{
  "best_metric": 0.8499902210052807,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-988",
  "epoch": 19.046138211382114,
  "eval_steps": 500,
  "global_step": 4920,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020325203252032522,
      "grad_norm": 4.696896553039551,
      "learning_rate": 1.0162601626016261e-06,
      "loss": 1.0589,
      "step": 10
    },
    {
      "epoch": 0.0040650406504065045,
      "grad_norm": 5.456978797912598,
      "learning_rate": 2.0325203252032523e-06,
      "loss": 1.0583,
      "step": 20
    },
    {
      "epoch": 0.006097560975609756,
      "grad_norm": 7.194735527038574,
      "learning_rate": 3.0487804878048782e-06,
      "loss": 1.0194,
      "step": 30
    },
    {
      "epoch": 0.008130081300813009,
      "grad_norm": 6.02046012878418,
      "learning_rate": 4.0650406504065046e-06,
      "loss": 1.0018,
      "step": 40
    },
    {
      "epoch": 0.01016260162601626,
      "grad_norm": 7.232560157775879,
      "learning_rate": 5.08130081300813e-06,
      "loss": 0.9331,
      "step": 50
    },
    {
      "epoch": 0.012195121951219513,
      "grad_norm": 5.583743095397949,
      "learning_rate": 6.0975609756097564e-06,
      "loss": 0.8941,
      "step": 60
    },
    {
      "epoch": 0.014227642276422764,
      "grad_norm": 4.490896224975586,
      "learning_rate": 7.113821138211382e-06,
      "loss": 0.8663,
      "step": 70
    },
    {
      "epoch": 0.016260162601626018,
      "grad_norm": 8.27023983001709,
      "learning_rate": 8.130081300813009e-06,
      "loss": 0.8057,
      "step": 80
    },
    {
      "epoch": 0.018292682926829267,
      "grad_norm": 5.526022911071777,
      "learning_rate": 9.146341463414634e-06,
      "loss": 0.7339,
      "step": 90
    },
    {
      "epoch": 0.02032520325203252,
      "grad_norm": 6.2672119140625,
      "learning_rate": 1.016260162601626e-05,
      "loss": 0.6207,
      "step": 100
    },
    {
      "epoch": 0.022357723577235773,
      "grad_norm": 5.983646869659424,
      "learning_rate": 1.1178861788617887e-05,
      "loss": 0.5178,
      "step": 110
    },
    {
      "epoch": 0.024390243902439025,
      "grad_norm": 7.784273147583008,
      "learning_rate": 1.2195121951219513e-05,
      "loss": 0.4967,
      "step": 120
    },
    {
      "epoch": 0.026422764227642278,
      "grad_norm": 9.177685737609863,
      "learning_rate": 1.321138211382114e-05,
      "loss": 0.4355,
      "step": 130
    },
    {
      "epoch": 0.028455284552845527,
      "grad_norm": 7.3704833984375,
      "learning_rate": 1.4227642276422764e-05,
      "loss": 0.425,
      "step": 140
    },
    {
      "epoch": 0.03048780487804878,
      "grad_norm": 6.048780918121338,
      "learning_rate": 1.524390243902439e-05,
      "loss": 0.3697,
      "step": 150
    },
    {
      "epoch": 0.032520325203252036,
      "grad_norm": 9.803858757019043,
      "learning_rate": 1.6260162601626018e-05,
      "loss": 0.4005,
      "step": 160
    },
    {
      "epoch": 0.034552845528455285,
      "grad_norm": 7.750985622406006,
      "learning_rate": 1.7276422764227643e-05,
      "loss": 0.3218,
      "step": 170
    },
    {
      "epoch": 0.036585365853658534,
      "grad_norm": 4.0331268310546875,
      "learning_rate": 1.8292682926829268e-05,
      "loss": 0.3925,
      "step": 180
    },
    {
      "epoch": 0.03861788617886179,
      "grad_norm": 11.129854202270508,
      "learning_rate": 1.9308943089430896e-05,
      "loss": 0.3311,
      "step": 190
    },
    {
      "epoch": 0.04065040650406504,
      "grad_norm": 5.826282978057861,
      "learning_rate": 2.032520325203252e-05,
      "loss": 0.385,
      "step": 200
    },
    {
      "epoch": 0.042682926829268296,
      "grad_norm": 3.613365888595581,
      "learning_rate": 2.134146341463415e-05,
      "loss": 0.4435,
      "step": 210
    },
    {
      "epoch": 0.044715447154471545,
      "grad_norm": 5.127413272857666,
      "learning_rate": 2.2357723577235773e-05,
      "loss": 0.4089,
      "step": 220
    },
    {
      "epoch": 0.046747967479674794,
      "grad_norm": 7.485469341278076,
      "learning_rate": 2.3373983739837398e-05,
      "loss": 0.3032,
      "step": 230
    },
    {
      "epoch": 0.04878048780487805,
      "grad_norm": 5.982870101928711,
      "learning_rate": 2.4390243902439026e-05,
      "loss": 0.3212,
      "step": 240
    },
    {
      "epoch": 0.050203252032520324,
      "eval_accuracy": 0.7897516135341287,
      "eval_loss": 0.5764808058738708,
      "eval_runtime": 4139.6988,
      "eval_samples_per_second": 1.235,
