 4295
},
{
"epoch": 6.19,
"grad_norm": 3.905503034591675,
"learning_rate": 2.113352545629203e-05,
"loss": 0.1853,
"step": 4300
},
{
"epoch": 6.2,
"grad_norm": 2.487541675567627,
"learning_rate": 2.1093499839897536e-05,
"loss": 0.1671,
"step": 4305
},
{
"epoch": 6.2,
"grad_norm": 4.602748870849609,
"learning_rate": 2.1053474223503043e-05,
"loss": 0.1677,
"step": 4310
},
{
"epoch": 6.21,
"grad_norm": 3.4200439453125,
"learning_rate": 2.1013448607108553e-05,
"loss": 0.169,
"step": 4315
},
{
"epoch": 6.22,
"grad_norm": 3.3340063095092773,
"learning_rate": 2.097342299071406e-05,
"loss": 0.1686,
"step": 4320
},
{
"epoch": 6.23,
"grad_norm": 6.540536403656006,
"learning_rate": 2.0933397374319567e-05,
"loss": 0.1712,
"step": 4325
},
{
"epoch": 6.23,
"grad_norm": 3.8900535106658936,
"learning_rate": 2.0893371757925074e-05,
"loss": 0.1524,
"step": 4330
},
{
"epoch": 6.24,
"grad_norm": 3.2148051261901855,
"learning_rate": 2.085334614153058e-05,
"loss": 0.1393,
"step": 4335
},
{
"epoch": 6.25,
"grad_norm": 5.575131416320801,
"learning_rate": 2.0813320525136088e-05,
"loss": 0.2153,
"step": 4340
},
{
"epoch": 6.25,
"grad_norm": 3.013366222381592,
"learning_rate": 2.0773294908741595e-05,
"loss": 0.1949,
"step": 4345
},
{
"epoch": 6.26,
"grad_norm": 5.952768325805664,
"learning_rate": 2.0733269292347105e-05,
"loss": 0.2062,
"step": 4350
},
{
"epoch": 6.27,
"grad_norm": 2.8933584690093994,
"learning_rate": 2.0693243675952612e-05,
"loss": 0.1815,
"step": 4355
},
{
"epoch": 6.28,
"grad_norm": 4.637942790985107,
"learning_rate": 2.065321805955812e-05,
"loss": 0.1566,
"step": 4360
},
{
"epoch": 6.28,
"grad_norm": 3.233267068862915,
"learning_rate": 2.0613192443163626e-05,
"loss": 0.1657,
"step": 4365
},
{
"epoch": 6.29,
"grad_norm": 3.421495199203491,
"learning_rate": 2.0573166826769133e-05,
"loss": 0.1737,
"step": 4370
},
{
"epoch": 6.3,
"grad_norm": 3.9024555683135986,
"learning_rate": 2.053314121037464e-05,
"loss": 0.1997,
"step": 4375
},
{
"epoch": 6.3,
"grad_norm": 3.0292460918426514,
"learning_rate": 2.0493115593980146e-05,
"loss": 0.1685,
"step": 4380
},
{
"epoch": 6.31,
"grad_norm": 3.764313220977783,
"learning_rate": 2.0453089977585657e-05,
"loss": 0.1926,
"step": 4385
},
{
"epoch": 6.32,
"grad_norm": 3.7476236820220947,
"learning_rate": 2.0413064361191164e-05,
"loss": 0.155,
"step": 4390
},
{
"epoch": 6.33,
"grad_norm": 3.0338127613067627,
"learning_rate": 2.037303874479667e-05,
"loss": 0.1698,
"step": 4395
},
{
"epoch": 6.33,
"grad_norm": 3.0882599353790283,
"learning_rate": 2.0333013128402177e-05,
"loss": 0.1895,
"step": 4400
},
{
"epoch": 6.34,
"grad_norm": 3.8048288822174072,
"learning_rate": 2.0292987512007684e-05,
"loss": 0.2199,
"step": 4405
},
{
"epoch": 6.35,
"grad_norm": 4.483154296875,
"learning_rate": 2.025296189561319e-05,
"loss": 0.1615,
"step": 4410
},
{
"epoch": 6.35,
"grad_norm": 2.9506173133850098,
"learning_rate": 2.02129362792187e-05,
"loss": 0.1447,
"step": 4415
},
{
"epoch": 6.36,
"grad_norm": 6.518076419830322,
"learning_rate": 2.017291066282421e-05,
"loss": 0.1792,
"step": 4420
},
{
"epoch": 6.37,
"grad_norm": 2.7018208503723145,
"learning_rate": 2.013288504642972e-05,
"loss": 0.1908,
"step": 4425
},
{
"epoch": 6.38,
"grad_norm": 3.325758457183838,
"learning_rate": 2.0092859430035225e-05,
"loss": 0.1714,
"step": 4430
},
{
"epoch": 6.38,
"grad_norm": 3.347137928009033,
"learning_rate": 2.0052833813640732e-05,
"loss": 0.1641,
"step": 4435
},
{
"epoch": 6.39,
"grad_norm": 3.994041681289673,
"learning_rate": 2.001280819724624e-05,
"loss": 0.1857,
"step": 4440
},
{
"epoch": 6.4,
"grad_norm": 4.458554744720459,
"learning_rate": 1.9972782580851746e-05,
"loss": 0.1762,
"step": 4445
},
{
"epoch": 6.41,
"grad_norm": 4.118882656097412,
"learning_rate": 1.9932756964457253e-05,
"loss": 0.2023,
"step": 4450
},
{
"epoch": 6.41,
"grad_norm": 3.061014175415039,
"learning_rate": 1.9892731348062763e-05,
"loss": 0.1874,
"step": 4455
},
{
"epoch": 6.42,
"grad_norm": 3.94010853767395,
"learning_rate": 1.985270573166827e-05,
"loss": 0.1926,
"step": 4460
},
{
"epoch": 6.43,
"grad_norm": 3.748819351196289,
"learning_rate": 1.9812680115273777e-05,
"loss": 0.1585,
"step": 4465
},
{
"epoch": 6.43,
"grad_norm": 4.36990213394165,
"learning_rate": 1.9772654498879284e-05,
"loss": 0.1674,
"step": 4470
},
{
"epoch": 6.44,
"grad_norm": 4.165234565734863,
"learning_rate": 1.973262888248479e-05,
"loss": 0.1858,
"step": 4475
},
{
"epoch": 6.45,
"grad_norm": 5.72090482711792,
"learning_rate": 1.9692603266090298e-05,
"loss": 0.1881,
"step": 4480
},
{
"epoch": 6.46,
"grad_norm": 3.430233955383301,
"learning_rate": 1.9652577649695805e-05,
"loss": 0.188,
"step": 4485
},
{
"epoch": 6.46,
"grad_norm": 4.649074554443359,
"learning_rate": 1.961255203330131e-05,
"loss": 0.1757,
"step": 4490
},
{
"epoch": 6.47,
"grad_norm": 4.929919242858887,
"learning_rate": 1.9572526416906822e-05,
"loss": 0.1574,
"step": 4495
},
{
"epoch": 6.48,
"grad_norm": 4.103848934173584,
"learning_rate": 1.953250080051233e-05,
"loss": 0.1382,
"step": 4500
},
{
"epoch": 6.48,
"grad_norm": 3.4986753463745117,
"learning_rate": 1.9492475184117836e-05,
"loss": 0.1716,
"step": 4505
},
{
"epoch": 6.49,
"grad_norm": 3.33921217918396,
"learning_rate": 1.9452449567723343e-05,
"loss": 0.1825,
"step": 4510
},
{
"epoch": 6.5,
"grad_norm": 3.54250431060791,
"learning_rate": 1.941242395132885e-05,
"loss": 0.2003,
"step": 4515
},
{
"epoch": 6.51,
"grad_norm": 4.161519527435303,
"learning_rate": 1.9372398334934356e-05,
"loss": 0.1651,
"step": 4520
},
{
"epoch": 6.51,
"grad_norm": 3.6889195442199707,
"learning_rate": 1.9332372718539867e-05,
"loss": 0.1684,
"step": 4525
},
{
"epoch": 6.52,
"grad_norm": 2.6100666522979736,
"learning_rate": 1.9292347102145374e-05,
"loss": 0.1705,
"step": 4530
},
{
"epoch": 6.53,
"grad_norm": 4.2136945724487305,
"learning_rate": 1.925232148575088e-05,
"loss": 0.1724,
"step": 4535
},
{
"epoch": 6.53,
"grad_norm": 3.131673812866211,
"learning_rate": 1.921229586935639e-05,
"loss": 0.1764,
"step": 4540
},
{
"epoch": 6.54,
"grad_norm": 3.179766893386841,
"learning_rate": 1.9172270252961898e-05,
"loss": 0.1691,
"step": 4545
},
{
"epoch": 6.55,
"grad_norm": 3.707369089126587,
"learning_rate": 1.9132244636567405e-05,
"loss": 0.1568,
"step": 4550
},
{
"epoch": 6.56,
"grad_norm": 4.39201545715332,
"learning_rate": 1.909221902017291e-05,
"loss": 0.1556,
"step": 4555
},
{
"epoch": 6.56,
"grad_norm": 4.148761749267578,
"learning_rate": 1.905219340377842e-05,
"loss": 0.1541,
"step": 4560
},
{
"epoch": 6.57,
"grad_norm": 3.5763509273529053,
"learning_rate": 1.901216778738393e-05,
"loss": 0.18,
"step": 4565
},
{
"epoch": 6.58,
"grad_norm": 3.047024726867676,
"learning_rate": 1.8972142170989436e-05,
"loss": 0.1967,
"step": 4570
},
{
"epoch": 6.59,
"grad_norm": 4.121464252471924,
"learning_rate": 1.8932116554594942e-05,
"loss": 0.1543,
"step": 4575
},
{
"epoch": 6.59,
"grad_norm": 2.1216299533843994,
"learning_rate": 1.889209093820045e-05,
"loss": 0.1663,
"step": 4580
},
{
"epoch": 6.6,
"grad_norm": 2.798755645751953,
"learning_rate": 1.8852065321805956e-05,
"loss": 0.1794,
"step": 4585
},
{
"epoch": 6.61,
"grad_norm": 3.9779279232025146,
"learning_rate": 1.8812039705411463e-05,
"loss": 0.2191,
"step": 4590
},
{
"epoch": 6.61,
"grad_norm": 3.798940658569336,
"learning_rate": 1.877201408901697e-05,
"loss": 0.1791,
"step": 4595
},
{
"epoch": 6.62,
"grad_norm": 4.1768951416015625,
"learning_rate": 1.873198847262248e-05,
"loss": 0.1853,
"step": 4600
},
{
"epoch": 6.63,
"grad_norm": 2.557591199874878,
"learning_rate": 1.8691962856227987e-05,
"loss": 0.1693,
"step": 4605
},
{
"epoch": 6.64,
"grad_norm": 3.179755687713623,
"learning_rate": 1.8651937239833494e-05,
"loss": 0.1703,
"step": 4610
},
{
"epoch": 6.64,
"grad_norm": 2.7072629928588867,
"learning_rate": 1.8611911623439e-05,
"loss": 0.1592,
"step": 4615
},
{
"epoch": 6.65,
"grad_norm": 4.008791923522949,
"learning_rate": 1.8571886007044508e-05,
"loss": 0.1781,
"step": 4620
},
{
"epoch": 6.66,
"grad_norm": 3.89431095123291,
"learning_rate": 1.8531860390650015e-05,
"loss": 0.1973,
"step": 4625
},
{
"epoch": 6.66,
"grad_norm": 3.258575201034546,
"learning_rate": 1.8491834774255522e-05,
"loss": 0.143,
"step": 4630
},
{
"epoch": 6.67,
"grad_norm": 6.140786647796631,
"learning_rate": 1.8451809157861032e-05,
"loss": 0.1682,
"step": 4635
},
{
"epoch": 6.68,
"grad_norm": 2.957918405532837,
"learning_rate": 1.841178354146654e-05,
"loss": 0.1977,
"step": 4640
},
{
"epoch": 6.69,
"grad_norm": 4.5869574546813965,
"learning_rate": 1.8371757925072046e-05,
"loss": 0.1443,
"step": 4645
},
{
"epoch": 6.69,
"grad_norm": 2.5755558013916016,
"learning_rate": 1.8331732308677556e-05,
"loss": 0.2153,
"step": 4650
},
{
"epoch": 6.7,
"grad_norm": 3.475339889526367,
"learning_rate": 1.8291706692283063e-05,
"loss": 0.1443,
"step": 4655
},
{
"epoch": 6.71,
"grad_norm": 2.79864501953125,
"learning_rate": 1.825168107588857e-05,
"loss": 0.1564,
"step": 4660
},
{
"epoch": 6.71,
"grad_norm": 2.523737668991089,
"learning_rate": 1.8211655459494077e-05,
"loss": 0.1609,
"step": 4665
},
{
"epoch": 6.72,
"grad_norm": 4.166416645050049,
"learning_rate": 1.8171629843099587e-05,
"loss": 0.1926,
"step": 4670
},
{
"epoch": 6.73,
"grad_norm": 3.4169209003448486,
"learning_rate": 1.8131604226705094e-05,
"loss": 0.1841,
"step": 4675
},
{
"epoch": 6.74,
"grad_norm": 4.482958793640137,
"learning_rate": 1.80915786103106e-05,
"loss": 0.1699,
"step": 4680
},
{
"epoch": 6.74,
"grad_norm": 3.656352996826172,
"learning_rate": 1.8051552993916108e-05,
"loss": 0.1883,
"step": 4685
},
{
"epoch": 6.75,
"grad_norm": 3.9258341789245605,
"learning_rate": 1.8011527377521615e-05,
"loss": 0.1847,
"step": 4690
},
{
"epoch": 6.76,
"grad_norm": 3.6343822479248047,
"learning_rate": 1.797150176112712e-05,
"loss": 0.2101,
"step": 4695
},
{
"epoch": 6.77,
"grad_norm": 4.363142013549805,
"learning_rate": 1.793147614473263e-05,
"loss": 0.1752,
"step": 4700
},
{
"epoch": 6.77,
"grad_norm": 3.392268419265747,
"learning_rate": 1.789145052833814e-05,
"loss": 0.1807,
"step": 4705
},
{
"epoch": 6.78,
"grad_norm": 2.219773054122925,
"learning_rate": 1.7851424911943646e-05,
"loss": 0.1176,
"step": 4710
},
{
"epoch": 6.79,
"grad_norm": 2.391719102859497,
"learning_rate": 1.7811399295549153e-05,
"loss": 0.1475,
"step": 4715
},
{
"epoch": 6.79,
"grad_norm": 3.7803821563720703,
"learning_rate": 1.777137367915466e-05,
"loss": 0.179,
"step": 4720
},
{
"epoch": 6.8,
"grad_norm": 4.069080352783203,
"learning_rate": 1.7731348062760166e-05,
"loss": 0.199,
"step": 4725
},
{
"epoch": 6.81,
"grad_norm": 2.244948387145996,
"learning_rate": 1.7691322446365673e-05,
"loss": 0.15,
"step": 4730
},
{
"epoch": 6.82,
"grad_norm": 2.642184019088745,
"learning_rate": 1.765129682997118e-05,
"loss": 0.1525,
"step": 4735
},
{
"epoch": 6.82,
"grad_norm": 3.719548463821411,
"learning_rate": 1.761127121357669e-05,
"loss": 0.1744,
"step": 4740
},
{
"epoch": 6.83,
"grad_norm": 4.889650821685791,
"learning_rate": 1.7571245597182197e-05,
"loss": 0.1936,
"step": 4745
},
{
"epoch": 6.84,
"grad_norm": 4.086753845214844,
"learning_rate": 1.7531219980787704e-05,
"loss": 0.1745,
"step": 4750
},
{
"epoch": 6.84,
"grad_norm": 3.039050817489624,
"learning_rate": 1.749119436439321e-05,
"loss": 0.1442,
"step": 4755
},
{
"epoch": 6.85,
"grad_norm": 3.0932884216308594,
"learning_rate": 1.7451168747998718e-05,
"loss": 0.1575,
"step": 4760
},
{
"epoch": 6.86,
"grad_norm": 14.424320220947266,
"learning_rate": 1.741114313160423e-05,
"loss": 0.1479,
"step": 4765
},
{
"epoch": 6.87,
"grad_norm": 6.001073837280273,
"learning_rate": 1.7371117515209735e-05,
"loss": 0.1448,
"step": 4770
},
{
"epoch": 6.87,
"grad_norm": 4.473045349121094,
"learning_rate": 1.7331091898815242e-05,
"loss": 0.1468,
"step": 4775
},
{
"epoch": 6.88,
"grad_norm": 3.3458259105682373,
"learning_rate": 1.7291066282420752e-05,
"loss": 0.142,
"step": 4780
},
{
"epoch": 6.89,
"grad_norm": 4.793277263641357,
"learning_rate": 1.725104066602626e-05,
"loss": 0.1837,
"step": 4785
},
{
"epoch": 6.89,
"grad_norm": 3.12072491645813,
"learning_rate": 1.7211015049631766e-05,
"loss": 0.2176,
"step": 4790
},
{
"epoch": 6.9,
"grad_norm": 2.7718887329101562,
"learning_rate": 1.7170989433237273e-05,
"loss": 0.1801,
"step": 4795
},
{
"epoch": 6.91,
"grad_norm": 4.768100738525391,
"learning_rate": 1.713096381684278e-05,
"loss": 0.2163,
"step": 4800
},
{
"epoch": 6.92,
"grad_norm": 3.0787980556488037,
"learning_rate": 1.7090938200448287e-05,
"loss": 0.1444,
"step": 4805
},
{
"epoch": 6.92,
"grad_norm": 4.640768051147461,
"learning_rate": 1.7050912584053797e-05,
"loss": 0.1869,
"step": 4810
},
{
"epoch": 6.93,
"grad_norm": 2.0733461380004883,
"learning_rate": 1.7010886967659304e-05,
"loss": 0.1588,
"step": 4815
},
{
"epoch": 6.94,
"grad_norm": 4.384278774261475,
"learning_rate": 1.697086135126481e-05,
"loss": 0.195,
"step": 4820
},
{
"epoch": 6.94,
"grad_norm": 4.83052396774292,
"learning_rate": 1.6930835734870318e-05,
"loss": 0.1871,
"step": 4825
},
{
"epoch": 6.95,
"grad_norm": 3.9045228958129883,
"learning_rate": 1.6890810118475825e-05,
"loss": 0.1829,
"step": 4830
},
{
"epoch": 6.96,
"grad_norm": 2.1931328773498535,
"learning_rate": 1.685078450208133e-05,
"loss": 0.1697,
"step": 4835
},
{
"epoch": 6.97,
"grad_norm": 5.390465259552002,
"learning_rate": 1.681075888568684e-05,
"loss": 0.1787,
"step": 4840
},
{
"epoch": 6.97,
"grad_norm": 2.7756080627441406,
"learning_rate": 1.677073326929235e-05,
"loss": 0.1668,
"step": 4845
},
{
"epoch": 6.98,
"grad_norm": 3.4794211387634277,
"learning_rate": 1.6730707652897856e-05,
"loss": 0.1443,
"step": 4850
},
{
"epoch": 6.99,
"grad_norm": 2.9568474292755127,
"learning_rate": 1.6690682036503363e-05,
"loss": 0.124,
"step": 4855
},
{
"epoch": 7.0,
"grad_norm": 2.9625253677368164,
"learning_rate": 1.665065642010887e-05,
"loss": 0.19,
"step": 4860
},
{
"epoch": 7.0,
"eval_accuracy": 0.9533400809716599,
"eval_loss": 0.13323618471622467,
"eval_runtime": 26.9803,
"eval_samples_per_second": 366.194,
"eval_steps_per_second": 11.453,
"step": 4863
},
{
"epoch": 7.0,
"grad_norm": 3.4989819526672363,
"learning_rate": 1.6610630803714376e-05,
"loss": 0.1874,
"step": 4865
},
{
"epoch": 7.01,
"grad_norm": 3.2885279655456543,
"learning_rate": 1.6570605187319883e-05,
"loss": 0.1561,
"step": 4870
},
{
"epoch": 7.02,
"grad_norm": 3.600435972213745,
"learning_rate": 1.653057957092539e-05,
"loss": 0.1705,
"step": 4875
},
{
"epoch": 7.02,
"grad_norm": 4.295845031738281,
"learning_rate": 1.64905539545309e-05,
"loss": 0.1646,
"step": 4880
},
{
"epoch": 7.03,
"grad_norm": 2.21591854095459,
"learning_rate": 1.6450528338136407e-05,
"loss": 0.1434,
"step": 4885
},
{
"epoch": 7.04,
"grad_norm": 3.466616153717041,
"learning_rate": 1.6410502721741918e-05,
"loss": 0.1694,
"step": 4890
},
{
"epoch": 7.05,
"grad_norm": 3.914956569671631,
"learning_rate": 1.6370477105347425e-05,
"loss": 0.1535,
"step": 4895
},
{
"epoch": 7.05,
"grad_norm": 4.452558994293213,
"learning_rate": 1.633045148895293e-05,
"loss": 0.1463,
"step": 4900
},
{
"epoch": 7.06,
"grad_norm": 3.224085807800293,
"learning_rate": 1.629042587255844e-05,
"loss": 0.1182,
"step": 4905
},
{
"epoch": 7.07,
"grad_norm": 3.1381661891937256,
"learning_rate": 1.6250400256163945e-05,
"loss": 0.1627,
"step": 4910
},
{
"epoch": 7.07,
"grad_norm": 3.2462520599365234,
"learning_rate": 1.6210374639769456e-05,
"loss": 0.1611,
"step": 4915
},
{
"epoch": 7.08,
"grad_norm": 2.3595752716064453,
"learning_rate": 1.6170349023374962e-05,
"loss": 0.1383,
"step": 4920
},
{
"epoch": 7.09,
"grad_norm": 2.8666834831237793,
"learning_rate": 1.613032340698047e-05,
"loss": 0.1735,
"step": 4925
},
{
"epoch": 7.1,
"grad_norm": 4.893600940704346,
"learning_rate": 1.6090297790585976e-05,
"loss": 0.1909,
"step": 4930
},
{
"epoch": 7.1,
"grad_norm": 3.570570230484009,
"learning_rate": 1.6050272174191483e-05,
"loss": 0.1695,
"step": 4935
},
{
"epoch": 7.11,
"grad_norm": 2.665599822998047,
"learning_rate": 1.601024655779699e-05,
"loss": 0.1534,
"step": 4940
},
{
"epoch": 7.12,
"grad_norm": 5.707326412200928,
"learning_rate": 1.5970220941402497e-05,
"loss": 0.1687,
"step": 4945
},
{
"epoch": 7.12,
"grad_norm": 3.2267582416534424,
"learning_rate": 1.5930195325008007e-05,
"loss": 0.16,
"step": 4950
},
{
"epoch": 7.13,
"grad_norm": 5.348135948181152,
"learning_rate": 1.5890169708613514e-05,
"loss": 0.1575,
"step": 4955
},
{
"epoch": 7.14,
"grad_norm": 2.868098497390747,
"learning_rate": 1.585014409221902e-05,
"loss": 0.1875,
"step": 4960
},
{
"epoch": 7.15,
"grad_norm": 4.623507976531982,
"learning_rate": 1.5810118475824528e-05,
"loss": 0.2132,
"step": 4965
},
{
"epoch": 7.15,
"grad_norm": 3.894582509994507,
"learning_rate": 1.5770092859430035e-05,
"loss": 0.202,
"step": 4970
},
{
"epoch": 7.16,
"grad_norm": 4.989294052124023,
"learning_rate": 1.5730067243035542e-05,
"loss": 0.1773,
"step": 4975
},
{
"epoch": 7.17,
"grad_norm": 2.993002414703369,
"learning_rate": 1.569004162664105e-05,
"loss": 0.1961,
"step": 4980
},
{
"epoch": 7.18,
"grad_norm": 3.3114404678344727,
"learning_rate": 1.565001601024656e-05,
"loss": 0.1341,
"step": 4985
},
{
"epoch": 7.18,
"grad_norm": 4.713860988616943,
"learning_rate": 1.5609990393852066e-05,
"loss": 0.1367,
"step": 4990
},
{
"epoch": 7.19,
"grad_norm": 3.016030788421631,
"learning_rate": 1.5569964777457573e-05,
"loss": 0.1768,
"step": 4995
},
{
"epoch": 7.2,
"grad_norm": 3.965454578399658,
"learning_rate": 1.552993916106308e-05,
"loss": 0.1454,
"step": 5000
},
{
"epoch": 7.2,
"grad_norm": 2.214106798171997,
"learning_rate": 1.548991354466859e-05,
"loss": 0.1392,
"step": 5005
},
{
"epoch": 7.21,
"grad_norm": 3.117654800415039,
"learning_rate": 1.5449887928274097e-05,
"loss": 0.1716,
"step": 5010
},
{
"epoch": 7.22,
"grad_norm": 4.0507493019104,
"learning_rate": 1.5409862311879604e-05,
"loss": 0.175,
"step": 5015
},
{
"epoch": 7.23,
"grad_norm": 4.791309356689453,
"learning_rate": 1.5369836695485114e-05,
"loss": 0.1885,
"step": 5020
},
{
"epoch": 7.23,
"grad_norm": 4.158595561981201,
"learning_rate": 1.532981107909062e-05,
"loss": 0.1712,
"step": 5025
},
{
"epoch": 7.24,
"grad_norm": 3.812695264816284,
"learning_rate": 1.5289785462696128e-05,
"loss": 0.1444,
"step": 5030
},
{
"epoch": 7.25,
"grad_norm": 3.045092821121216,
"learning_rate": 1.5249759846301635e-05,
"loss": 0.183,
"step": 5035
},
{
"epoch": 7.25,
"grad_norm": 2.478226661682129,
"learning_rate": 1.5209734229907142e-05,
"loss": 0.1099,
"step": 5040
},
{
"epoch": 7.26,
"grad_norm": 5.668630599975586,
"learning_rate": 1.5169708613512648e-05,
"loss": 0.1812,
"step": 5045
},
{
"epoch": 7.27,
"grad_norm": 4.244511604309082,
"learning_rate": 1.5129682997118155e-05,
"loss": 0.1864,
"step": 5050
},
{
"epoch": 7.28,
"grad_norm": 2.8045542240142822,
"learning_rate": 1.5089657380723666e-05,
"loss": 0.1713,
"step": 5055
},
{
"epoch": 7.28,
"grad_norm": 2.768808126449585,
"learning_rate": 1.5049631764329173e-05,
"loss": 0.1761,
"step": 5060
},
{
"epoch": 7.29,
"grad_norm": 3.726841449737549,
"learning_rate": 1.500960614793468e-05,
"loss": 0.1605,
"step": 5065
},
{
"epoch": 7.3,
"grad_norm": 4.635730743408203,
"learning_rate": 1.4969580531540186e-05,
"loss": 0.179,
"step": 5070
},
{
"epoch": 7.3,
"grad_norm": 3.83880877494812,
"learning_rate": 1.4929554915145693e-05,
"loss": 0.1506,
"step": 5075
},
{
"epoch": 7.31,
"grad_norm": 2.271193504333496,
"learning_rate": 1.48895292987512e-05,
"loss": 0.1327,
"step": 5080
},
{
"epoch": 7.32,
"grad_norm": 2.6322343349456787,
"learning_rate": 1.4849503682356709e-05,
"loss": 0.1543,
"step": 5085
},
{
"epoch": 7.33,
"grad_norm": 4.538561820983887,
"learning_rate": 1.4809478065962217e-05,
"loss": 0.1797,
"step": 5090
},
{
"epoch": 7.33,
"grad_norm": 2.668684720993042,
"learning_rate": 1.4769452449567724e-05,
"loss": 0.1601,
"step": 5095
},
{
"epoch": 7.34,
"grad_norm": 3.4693799018859863,
"learning_rate": 1.4729426833173233e-05,
"loss": 0.1713,
"step": 5100
},
{
"epoch": 7.35,
"grad_norm": 3.7429068088531494,
"learning_rate": 1.468940121677874e-05,
"loss": 0.1934,
"step": 5105
},
{
"epoch": 7.36,
"grad_norm": 2.9476771354675293,
"learning_rate": 1.4649375600384247e-05,
"loss": 0.1566,
"step": 5110
},
{
"epoch": 7.36,
"grad_norm": 3.614051103591919,
"learning_rate": 1.4609349983989754e-05,
"loss": 0.1706,
"step": 5115
},
{
"epoch": 7.37,
"grad_norm": 5.335232734680176,
"learning_rate": 1.456932436759526e-05,
"loss": 0.2092,
"step": 5120
},
{
"epoch": 7.38,
"grad_norm": 2.373945474624634,
"learning_rate": 1.452929875120077e-05,
"loss": 0.1533,
"step": 5125
},
{
"epoch": 7.38,
"grad_norm": 4.9412312507629395,
"learning_rate": 1.4489273134806278e-05,
"loss": 0.1823,
"step": 5130
},
{
"epoch": 7.39,
"grad_norm": 4.234364032745361,
"learning_rate": 1.4449247518411785e-05,
"loss": 0.1649,
"step": 5135
},
{
"epoch": 7.4,
"grad_norm": 3.623694896697998,
"learning_rate": 1.4409221902017291e-05,
"loss": 0.1746,
"step": 5140
},
{
"epoch": 7.41,
"grad_norm": 2.8488681316375732,
"learning_rate": 1.4369196285622798e-05,
"loss": 0.1589,
"step": 5145
},
{
"epoch": 7.41,
"grad_norm": 4.376282691955566,
"learning_rate": 1.4329170669228307e-05,
"loss": 0.1619,
"step": 5150
},
{
"epoch": 7.42,
"grad_norm": 4.9292426109313965,
"learning_rate": 1.4289145052833814e-05,
"loss": 0.1638,
"step": 5155
},
{
"epoch": 7.43,
"grad_norm": 5.3557658195495605,
"learning_rate": 1.4249119436439322e-05,
"loss": 0.1841,
"step": 5160
},
{
"epoch": 7.43,
"grad_norm": 5.3310394287109375,
"learning_rate": 1.4209093820044831e-05,
"loss": 0.1746,
"step": 5165
},
{
"epoch": 7.44,
"grad_norm": 4.013771057128906,
"learning_rate": 1.4169068203650338e-05,
"loss": 0.1853,
"step": 5170
},
{
"epoch": 7.45,
"grad_norm": 3.4680042266845703,
"learning_rate": 1.4129042587255845e-05,
"loss": 0.1796,
"step": 5175
},
{
"epoch": 7.46,
"grad_norm": 3.229072332382202,
"learning_rate": 1.4089016970861352e-05,
"loss": 0.1555,
"step": 5180
},
{
"epoch": 7.46,
"grad_norm": 2.8179924488067627,
"learning_rate": 1.4048991354466859e-05,
"loss": 0.1861,
"step": 5185
},
{
"epoch": 7.47,
"grad_norm": 2.785851001739502,
"learning_rate": 1.4008965738072365e-05,
"loss": 0.16,
"step": 5190
},
{
"epoch": 7.48,
"grad_norm": 2.7952377796173096,
"learning_rate": 1.3968940121677872e-05,
"loss": 0.1587,
"step": 5195
},
{
"epoch": 7.48,
"grad_norm": 4.441540241241455,
"learning_rate": 1.3928914505283383e-05,
"loss": 0.1741,
"step": 5200
},
{
"epoch": 7.49,
"grad_norm": 4.560239791870117,
"learning_rate": 1.388888888888889e-05,
"loss": 0.1657,
"step": 5205
},
{
"epoch": 7.5,
"grad_norm": 2.9085419178009033,
"learning_rate": 1.3848863272494398e-05,
"loss": 0.1391,
"step": 5210
},
{
"epoch": 7.51,
"grad_norm": 2.2866363525390625,
"learning_rate": 1.3808837656099905e-05,
"loss": 0.1297,
"step": 5215
},
{
"epoch": 7.51,
"grad_norm": 3.377863645553589,
"learning_rate": 1.3768812039705412e-05,
"loss": 0.1747,
"step": 5220
},
{
"epoch": 7.52,
"grad_norm": 2.285461187362671,
"learning_rate": 1.3728786423310919e-05,
"loss": 0.1473,
"step": 5225
},
{
"epoch": 7.53,
"grad_norm": 3.6761443614959717,
"learning_rate": 1.3688760806916426e-05,
"loss": 0.1518,
"step": 5230
},
{
"epoch": 7.54,
"grad_norm": 2.0515265464782715,
"learning_rate": 1.3648735190521936e-05,
"loss": 0.1755,
"step": 5235
},
{
"epoch": 7.54,
"grad_norm": 3.5392673015594482,
"learning_rate": 1.3608709574127443e-05,
"loss": 0.1687,
"step": 5240
},
{
"epoch": 7.55,
"grad_norm": 4.235048770904541,
"learning_rate": 1.356868395773295e-05,
"loss": 0.159,
"step": 5245
},
{
"epoch": 7.56,
"grad_norm": 3.9819908142089844,
"learning_rate": 1.3528658341338457e-05,
"loss": 0.1388,
"step": 5250
},
{
"epoch": 7.56,
"grad_norm": 4.2567524909973145,
"learning_rate": 1.3488632724943964e-05,
"loss": 0.1855,
"step": 5255
},
{
"epoch": 7.57,
"grad_norm": 3.7046706676483154,
"learning_rate": 1.3448607108549472e-05,
"loss": 0.1705,
"step": 5260
},
{
"epoch": 7.58,
"grad_norm": 5.374367713928223,
"learning_rate": 1.3408581492154979e-05,
"loss": 0.1558,
"step": 5265
},
{
"epoch": 7.59,
"grad_norm": 3.5017898082733154,
"learning_rate": 1.3368555875760488e-05,
"loss": 0.1665,
"step": 5270
},
{
"epoch": 7.59,
"grad_norm": 2.907083034515381,
"learning_rate": 1.3328530259365996e-05,
"loss": 0.1407,
"step": 5275
},
{
"epoch": 7.6,
"grad_norm": 3.715378522872925,
"learning_rate": 1.3288504642971503e-05,
"loss": 0.1519,
"step": 5280
},
{
"epoch": 7.61,
"grad_norm": 5.481036186218262,
"learning_rate": 1.324847902657701e-05,
"loss": 0.1559,
"step": 5285
},
{
"epoch": 7.61,
"grad_norm": 2.949070692062378,
"learning_rate": 1.3208453410182517e-05,
"loss": 0.1673,
"step": 5290
},
{
"epoch": 7.62,
"grad_norm": 2.929731845855713,
"learning_rate": 1.3168427793788024e-05,
"loss": 0.1592,
"step": 5295
},
{
"epoch": 7.63,
"grad_norm": 4.555692672729492,
"learning_rate": 1.312840217739353e-05,
"loss": 0.2078,
"step": 5300
},
{
"epoch": 7.64,
"grad_norm": 3.130526065826416,
"learning_rate": 1.3088376560999041e-05,
"loss": 0.1381,
"step": 5305
},
{
"epoch": 7.64,
"grad_norm": 4.921693325042725,
"learning_rate": 1.3048350944604548e-05,
"loss": 0.2067,
"step": 5310
},
{
"epoch": 7.65,
"grad_norm": 3.4057352542877197,
"learning_rate": 1.3008325328210055e-05,
"loss": 0.1232,
"step": 5315
},
{
"epoch": 7.66,
"grad_norm": 6.398148536682129,
"learning_rate": 1.2968299711815562e-05,
"loss": 0.1854,
"step": 5320
},
{
"epoch": 7.66,
"grad_norm": 4.1279754638671875,
"learning_rate": 1.292827409542107e-05,
"loss": 0.1674,
"step": 5325
},
{
"epoch": 7.67,
"grad_norm": 3.3673510551452637,
"learning_rate": 1.2888248479026577e-05,
"loss": 0.144,
"step": 5330
},
{
"epoch": 7.68,
"grad_norm": 3.9797935485839844,
"learning_rate": 1.2848222862632084e-05,
"loss": 0.1506,
"step": 5335
},
{
"epoch": 7.69,
"grad_norm": 4.889264106750488,
"learning_rate": 1.2808197246237594e-05,
"loss": 0.1409,
"step": 5340
},
{
"epoch": 7.69,
"grad_norm": 3.3350377082824707,
"learning_rate": 1.2768171629843101e-05,
"loss": 0.1267,
"step": 5345
},
{
"epoch": 7.7,
"grad_norm": 3.927999496459961,
"learning_rate": 1.2728146013448608e-05,
"loss": 0.1484,
"step": 5350
},
{
"epoch": 7.71,
"grad_norm": 3.547835350036621,
"learning_rate": 1.2688120397054115e-05,
"loss": 0.1444,
"step": 5355
},
{
"epoch": 7.72,
"grad_norm": 3.605868101119995,
"learning_rate": 1.2648094780659622e-05,
"loss": 0.1524,
"step": 5360
},
{
"epoch": 7.72,
"grad_norm": 4.846922397613525,
"learning_rate": 1.2608069164265129e-05,
"loss": 0.1815,
"step": 5365
},
{
"epoch": 7.73,
"grad_norm": 3.1063039302825928,
"learning_rate": 1.2568043547870636e-05,
"loss": 0.1288,
"step": 5370
},
{
"epoch": 7.74,
"grad_norm": 3.755476236343384,
"learning_rate": 1.2528017931476146e-05,
"loss": 0.1703,
"step": 5375
},
{
"epoch": 7.74,
"grad_norm": 2.933077096939087,
"learning_rate": 1.2487992315081653e-05,
"loss": 0.168,
"step": 5380
},
{
"epoch": 7.75,
"grad_norm": 3.11079478263855,
"learning_rate": 1.244796669868716e-05,
"loss": 0.161,
"step": 5385
},
{
"epoch": 7.76,
"grad_norm": 3.155428409576416,
"learning_rate": 1.2407941082292668e-05,
"loss": 0.1903,
"step": 5390
},
{
"epoch": 7.77,
"grad_norm": 4.022368907928467,
"learning_rate": 1.2367915465898175e-05,
"loss": 0.1941,
"step": 5395
},
{
"epoch": 7.77,
"grad_norm": 3.187999725341797,
"learning_rate": 1.2327889849503684e-05,
"loss": 0.1445,
"step": 5400
},
{
"epoch": 7.78,
"grad_norm": 4.614522457122803,
"learning_rate": 1.2287864233109191e-05,
"loss": 0.145,
"step": 5405
},
{
"epoch": 7.79,
"grad_norm": 4.4915266036987305,
"learning_rate": 1.2247838616714698e-05,
"loss": 0.1815,
"step": 5410
},
{
"epoch": 7.79,
"grad_norm": 3.4047083854675293,
"learning_rate": 1.2207813000320205e-05,
"loss": 0.1781,
"step": 5415
},
{
"epoch": 7.8,
"grad_norm": 3.843172550201416,
"learning_rate": 1.2167787383925713e-05,
"loss": 0.1601,
"step": 5420
},
{
"epoch": 7.81,
"grad_norm": 3.592343330383301,
"learning_rate": 1.212776176753122e-05,
"loss": 0.1948,
"step": 5425
},
{
"epoch": 7.82,
"grad_norm": 6.0892133712768555,
"learning_rate": 1.2087736151136727e-05,
"loss": 0.1562,
"step": 5430
},
{
"epoch": 7.82,
"grad_norm": 2.9761712551116943,
"learning_rate": 1.2047710534742236e-05,
"loss": 0.1584,
"step": 5435
},
{
"epoch": 7.83,
"grad_norm": 4.459796905517578,
"learning_rate": 1.2007684918347743e-05,
"loss": 0.1492,
"step": 5440
},
{
"epoch": 7.84,
"grad_norm": 3.517129421234131,
"learning_rate": 1.1967659301953251e-05,
"loss": 0.1892,
"step": 5445
},
{
"epoch": 7.84,
"grad_norm": 3.708636999130249,
"learning_rate": 1.1927633685558758e-05,
"loss": 0.1644,
"step": 5450
},
{
"epoch": 7.85,
"grad_norm": 4.543480396270752,
"learning_rate": 1.1887608069164267e-05,
"loss": 0.1459,
"step": 5455
},
{
"epoch": 7.86,
"grad_norm": 3.9120936393737793,
"learning_rate": 1.1847582452769774e-05,
"loss": 0.1597,
"step": 5460
},
{
"epoch": 7.87,
"grad_norm": 4.939571380615234,
"learning_rate": 1.180755683637528e-05,
"loss": 0.2059,
"step": 5465
},
{
"epoch": 7.87,
"grad_norm": 3.939115047454834,
"learning_rate": 1.1767531219980789e-05,
"loss": 0.16,
"step": 5470
},
{
"epoch": 7.88,
"grad_norm": 3.2744674682617188,
"learning_rate": 1.1727505603586296e-05,
"loss": 0.1793,
"step": 5475
},
{
"epoch": 7.89,
"grad_norm": 4.565367698669434,
"learning_rate": 1.1687479987191803e-05,
"loss": 0.1982,
"step": 5480
},
{
"epoch": 7.89,
"grad_norm": 5.799792289733887,
"learning_rate": 1.164745437079731e-05,
"loss": 0.1593,
"step": 5485
},
{
"epoch": 7.9,
"grad_norm": 2.6292014122009277,
"learning_rate": 1.1607428754402818e-05,
"loss": 0.1484,
"step": 5490
},
{
"epoch": 7.91,
"grad_norm": 4.108443260192871,
"learning_rate": 1.1567403138008325e-05,
"loss": 0.1535,
"step": 5495
},
{
"epoch": 7.92,
"grad_norm": 3.3221068382263184,
"learning_rate": 1.1527377521613834e-05,
"loss": 0.1611,
"step": 5500
},
{
"epoch": 7.92,
"grad_norm": 3.2103445529937744,
"learning_rate": 1.148735190521934e-05,
"loss": 0.1159,
"step": 5505
},
{
"epoch": 7.93,
"grad_norm": 4.046101093292236,
"learning_rate": 1.144732628882485e-05,
"loss": 0.1895,
"step": 5510
},
{
"epoch": 7.94,
"grad_norm": 5.981593608856201,
"learning_rate": 1.1407300672430356e-05,
"loss": 0.1748,
"step": 5515
},
{
"epoch": 7.95,
"grad_norm": 3.193052291870117,
"learning_rate": 1.1367275056035863e-05,
"loss": 0.1463,
"step": 5520
},
{
"epoch": 7.95,
"grad_norm": 4.645889759063721,
"learning_rate": 1.1327249439641372e-05,
"loss": 0.162,
"step": 5525
},
{
"epoch": 7.96,
"grad_norm": 3.167149543762207,
"learning_rate": 1.1287223823246879e-05,
"loss": 0.1663,
"step": 5530
},
{
"epoch": 7.97,
"grad_norm": 5.2093048095703125,
"learning_rate": 1.1247198206852385e-05,
"loss": 0.1256,
"step": 5535
},
{
"epoch": 7.97,
"grad_norm": 3.816267728805542,
"learning_rate": 1.1207172590457894e-05,
"loss": 0.1713,
"step": 5540
},
{
"epoch": 7.98,
"grad_norm": 3.6065261363983154,
"learning_rate": 1.1167146974063401e-05,
"loss": 0.1593,
"step": 5545
},
{
"epoch": 7.99,
"grad_norm": 3.608593463897705,
"learning_rate": 1.1127121357668908e-05,
"loss": 0.1727,
"step": 5550
},
{
"epoch": 8.0,
"grad_norm": 2.9817004203796387,
"learning_rate": 1.1087095741274415e-05,
"loss": 0.1424,
"step": 5555
},
{
"epoch": 8.0,
"eval_accuracy": 0.9548582995951417,
"eval_loss": 0.1296539306640625,
"eval_runtime": 26.6571,
"eval_samples_per_second": 370.634,
"eval_steps_per_second": 11.592,
"step": 5558
},
{
"epoch": 8.0,
"grad_norm": 7.694788455963135,
"learning_rate": 1.1047070124879923e-05,
"loss": 0.1771,
"step": 5560
},
{
"epoch": 8.01,
"grad_norm": 2.9499638080596924,
"learning_rate": 1.1007044508485432e-05,
"loss": 0.1576,
"step": 5565
},
{
"epoch": 8.02,
"grad_norm": 3.4104106426239014,
"learning_rate": 1.0967018892090939e-05,
"loss": 0.1495,
"step": 5570
},
{
"epoch": 8.02,
"grad_norm": 4.150513648986816,
"learning_rate": 1.0926993275696447e-05,
"loss": 0.1456,
"step": 5575
},
{
"epoch": 8.03,
"grad_norm": 4.284981727600098,
"learning_rate": 1.0886967659301954e-05,
"loss": 0.1184,
"step": 5580
},
{
"epoch": 8.04,
"grad_norm": 5.80072546005249,
"learning_rate": 1.0846942042907461e-05,
"loss": 0.188,
"step": 5585
},
{
"epoch": 8.05,
"grad_norm": 3.601327419281006,
"learning_rate": 1.0806916426512968e-05,
"loss": 0.1397,
"step": 5590
},
{
"epoch": 8.05,
"grad_norm": 3.128159999847412,
"learning_rate": 1.0766890810118477e-05,
"loss": 0.1188,
"step": 5595
},
{
"epoch": 8.06,
"grad_norm": 4.803286552429199,
"learning_rate": 1.0726865193723984e-05,
"loss": 0.199,
"step": 5600
},
{
"epoch": 8.07,
"grad_norm": 2.26747727394104,
"learning_rate": 1.068683957732949e-05,
"loss": 0.1105,
"step": 5605
},
{
"epoch": 8.07,
"grad_norm": 3.7966387271881104,
"learning_rate": 1.0646813960934999e-05,
"loss": 0.1747,
"step": 5610
},
{
"epoch": 8.08,
"grad_norm": 2.617194652557373,
"learning_rate": 1.0606788344540506e-05,
"loss": 0.1529,
"step": 5615
},
{
"epoch": 8.09,
"grad_norm": 3.259854793548584,
"learning_rate": 1.0566762728146015e-05,
"loss": 0.1641,
"step": 5620
},
{
"epoch": 8.1,
"grad_norm": 4.257889270782471,
"learning_rate": 1.0526737111751522e-05,
"loss": 0.1633,
"step": 5625
},
{
"epoch": 8.1,
"grad_norm": 2.608647108078003,
"learning_rate": 1.048671149535703e-05,
"loss": 0.1448,
"step": 5630
},
{
"epoch": 8.11,
"grad_norm": 3.398613929748535,
"learning_rate": 1.0446685878962537e-05,
"loss": 0.1447,
"step": 5635
},
{
"epoch": 8.12,
"grad_norm": 4.047391891479492,
"learning_rate": 1.0406660262568044e-05,
"loss": 0.142,
"step": 5640
},
{
"epoch": 8.13,
"grad_norm": 4.146562576293945,
"learning_rate": 1.0366634646173552e-05,
"loss": 0.131,
"step": 5645
},
{
"epoch": 8.13,
"grad_norm": 3.4250409603118896,
"learning_rate": 1.032660902977906e-05,
"loss": 0.1732,
"step": 5650
},
{
"epoch": 8.14,
"grad_norm": 3.730803966522217,
"learning_rate": 1.0286583413384566e-05,
"loss": 0.163,
"step": 5655
},
{
"epoch": 8.15,
"grad_norm": 5.8135504722595215,
"learning_rate": 1.0246557796990073e-05,
"loss": 0.1422,
"step": 5660
},
{
"epoch": 8.15,
"grad_norm": 6.937263488769531,
"learning_rate": 1.0206532180595582e-05,
"loss": 0.1504,
"step": 5665
},
{
"epoch": 8.16,
"grad_norm": 5.335781097412109,
"learning_rate": 1.0166506564201089e-05,
"loss": 0.1636,
"step": 5670
},
{
"epoch": 8.17,
"grad_norm": 5.222043037414551,
"learning_rate": 1.0126480947806596e-05,
"loss": 0.1459,
"step": 5675
},
{
"epoch": 8.18,
"grad_norm": 7.020112037658691,
"learning_rate": 1.0086455331412104e-05,
"loss": 0.1235,
"step": 5680
},
{
"epoch": 8.18,
"grad_norm": 5.291160583496094,
"learning_rate": 1.0046429715017613e-05,
"loss": 0.1816,
"step": 5685
},
{
"epoch": 8.19,
"grad_norm": 5.9437575340271,
"learning_rate": 1.000640409862312e-05,
"loss": 0.1886,
"step": 5690
},
{
"epoch": 8.2,
"grad_norm": 5.075570106506348,
"learning_rate": 9.966378482228627e-06,
"loss": 0.1766,
"step": 5695
},
{
"epoch": 8.2,
"grad_norm": 4.061030387878418,
"learning_rate": 9.926352865834135e-06,
"loss": 0.1349,
"step": 5700
},
{
"epoch": 8.21,
"grad_norm": 3.3227720260620117,
"learning_rate": 9.886327249439642e-06,
"loss": 0.1564,
"step": 5705
},
{
"epoch": 8.22,
"grad_norm": 5.788105487823486,
"learning_rate": 9.846301633045149e-06,
"loss": 0.1792,
"step": 5710
},
{
"epoch": 8.23,
"grad_norm": 4.576950550079346,
"learning_rate": 9.806276016650656e-06,
"loss": 0.1707,
"step": 5715
},
{
"epoch": 8.23,
"grad_norm": 1.9181571006774902,
"learning_rate": 9.766250400256164e-06,
"loss": 0.1588,
"step": 5720
},
{
"epoch": 8.24,
"grad_norm": 4.4835638999938965,
"learning_rate": 9.726224783861671e-06,
"loss": 0.1474,
"step": 5725
},
{
"epoch": 8.25,
"grad_norm": 3.0491483211517334,
"learning_rate": 9.686199167467178e-06,
"loss": 0.1391,
"step": 5730
},
{
"epoch": 8.25,
"grad_norm": 6.024807929992676,
"learning_rate": 9.646173551072687e-06,
"loss": 0.1819,
"step": 5735
},
{
"epoch": 8.26,
"grad_norm": 3.9458465576171875,
"learning_rate": 9.606147934678195e-06,
"loss": 0.1176,
"step": 5740
},
{
"epoch": 8.27,
"grad_norm": 5.939620494842529,
"learning_rate": 9.566122318283702e-06,
"loss": 0.1519,
"step": 5745
},
{
"epoch": 8.28,
"grad_norm": 2.2622225284576416,
"learning_rate": 9.52609670188921e-06,
"loss": 0.152,
"step": 5750
},
{
"epoch": 8.28,
"grad_norm": 3.9069314002990723,
"learning_rate": 9.486071085494718e-06,
"loss": 0.1556,
"step": 5755
},
{
"epoch": 8.29,
"grad_norm": 3.782322645187378,
"learning_rate": 9.446045469100225e-06,
"loss": 0.1863,
"step": 5760
},
{
"epoch": 8.3,
"grad_norm": 3.462238311767578,
"learning_rate": 9.406019852705732e-06,
"loss": 0.1217,
"step": 5765
},
{
"epoch": 8.31,
"grad_norm": 3.4270153045654297,
"learning_rate": 9.36599423631124e-06,
"loss": 0.1562,
"step": 5770
},
{
"epoch": 8.31,
"grad_norm": 4.370534420013428,
"learning_rate": 9.325968619916747e-06,
"loss": 0.1868,
"step": 5775
},
{
"epoch": 8.32,
"grad_norm": 5.722720623016357,
"learning_rate": 9.285943003522254e-06,
"loss": 0.1451,
"step": 5780
},
{
"epoch": 8.33,
"grad_norm": 4.850919723510742,
"learning_rate": 9.245917387127761e-06,
"loss": 0.1617,
"step": 5785
},
{
"epoch": 8.33,
"grad_norm": 3.516965627670288,
"learning_rate": 9.20589177073327e-06,
"loss": 0.1658,
"step": 5790
},
{
"epoch": 8.34,
"grad_norm": 3.0313217639923096,
"learning_rate": 9.165866154338778e-06,
"loss": 0.1594,
"step": 5795
},
{
"epoch": 8.35,
"grad_norm": 3.095097780227661,
"learning_rate": 9.125840537944285e-06,
"loss": 0.1256,
"step": 5800
},
{
"epoch": 8.36,
"grad_norm": 4.62247371673584,
"learning_rate": 9.085814921549794e-06,
"loss": 0.1664,
"step": 5805
},
{
"epoch": 8.36,
"grad_norm": 4.7111310958862305,
"learning_rate": 9.0457893051553e-06,
"loss": 0.1656,
"step": 5810
},
{
"epoch": 8.37,
"grad_norm": 3.2498939037323,
"learning_rate": 9.005763688760807e-06,
"loss": 0.164,
"step": 5815
},
{
"epoch": 8.38,
"grad_norm": 3.6476857662200928,
"learning_rate": 8.965738072366314e-06,
"loss": 0.1757,
"step": 5820
},
{
"epoch": 8.38,
"grad_norm": 6.174407482147217,
"learning_rate": 8.925712455971823e-06,
"loss": 0.1213,
"step": 5825
},
{
"epoch": 8.39,
"grad_norm": 4.54398775100708,
"learning_rate": 8.88568683957733e-06,
"loss": 0.1786,
"step": 5830
},
{
"epoch": 8.4,
"grad_norm": 2.651683807373047,
"learning_rate": 8.845661223182837e-06,
"loss": 0.1537,
"step": 5835
},
{
"epoch": 8.41,
"grad_norm": 4.218119144439697,
"learning_rate": 8.805635606788345e-06,
"loss": 0.1579,
"step": 5840
},
{
"epoch": 8.41,
"grad_norm": 3.273021936416626,
"learning_rate": 8.765609990393852e-06,
"loss": 0.154,
"step": 5845
},
{
"epoch": 8.42,
"grad_norm": 7.346030235290527,
"learning_rate": 8.725584373999359e-06,
"loss": 0.1743,
"step": 5850
},
{
"epoch": 8.43,
"grad_norm": 4.279568672180176,
"learning_rate": 8.685558757604868e-06,
"loss": 0.1704,
"step": 5855
},
{
"epoch": 8.43,
"grad_norm": 4.570573329925537,
"learning_rate": 8.645533141210376e-06,
"loss": 0.1697,
"step": 5860
},
{
"epoch": 8.44,
"grad_norm": 2.7568442821502686,
"learning_rate": 8.605507524815883e-06,
"loss": 0.1693,
"step": 5865
},
{
"epoch": 8.45,
"grad_norm": 3.1747210025787354,
"learning_rate": 8.56548190842139e-06,
"loss": 0.141,
"step": 5870
},
{
"epoch": 8.46,
"grad_norm": 3.326231002807617,
"learning_rate": 8.525456292026899e-06,
"loss": 0.1191,
"step": 5875
},
{
"epoch": 8.46,
"grad_norm": 5.524628639221191,
"learning_rate": 8.485430675632405e-06,
"loss": 0.1454,
"step": 5880
},
{
"epoch": 8.47,
"grad_norm": 2.5818092823028564,
"learning_rate": 8.445405059237912e-06,
"loss": 0.1409,
"step": 5885
},
{
"epoch": 8.48,
"grad_norm": 6.412874698638916,
"learning_rate": 8.40537944284342e-06,
"loss": 0.1661,
"step": 5890
},
{
"epoch": 8.49,
"grad_norm": 2.9045023918151855,
"learning_rate": 8.365353826448928e-06,
"loss": 0.1251,
"step": 5895
},
{
"epoch": 8.49,
"grad_norm": 4.160517692565918,
"learning_rate": 8.325328210054435e-06,
"loss": 0.1297,
"step": 5900
},
{
"epoch": 8.5,
"grad_norm": 5.732846736907959,
"learning_rate": 8.285302593659942e-06,
"loss": 0.1167,
"step": 5905
},
{
"epoch": 8.51,
"grad_norm": 2.2596616744995117,
"learning_rate": 8.24527697726545e-06,
"loss": 0.116,
"step": 5910
},
{
"epoch": 8.51,
"grad_norm": 4.199057102203369,
"learning_rate": 8.205251360870959e-06,
"loss": 0.167,
"step": 5915
},
{
"epoch": 8.52,
"grad_norm": 3.995457649230957,
"learning_rate": 8.165225744476466e-06,
"loss": 0.1513,
"step": 5920
},
{
"epoch": 8.53,
"grad_norm": 3.3315746784210205,
"learning_rate": 8.125200128081973e-06,
"loss": 0.1438,
"step": 5925
},
{
"epoch": 8.54,
"grad_norm": 4.279821872711182,
"learning_rate": 8.085174511687481e-06,
"loss": 0.1544,
"step": 5930
},
{
"epoch": 8.54,
"grad_norm": 4.037819862365723,
"learning_rate": 8.045148895292988e-06,
"loss": 0.1424,
"step": 5935
},
{
"epoch": 8.55,
"grad_norm": 2.268348455429077,
"learning_rate": 8.005123278898495e-06,
"loss": 0.136,
"step": 5940
},
{
"epoch": 8.56,
"grad_norm": 3.2165260314941406,
"learning_rate": 7.965097662504004e-06,
"loss": 0.1645,
"step": 5945
},
{
"epoch": 8.56,
"grad_norm": 2.8319571018218994,
"learning_rate": 7.92507204610951e-06,
"loss": 0.1517,
"step": 5950
},
{
"epoch": 8.57,
"grad_norm": 2.6142780780792236,
"learning_rate": 7.885046429715017e-06,
"loss": 0.1181,
"step": 5955
},
{
"epoch": 8.58,
"grad_norm": 4.016239643096924,
"learning_rate": 7.845020813320524e-06,
"loss": 0.141,
"step": 5960
},
{
"epoch": 8.59,
"grad_norm": 2.4863009452819824,
"learning_rate": 7.804995196926033e-06,
"loss": 0.1556,
"step": 5965
},
{
"epoch": 8.59,
"grad_norm": 7.284638404846191,
"learning_rate": 7.76496958053154e-06,
"loss": 0.1665,
"step": 5970
},
{
"epoch": 8.6,
"grad_norm": 3.67448091506958,
"learning_rate": 7.724943964137048e-06,
"loss": 0.1473,
"step": 5975
},
{
"epoch": 8.61,
"grad_norm": 3.4379279613494873,
"learning_rate": 7.684918347742557e-06,
"loss": 0.1304,
"step": 5980
},
{
"epoch": 8.61,
"grad_norm": 3.080416440963745,
"learning_rate": 7.644892731348064e-06,
"loss": 0.1304,
"step": 5985
},
{
"epoch": 8.62,
"grad_norm": 4.215944290161133,
"learning_rate": 7.604867114953571e-06,
"loss": 0.1884,
"step": 5990
},
{
"epoch": 8.63,
"grad_norm": 3.3756349086761475,
"learning_rate": 7.564841498559078e-06,
"loss": 0.1415,
"step": 5995
},
{
"epoch": 8.64,
"grad_norm": 4.051993370056152,
"learning_rate": 7.524815882164586e-06,
"loss": 0.1318,
"step": 6000
},
{
"epoch": 8.64,
"grad_norm": 3.0433032512664795,
"learning_rate": 7.484790265770093e-06,
"loss": 0.1427,
"step": 6005
},
{
"epoch": 8.65,
"grad_norm": 2.945946216583252,
"learning_rate": 7.4447646493756e-06,
"loss": 0.1374,
"step": 6010
},
{
"epoch": 8.66,
"grad_norm": 3.9186439514160156,
"learning_rate": 7.404739032981109e-06,
"loss": 0.167,
"step": 6015
},
{
"epoch": 8.66,
"grad_norm": 3.6851704120635986,
"learning_rate": 7.364713416586616e-06,
"loss": 0.1658,
"step": 6020
},
{
"epoch": 8.67,
"grad_norm": 5.047458171844482,
"learning_rate": 7.324687800192123e-06,
"loss": 0.1568,
"step": 6025
},
{
"epoch": 8.68,
"grad_norm": 3.574936628341675,
"learning_rate": 7.28466218379763e-06,
"loss": 0.1508,
"step": 6030
},
{
"epoch": 8.69,
"grad_norm": 4.4691057205200195,
"learning_rate": 7.244636567403139e-06,
"loss": 0.1467,
"step": 6035
},
{
"epoch": 8.69,
"grad_norm": 2.8131790161132812,
"learning_rate": 7.204610951008646e-06,
"loss": 0.1948,
"step": 6040
},
{
"epoch": 8.7,
"grad_norm": 4.179603576660156,
"learning_rate": 7.1645853346141535e-06,
"loss": 0.1824,
"step": 6045
},
{
"epoch": 8.71,
"grad_norm": 3.927391290664673,
"learning_rate": 7.124559718219661e-06,
"loss": 0.1436,
"step": 6050
},
{
"epoch": 8.72,
"grad_norm": 4.930620193481445,
"learning_rate": 7.084534101825169e-06,
"loss": 0.1708,
"step": 6055
},
{
"epoch": 8.72,
"grad_norm": 2.9689087867736816,
"learning_rate": 7.044508485430676e-06,
"loss": 0.1535,
"step": 6060
},
{
"epoch": 8.73,
"grad_norm": 4.4355316162109375,
"learning_rate": 7.004482869036183e-06,
"loss": 0.1758,
"step": 6065
},
{
"epoch": 8.74,
"grad_norm": 4.122786045074463,
"learning_rate": 6.964457252641691e-06,
"loss": 0.1496,
"step": 6070
},
{
"epoch": 8.74,
"grad_norm": 6.657648086547852,
"learning_rate": 6.924431636247199e-06,
"loss": 0.1725,
"step": 6075
},
{
"epoch": 8.75,
"grad_norm": 4.54775333404541,
"learning_rate": 6.884406019852706e-06,
"loss": 0.158,
"step": 6080
},
{
"epoch": 8.76,
"grad_norm": 2.5869054794311523,
"learning_rate": 6.844380403458213e-06,
"loss": 0.1509,
"step": 6085
},
{
"epoch": 8.77,
"grad_norm": 5.620914459228516,
"learning_rate": 6.8043547870637215e-06,
"loss": 0.1647,
"step": 6090
},
{
"epoch": 8.77,
"grad_norm": 3.599468946456909,
"learning_rate": 6.764329170669228e-06,
"loss": 0.1253,
"step": 6095
},
{
"epoch": 8.78,
"grad_norm": 3.4432973861694336,
"learning_rate": 6.724303554274736e-06,
"loss": 0.1452,
"step": 6100
},
{
"epoch": 8.79,
"grad_norm": 2.940871477127075,
"learning_rate": 6.684277937880244e-06,
"loss": 0.1382,
"step": 6105
},
{
"epoch": 8.79,
"grad_norm": 3.9849982261657715,
"learning_rate": 6.644252321485752e-06,
"loss": 0.1394,
"step": 6110
},
{
"epoch": 8.8,
"grad_norm": 2.607046604156494,
"learning_rate": 6.6042267050912585e-06,
"loss": 0.1434,
"step": 6115
},
{
"epoch": 8.81,
"grad_norm": 3.1801774501800537,
"learning_rate": 6.564201088696765e-06,
"loss": 0.126,
"step": 6120
},
{
"epoch": 8.82,
"grad_norm": 3.0512826442718506,
"learning_rate": 6.524175472302274e-06,
"loss": 0.1625,
"step": 6125
},
{
"epoch": 8.82,
"grad_norm": 5.523614883422852,
"learning_rate": 6.484149855907781e-06,
"loss": 0.1591,
"step": 6130
},
{
"epoch": 8.83,
"grad_norm": 5.236200332641602,
"learning_rate": 6.444124239513289e-06,
"loss": 0.1653,
"step": 6135
},
{
"epoch": 8.84,
"grad_norm": 4.800531387329102,
"learning_rate": 6.404098623118797e-06,
"loss": 0.1491,
"step": 6140
},
{
"epoch": 8.84,
"grad_norm": 3.7076640129089355,
"learning_rate": 6.364073006724304e-06,
"loss": 0.1459,
"step": 6145
},
{
"epoch": 8.85,
"grad_norm": 3.248276948928833,
"learning_rate": 6.324047390329811e-06,
"loss": 0.1281,
"step": 6150
},
{
"epoch": 8.86,
"grad_norm": 3.0434398651123047,
"learning_rate": 6.284021773935318e-06,
"loss": 0.1532,
"step": 6155
},
{
"epoch": 8.87,
"grad_norm": 3.669671058654785,
"learning_rate": 6.2439961575408265e-06,
"loss": 0.1308,
"step": 6160
},
{
"epoch": 8.87,
"grad_norm": 2.7335450649261475,
"learning_rate": 6.203970541146334e-06,
"loss": 0.1377,
"step": 6165
},
{
"epoch": 8.88,
"grad_norm": 3.6164846420288086,
"learning_rate": 6.163944924751842e-06,
"loss": 0.175,
"step": 6170
},
{
"epoch": 8.89,
"grad_norm": 3.5082285404205322,
"learning_rate": 6.123919308357349e-06,
"loss": 0.1921,
"step": 6175
},
{
"epoch": 8.9,
"grad_norm": 3.8012921810150146,
"learning_rate": 6.083893691962857e-06,
"loss": 0.1384,
"step": 6180
},
{
"epoch": 8.9,
"grad_norm": 2.176058053970337,
"learning_rate": 6.0438680755683635e-06,
"loss": 0.1364,
"step": 6185
},
{
"epoch": 8.91,
"grad_norm": 3.2312066555023193,
"learning_rate": 6.003842459173871e-06,
"loss": 0.1686,
"step": 6190
},
{
"epoch": 8.92,
"grad_norm": 5.053197860717773,
"learning_rate": 5.963816842779379e-06,
"loss": 0.1815,
"step": 6195
},
{
"epoch": 8.92,
"grad_norm": 3.445341110229492,
"learning_rate": 5.923791226384887e-06,
"loss": 0.1322,
"step": 6200
},
{
"epoch": 8.93,
"grad_norm": 4.0553669929504395,
"learning_rate": 5.8837656099903945e-06,
"loss": 0.1456,
"step": 6205
},
{
"epoch": 8.94,
"grad_norm": 3.9864542484283447,
"learning_rate": 5.843739993595901e-06,
"loss": 0.1617,
"step": 6210
},
{
"epoch": 8.95,
"grad_norm": 3.3293707370758057,
"learning_rate": 5.803714377201409e-06,
"loss": 0.1199,
"step": 6215
},
{
"epoch": 8.95,
"grad_norm": 3.733670949935913,
"learning_rate": 5.763688760806917e-06,
"loss": 0.1965,
"step": 6220
},
{
"epoch": 8.96,
"grad_norm": 3.420346260070801,
"learning_rate": 5.723663144412425e-06,
"loss": 0.1764,
"step": 6225
},
{
"epoch": 8.97,
"grad_norm": 4.500962257385254,
"learning_rate": 5.6836375280179315e-06,
"loss": 0.1632,
"step": 6230
},
{
"epoch": 8.97,
"grad_norm": 3.7348039150238037,
"learning_rate": 5.643611911623439e-06,
"loss": 0.1441,
"step": 6235
},
{
"epoch": 8.98,
"grad_norm": 4.5082502365112305,
"learning_rate": 5.603586295228947e-06,
"loss": 0.1602,
"step": 6240
},
{
"epoch": 8.99,
"grad_norm": 7.206320762634277,
"learning_rate": 5.563560678834454e-06,
"loss": 0.2032,
"step": 6245
},
{
"epoch": 9.0,
"grad_norm": 5.481276035308838,
"learning_rate": 5.523535062439962e-06,
"loss": 0.1555,
"step": 6250
},
{
"epoch": 9.0,
"eval_accuracy": 0.9551619433198381,
"eval_loss": 0.13027772307395935,
"eval_runtime": 26.8168,
"eval_samples_per_second": 368.425,
"eval_steps_per_second": 11.523,
"step": 6252
},
{
"epoch": 9.0,
"grad_norm": 4.276639938354492,
"learning_rate": 5.4835094460454694e-06,
"loss": 0.1455,
"step": 6255
},
{
"epoch": 9.01,
"grad_norm": 3.9856693744659424,
"learning_rate": 5.443483829650977e-06,
"loss": 0.1415,
"step": 6260
},
{
"epoch": 9.02,
"grad_norm": 3.2395150661468506,
"learning_rate": 5.403458213256484e-06,
"loss": 0.146,
"step": 6265
},
{
"epoch": 9.02,
"grad_norm": 1.18377685546875,
"learning_rate": 5.363432596861992e-06,
"loss": 0.1048,
"step": 6270
},
{
"epoch": 9.03,
"grad_norm": 4.3180928230285645,
"learning_rate": 5.3234069804674996e-06,
"loss": 0.1623,
"step": 6275
},
{
"epoch": 9.04,
"grad_norm": 2.203953742980957,
"learning_rate": 5.283381364073007e-06,
"loss": 0.1763,
"step": 6280
},
{
"epoch": 9.05,
"grad_norm": 4.12826681137085,
"learning_rate": 5.243355747678515e-06,
"loss": 0.155,
"step": 6285
},
{
"epoch": 9.05,
"grad_norm": 3.8807485103607178,
"learning_rate": 5.203330131284022e-06,
"loss": 0.1222,
"step": 6290
},
{
"epoch": 9.06,
"grad_norm": 3.572359085083008,
"learning_rate": 5.16330451488953e-06,
"loss": 0.1359,
"step": 6295
},
{
"epoch": 9.07,
"grad_norm": 2.7034575939178467,
"learning_rate": 5.123278898495037e-06,
"loss": 0.128,
"step": 6300
},
{
"epoch": 9.08,
"grad_norm": 4.068083763122559,
"learning_rate": 5.083253282100544e-06,
"loss": 0.1842,
"step": 6305
},
{
"epoch": 9.08,
"grad_norm": 3.0719077587127686,
"learning_rate": 5.043227665706052e-06,
"loss": 0.1435,
"step": 6310
},
{
"epoch": 9.09,
"grad_norm": 2.3211283683776855,
"learning_rate": 5.00320204931156e-06,
"loss": 0.1465,
"step": 6315
},
{
"epoch": 9.1,
"grad_norm": 3.4732654094696045,
"learning_rate": 4.9631764329170676e-06,
"loss": 0.1643,
"step": 6320
},
{
"epoch": 9.1,
"grad_norm": 3.3369507789611816,
"learning_rate": 4.9231508165225745e-06,
"loss": 0.1434,
"step": 6325
},
{
"epoch": 9.11,
"grad_norm": 3.845771312713623,
"learning_rate": 4.883125200128082e-06,
"loss": 0.1646,
"step": 6330
},
{
"epoch": 9.12,
"grad_norm": 3.8775432109832764,
"learning_rate": 4.843099583733589e-06,
"loss": 0.1551,
"step": 6335
},
{
"epoch": 9.13,
"grad_norm": 3.2787814140319824,
"learning_rate": 4.803073967339098e-06,
"loss": 0.1054,
"step": 6340
},
{
"epoch": 9.13,
"grad_norm": 4.375431060791016,
"learning_rate": 4.763048350944605e-06,
"loss": 0.1554,
"step": 6345
},
{
"epoch": 9.14,
"grad_norm": 4.358311653137207,
"learning_rate": 4.723022734550112e-06,
"loss": 0.1384,
"step": 6350
},
{
"epoch": 9.15,
"grad_norm": 3.3018763065338135,
"learning_rate": 4.68299711815562e-06,
"loss": 0.1683,
"step": 6355
},
{
"epoch": 9.15,
"grad_norm": 3.3858537673950195,
"learning_rate": 4.642971501761127e-06,
"loss": 0.1672,
"step": 6360
},
{
"epoch": 9.16,
"grad_norm": 3.9749813079833984,
"learning_rate": 4.602945885366635e-06,
"loss": 0.1638,
"step": 6365
},
{
"epoch": 9.17,
"grad_norm": 5.917661666870117,
"learning_rate": 4.5629202689721425e-06,
"loss": 0.1417,
"step": 6370
},
{
"epoch": 9.18,
"grad_norm": 3.9763858318328857,
"learning_rate": 4.52289465257765e-06,
"loss": 0.0966,
"step": 6375
},
{
"epoch": 9.18,
"grad_norm": 3.998041868209839,
"learning_rate": 4.482869036183157e-06,
"loss": 0.1367,
"step": 6380
},
{
"epoch": 9.19,
"grad_norm": 3.193472146987915,
"learning_rate": 4.442843419788665e-06,
"loss": 0.1464,
"step": 6385
},
{
"epoch": 9.2,
"grad_norm": 4.275448322296143,
"learning_rate": 4.402817803394173e-06,
"loss": 0.1398,
"step": 6390
},
{
"epoch": 9.2,
"grad_norm": 4.345200538635254,
"learning_rate": 4.3627921869996795e-06,
"loss": 0.1347,
"step": 6395
},
{
"epoch": 9.21,
"grad_norm": 3.47943115234375,
"learning_rate": 4.322766570605188e-06,
"loss": 0.1849,
"step": 6400
},
{
"epoch": 9.22,
"grad_norm": 3.6434147357940674,
"learning_rate": 4.282740954210695e-06,
"loss": 0.1694,
"step": 6405
},
{
"epoch": 9.23,
"grad_norm": 3.210238456726074,
"learning_rate": 4.242715337816203e-06,
"loss": 0.1491,
"step": 6410
},
{
"epoch": 9.23,
"grad_norm": 2.9370250701904297,
"learning_rate": 4.20268972142171e-06,
"loss": 0.1279,
"step": 6415
},
{
"epoch": 9.24,
"grad_norm": 3.1297073364257812,
"learning_rate": 4.162664105027217e-06,
"loss": 0.1501,
"step": 6420
},
{
"epoch": 9.25,
"grad_norm": 2.1068637371063232,
"learning_rate": 4.122638488632725e-06,
"loss": 0.1338,
"step": 6425
},
{
"epoch": 9.26,
"grad_norm": 2.867342472076416,
"learning_rate": 4.082612872238233e-06,
"loss": 0.1261,
"step": 6430
},
{
"epoch": 9.26,
"grad_norm": 4.779037952423096,
"learning_rate": 4.042587255843741e-06,
"loss": 0.1451,
"step": 6435
},
{
"epoch": 9.27,
"grad_norm": 4.961643218994141,
"learning_rate": 4.0025616394492475e-06,
"loss": 0.1724,
"step": 6440
},
{
"epoch": 9.28,
"grad_norm": 3.6226232051849365,
"learning_rate": 3.962536023054755e-06,
"loss": 0.1426,
"step": 6445
},
{
"epoch": 9.28,
"grad_norm": 2.5623698234558105,
"learning_rate": 3.922510406660262e-06,
"loss": 0.1524,
"step": 6450
},
{
"epoch": 9.29,
"grad_norm": 5.580244064331055,
"learning_rate": 3.88248479026577e-06,
"loss": 0.1684,
"step": 6455
},
{
"epoch": 9.3,
"grad_norm": 3.1150074005126953,
"learning_rate": 3.8424591738712785e-06,
"loss": 0.145,
"step": 6460
},
{
"epoch": 9.31,
"grad_norm": 2.4989027976989746,
"learning_rate": 3.8024335574767854e-06,
"loss": 0.1636,
"step": 6465
},
{
"epoch": 9.31,
"grad_norm": 2.90736722946167,
"learning_rate": 3.762407941082293e-06,
"loss": 0.1573,
"step": 6470
},
{
"epoch": 9.32,
"grad_norm": 2.772012710571289,
"learning_rate": 3.7223823246878e-06,
"loss": 0.1512,
"step": 6475
},
{
"epoch": 9.33,
"grad_norm": 2.553661346435547,
"learning_rate": 3.682356708293308e-06,
"loss": 0.1666,
"step": 6480
},
{
"epoch": 9.33,
"grad_norm": 3.670997381210327,
"learning_rate": 3.642331091898815e-06,
"loss": 0.2017,
"step": 6485
},
{
"epoch": 9.34,
"grad_norm": 3.2845664024353027,
"learning_rate": 3.602305475504323e-06,
"loss": 0.1749,
"step": 6490
},
{
"epoch": 9.35,
"grad_norm": 3.8596317768096924,
"learning_rate": 3.5622798591098306e-06,
"loss": 0.2147,
"step": 6495
},
{
"epoch": 9.36,
"grad_norm": 3.070552349090576,
"learning_rate": 3.522254242715338e-06,
"loss": 0.1353,
"step": 6500
},
{
"epoch": 9.36,
"grad_norm": 3.2333202362060547,
"learning_rate": 3.4822286263208457e-06,
"loss": 0.1722,
"step": 6505
},
{
"epoch": 9.37,
"grad_norm": 3.9244325160980225,
"learning_rate": 3.442203009926353e-06,
"loss": 0.1528,
"step": 6510
},
{
"epoch": 9.38,
"grad_norm": 4.576772689819336,
"learning_rate": 3.4021773935318607e-06,
"loss": 0.1257,
"step": 6515
},
{
"epoch": 9.38,
"grad_norm": 2.5687806606292725,
"learning_rate": 3.362151777137368e-06,
"loss": 0.1155,
"step": 6520
},
{
"epoch": 9.39,
"grad_norm": 3.3034443855285645,
"learning_rate": 3.322126160742876e-06,
"loss": 0.1803,
"step": 6525
},
{
"epoch": 9.4,
"grad_norm": 5.908419609069824,
"learning_rate": 3.2821005443483827e-06,
"loss": 0.1669,
"step": 6530
},
{
"epoch": 9.41,
"grad_norm": 3.6566712856292725,
"learning_rate": 3.2420749279538904e-06,
"loss": 0.1431,
"step": 6535
},
{
"epoch": 9.41,
"grad_norm": 3.199873208999634,
"learning_rate": 3.2020493115593986e-06,
"loss": 0.1709,
"step": 6540
},
{
"epoch": 9.42,
"grad_norm": 3.752716302871704,
"learning_rate": 3.1620236951649055e-06,
"loss": 0.146,
"step": 6545
},
{
"epoch": 9.43,
"grad_norm": 2.432011842727661,
"learning_rate": 3.1219980787704133e-06,
"loss": 0.1525,
"step": 6550
},
{
"epoch": 9.44,
"grad_norm": 2.615161895751953,
"learning_rate": 3.081972462375921e-06,
"loss": 0.1491,
"step": 6555
},
{
"epoch": 9.44,
"grad_norm": 3.136516571044922,
"learning_rate": 3.0419468459814283e-06,
"loss": 0.1221,
"step": 6560
},
{
"epoch": 9.45,
"grad_norm": 3.304135322570801,
"learning_rate": 3.0019212295869356e-06,
"loss": 0.1717,
"step": 6565
},
{
"epoch": 9.46,
"grad_norm": 5.078571796417236,
"learning_rate": 2.9618956131924434e-06,
"loss": 0.1656,
"step": 6570
},
{
"epoch": 9.46,
"grad_norm": 3.5032825469970703,
"learning_rate": 2.9218699967979507e-06,
"loss": 0.2004,
"step": 6575
},
{
"epoch": 9.47,
"grad_norm": 3.3871119022369385,
"learning_rate": 2.8818443804034585e-06,
"loss": 0.122,
"step": 6580
},
{
"epoch": 9.48,
"grad_norm": 3.4582040309906006,
"learning_rate": 2.8418187640089658e-06,
"loss": 0.1465,
"step": 6585
},
{
"epoch": 9.49,
"grad_norm": 2.7074215412139893,
"learning_rate": 2.8017931476144735e-06,
"loss": 0.159,
"step": 6590
},
{
"epoch": 9.49,
"grad_norm": 3.0131921768188477,
"learning_rate": 2.761767531219981e-06,
"loss": 0.1573,
"step": 6595
},
{
"epoch": 9.5,
"grad_norm": 3.2007336616516113,
"learning_rate": 2.7217419148254886e-06,
"loss": 0.1196,
"step": 6600
},
{
"epoch": 9.51,
"grad_norm": 3.2668557167053223,
"learning_rate": 2.681716298430996e-06,
"loss": 0.1251,
"step": 6605
},
{
"epoch": 9.51,
"grad_norm": 3.1396324634552,
"learning_rate": 2.6416906820365037e-06,
"loss": 0.1482,
"step": 6610
},
{
"epoch": 9.52,
"grad_norm": 3.94993257522583,
"learning_rate": 2.601665065642011e-06,
"loss": 0.1568,
"step": 6615
},
{
"epoch": 9.53,
"grad_norm": 3.8396835327148438,
"learning_rate": 2.5616394492475183e-06,
"loss": 0.1813,
"step": 6620
},
{
"epoch": 9.54,
"grad_norm": 6.602664947509766,
"learning_rate": 2.521613832853026e-06,
"loss": 0.1661,
"step": 6625
},
{
"epoch": 9.54,
"grad_norm": 4.1188225746154785,
"learning_rate": 2.4815882164585338e-06,
"loss": 0.1111,
"step": 6630
},
{
"epoch": 9.55,
"grad_norm": 3.777866840362549,
"learning_rate": 2.441562600064041e-06,
"loss": 0.1788,
"step": 6635
},
{
"epoch": 9.56,
"grad_norm": 4.2113542556762695,
"learning_rate": 2.401536983669549e-06,
"loss": 0.1694,
"step": 6640
},
{
"epoch": 9.56,
"grad_norm": 4.3874831199646,
"learning_rate": 2.361511367275056e-06,
"loss": 0.1448,
"step": 6645
},
{
"epoch": 9.57,
"grad_norm": 3.2576651573181152,
"learning_rate": 2.3214857508805635e-06,
"loss": 0.1353,
"step": 6650
},
{
"epoch": 9.58,
"grad_norm": 3.4032511711120605,
"learning_rate": 2.2814601344860712e-06,
"loss": 0.1045,
"step": 6655
},
{
"epoch": 9.59,
"grad_norm": 4.300445079803467,
"learning_rate": 2.2414345180915786e-06,
"loss": 0.1479,
"step": 6660
},
{
"epoch": 9.59,
"grad_norm": 3.0734059810638428,
"learning_rate": 2.2014089016970863e-06,
"loss": 0.1494,
"step": 6665
},
{
"epoch": 9.6,
"grad_norm": 5.5670952796936035,
"learning_rate": 2.161383285302594e-06,
"loss": 0.1565,
"step": 6670
},
{
"epoch": 9.61,
"grad_norm": 4.408038139343262,
"learning_rate": 2.1213576689081014e-06,
"loss": 0.1758,
"step": 6675
},
{
"epoch": 9.61,
"grad_norm": 4.501611709594727,
"learning_rate": 2.0813320525136087e-06,
"loss": 0.1658,
"step": 6680
},
{
"epoch": 9.62,
"grad_norm": 3.2457005977630615,
"learning_rate": 2.0413064361191164e-06,
"loss": 0.1599,
"step": 6685
},
{
"epoch": 9.63,
"grad_norm": 4.1629862785339355,
"learning_rate": 2.0012808197246238e-06,
"loss": 0.141,
"step": 6690
},
{
"epoch": 9.64,
"grad_norm": 2.0646541118621826,
"learning_rate": 1.961255203330131e-06,
"loss": 0.1197,
"step": 6695
},
{
"epoch": 9.64,
"grad_norm": 4.061581134796143,
"learning_rate": 1.9212295869356392e-06,
"loss": 0.1541,
"step": 6700
},
{
"epoch": 9.65,
"grad_norm": 4.780194282531738,
"learning_rate": 1.8812039705411466e-06,
"loss": 0.1247,
"step": 6705
},
{
"epoch": 9.66,
"grad_norm": 3.3698534965515137,
"learning_rate": 1.841178354146654e-06,
"loss": 0.1486,
"step": 6710
},
{
"epoch": 9.67,
"grad_norm": 2.801363945007324,
"learning_rate": 1.8011527377521614e-06,
"loss": 0.1459,
"step": 6715
},
{
"epoch": 9.67,
"grad_norm": 4.238241672515869,
"learning_rate": 1.761127121357669e-06,
"loss": 0.1772,
"step": 6720
},
{
"epoch": 9.68,
"grad_norm": 2.364281415939331,
"learning_rate": 1.7211015049631765e-06,
"loss": 0.1272,
"step": 6725
},
{
"epoch": 9.69,
"grad_norm": 4.3810038566589355,
"learning_rate": 1.681075888568684e-06,
"loss": 0.112,
"step": 6730
},
{
"epoch": 9.69,
"grad_norm": 2.3395895957946777,
"learning_rate": 1.6410502721741913e-06,
"loss": 0.1298,
"step": 6735
},
{
"epoch": 9.7,
"grad_norm": 3.7817649841308594,
"learning_rate": 1.6010246557796993e-06,
"loss": 0.1638,
"step": 6740
},
{
"epoch": 9.71,
"grad_norm": 2.224323272705078,
"learning_rate": 1.5609990393852066e-06,
"loss": 0.1213,
"step": 6745
},
{
"epoch": 9.72,
"grad_norm": 3.1445086002349854,
"learning_rate": 1.5209734229907142e-06,
"loss": 0.1252,
"step": 6750
},
{
"epoch": 9.72,
"grad_norm": 3.2290494441986084,
"learning_rate": 1.4809478065962217e-06,
"loss": 0.1635,
"step": 6755
},
{
"epoch": 9.73,
"grad_norm": 3.6329257488250732,
"learning_rate": 1.4409221902017292e-06,
"loss": 0.13,
"step": 6760
},
{
"epoch": 9.74,
"grad_norm": 3.1905136108398438,
"learning_rate": 1.4008965738072368e-06,
"loss": 0.1265,
"step": 6765
},
{
"epoch": 9.74,
"grad_norm": 5.263790607452393,
"learning_rate": 1.3608709574127443e-06,
"loss": 0.1498,
"step": 6770
},
{
"epoch": 9.75,
"grad_norm": 3.3033974170684814,
"learning_rate": 1.3208453410182518e-06,
"loss": 0.1371,
"step": 6775
},
{
"epoch": 9.76,
"grad_norm": 3.2312657833099365,
"learning_rate": 1.2808197246237591e-06,
"loss": 0.1604,
"step": 6780
},
{
"epoch": 9.77,
"grad_norm": 4.103578567504883,
"learning_rate": 1.2407941082292669e-06,
"loss": 0.1532,
"step": 6785
},
{
"epoch": 9.77,
"grad_norm": 3.236082077026367,
"learning_rate": 1.2007684918347744e-06,
"loss": 0.1199,
"step": 6790
},
{
"epoch": 9.78,
"grad_norm": 3.1800456047058105,
"learning_rate": 1.1607428754402817e-06,
"loss": 0.1527,
"step": 6795
},
{
"epoch": 9.79,
"grad_norm": 2.98972225189209,
"learning_rate": 1.1207172590457893e-06,
"loss": 0.1367,
"step": 6800
},
{
"epoch": 9.79,
"grad_norm": 3.6796061992645264,
"learning_rate": 1.080691642651297e-06,
"loss": 0.1639,
"step": 6805
},
{
"epoch": 9.8,
"grad_norm": 3.3537867069244385,
"learning_rate": 1.0406660262568043e-06,
"loss": 0.1696,
"step": 6810
},
{
"epoch": 9.81,
"grad_norm": 3.785534143447876,
"learning_rate": 1.0006404098623119e-06,
"loss": 0.1267,
"step": 6815
},
{
"epoch": 9.82,
"grad_norm": 3.4578371047973633,
"learning_rate": 9.606147934678196e-07,
"loss": 0.1542,
"step": 6820
},
{
"epoch": 9.82,
"grad_norm": 2.174834966659546,
"learning_rate": 9.20589177073327e-07,
"loss": 0.1908,
"step": 6825
},
{
"epoch": 9.83,
"grad_norm": 2.7609782218933105,
"learning_rate": 8.805635606788345e-07,
"loss": 0.152,
"step": 6830
},
{
"epoch": 9.84,
"grad_norm": 2.602288246154785,
"learning_rate": 8.40537944284342e-07,
"loss": 0.1785,
"step": 6835
},
{
"epoch": 9.85,
"grad_norm": 3.4062986373901367,
"learning_rate": 8.005123278898497e-07,
"loss": 0.175,
"step": 6840
},
{
"epoch": 9.85,
"grad_norm": 2.9411168098449707,
"learning_rate": 7.604867114953571e-07,
"loss": 0.1402,
"step": 6845
},
{
"epoch": 9.86,
"grad_norm": 4.67616605758667,
"learning_rate": 7.204610951008646e-07,
"loss": 0.1281,
"step": 6850
},
{
"epoch": 9.87,
"grad_norm": 4.2490763664245605,
"learning_rate": 6.804354787063721e-07,
"loss": 0.1378,
"step": 6855
},
{
"epoch": 9.87,
"grad_norm": 3.4150235652923584,
"learning_rate": 6.404098623118796e-07,
"loss": 0.15,
"step": 6860
},
{
"epoch": 9.88,
"grad_norm": 3.8505630493164062,
"learning_rate": 6.003842459173872e-07,
"loss": 0.1729,
"step": 6865
},
{
"epoch": 9.89,
"grad_norm": 3.3457560539245605,
"learning_rate": 5.603586295228946e-07,
"loss": 0.1047,
"step": 6870
},
{
"epoch": 9.9,
"grad_norm": 4.059729099273682,
"learning_rate": 5.203330131284022e-07,
"loss": 0.1271,
"step": 6875
},
{
"epoch": 9.9,
"grad_norm": 2.4905691146850586,
"learning_rate": 4.803073967339098e-07,
"loss": 0.1245,
"step": 6880
},
{
"epoch": 9.91,
"grad_norm": 5.17240571975708,
"learning_rate": 4.4028178033941724e-07,
"loss": 0.17,
"step": 6885
},
{
"epoch": 9.92,
"grad_norm": 3.049504518508911,
"learning_rate": 4.002561639449248e-07,
"loss": 0.161,
"step": 6890
},
{
"epoch": 9.92,
"grad_norm": 3.2398414611816406,
"learning_rate": 3.602305475504323e-07,
"loss": 0.1539,
"step": 6895
},
{
"epoch": 9.93,
"grad_norm": 3.9823930263519287,
"learning_rate": 3.202049311559398e-07,
"loss": 0.1569,
"step": 6900
},
{
"epoch": 9.94,
"grad_norm": 3.368542194366455,
"learning_rate": 2.801793147614473e-07,
"loss": 0.1652,
"step": 6905
},
{
"epoch": 9.95,
"grad_norm": 2.684319019317627,
"learning_rate": 2.401536983669549e-07,
"loss": 0.1253,
"step": 6910
},
{
"epoch": 9.95,
"grad_norm": 3.630676746368408,
"learning_rate": 2.001280819724624e-07,
"loss": 0.1713,
"step": 6915
},
{
"epoch": 9.96,
"grad_norm": 5.006500720977783,
"learning_rate": 1.601024655779699e-07,
"loss": 0.1505,
"step": 6920
},
{
"epoch": 9.97,
"grad_norm": 5.6927313804626465,
"learning_rate": 1.2007684918347745e-07,
"loss": 0.181,
"step": 6925
},
{
"epoch": 9.97,
"grad_norm": 2.7989654541015625,
"learning_rate": 8.005123278898495e-08,
"loss": 0.1651,
"step": 6930
},
{
"epoch": 9.98,
"grad_norm": 6.005398750305176,
"learning_rate": 4.0025616394492473e-08,
"loss": 0.1509,
"step": 6935
},
{
"epoch": 9.99,
"grad_norm": 3.4408304691314697,
"learning_rate": 0.0,
"loss": 0.1238,
"step": 6940
},
{
"epoch": 9.99,
"eval_accuracy": 0.9558704453441296,
"eval_loss": 0.1293453723192215,
"eval_runtime": 26.754,
"eval_samples_per_second": 369.291,
"eval_steps_per_second": 11.55,
"step": 6940
},
{
"epoch": 9.99,
"step": 6940,
"total_flos": 2.207935349023097e+19,
"train_loss": 0.24048639133100208,
"train_runtime": 4813.8727,
"train_samples_per_second": 184.697,
"train_steps_per_second": 1.442
}
],
"logging_steps": 5,
"max_steps": 6940,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 2.207935349023097e+19,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}