      "eval_steps_per_second": 0.077,
      "step": 247
    },
    {
      "epoch": 1.0006097560975609,
      "grad_norm": 5.0687432289123535,
      "learning_rate": 2.5406504065040654e-05,
      "loss": 0.4616,
      "step": 250
    },
    {
      "epoch": 1.0026422764227643,
      "grad_norm": 30.613248825073242,
      "learning_rate": 2.642276422764228e-05,
      "loss": 0.4082,
      "step": 260
    },
    {
      "epoch": 1.0046747967479674,
      "grad_norm": 3.021082639694214,
      "learning_rate": 2.7439024390243906e-05,
      "loss": 0.4118,
      "step": 270
    },
    {
      "epoch": 1.0067073170731706,
      "grad_norm": 10.735918998718262,
      "learning_rate": 2.8455284552845528e-05,
      "loss": 0.3345,
      "step": 280
    },
    {
      "epoch": 1.008739837398374,
      "grad_norm": 3.1623053550720215,
      "learning_rate": 2.947154471544716e-05,
      "loss": 0.3785,
      "step": 290
    },
    {
      "epoch": 1.0107723577235772,
      "grad_norm": 3.2091872692108154,
      "learning_rate": 3.048780487804878e-05,
      "loss": 0.3185,
      "step": 300
    },
    {
      "epoch": 1.0128048780487804,
      "grad_norm": 2.9003994464874268,
      "learning_rate": 3.150406504065041e-05,
      "loss": 0.1918,
      "step": 310
    },
    {
      "epoch": 1.0148373983739838,
      "grad_norm": 4.2118964195251465,
      "learning_rate": 3.2520325203252037e-05,
      "loss": 0.4376,
      "step": 320
    },
    {
      "epoch": 1.016869918699187,
      "grad_norm": 10.345478057861328,
      "learning_rate": 3.353658536585366e-05,
      "loss": 0.3226,
      "step": 330
    },
    {
      "epoch": 1.0189024390243901,
      "grad_norm": 4.225228786468506,
      "learning_rate": 3.4552845528455286e-05,
      "loss": 0.3029,
      "step": 340
    },
    {
      "epoch": 1.0209349593495936,
      "grad_norm": 11.051701545715332,
      "learning_rate": 3.556910569105692e-05,
      "loss": 0.2655,
      "step": 350
    },
    {
      "epoch": 1.0229674796747967,
      "grad_norm": 11.426584243774414,
      "learning_rate": 3.6585365853658535e-05,
      "loss": 0.4213,
      "step": 360
    },
    {
      "epoch": 1.025,
      "grad_norm": 2.8330910205841064,
      "learning_rate": 3.760162601626017e-05,
      "loss": 0.2909,
      "step": 370
    },
    {
      "epoch": 1.0270325203252033,
      "grad_norm": 4.851552486419678,
      "learning_rate": 3.861788617886179e-05,
      "loss": 0.418,
      "step": 380
    },
    {
      "epoch": 1.0290650406504065,
      "grad_norm": 18.32183265686035,
      "learning_rate": 3.9634146341463416e-05,
      "loss": 0.2941,
      "step": 390
    },
    {
      "epoch": 1.0310975609756097,
      "grad_norm": 5.6820597648620605,
      "learning_rate": 4.065040650406504e-05,
      "loss": 0.5664,
      "step": 400
    },
    {
      "epoch": 1.033130081300813,
      "grad_norm": 2.8251802921295166,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3508,
      "step": 410
    },
    {
      "epoch": 1.0351626016260163,
      "grad_norm": 3.7893946170806885,
      "learning_rate": 4.26829268292683e-05,
      "loss": 0.3236,
      "step": 420
    },
    {
      "epoch": 1.0371951219512194,
      "grad_norm": 1.7446439266204834,
      "learning_rate": 4.369918699186992e-05,
      "loss": 0.1824,
      "step": 430
    },
    {
      "epoch": 1.0392276422764228,
      "grad_norm": 1.4556201696395874,
      "learning_rate": 4.4715447154471546e-05,
      "loss": 0.3866,
      "step": 440
    },
    {
      "epoch": 1.041260162601626,
      "grad_norm": 6.709419250488281,
      "learning_rate": 4.573170731707318e-05,
      "loss": 0.313,
      "step": 450
    },
    {
      "epoch": 1.0432926829268292,
      "grad_norm": 2.0539255142211914,
      "learning_rate": 4.6747967479674795e-05,
      "loss": 0.3885,
      "step": 460
    },
    {
      "epoch": 1.0453252032520326,
      "grad_norm": 2.8965578079223633,
      "learning_rate": 4.776422764227643e-05,
      "loss": 0.2792,
      "step": 470
    },
    {
      "epoch": 1.0473577235772358,
      "grad_norm": 1.761131763458252,
      "learning_rate": 4.878048780487805e-05,
      "loss": 0.2827,
      "step": 480
    },
    {
      "epoch": 1.049390243902439,
      "grad_norm": 2.394094944000244,
      "learning_rate": 4.9796747967479676e-05,
      "loss": 0.3799,
      "step": 490
    },
    {
      "epoch": 1.0502032520325204,
      "eval_accuracy": 0.8042245257187561,
      "eval_loss": 0.4347330331802368,
      "eval_runtime": 3754.5008,
      "eval_samples_per_second": 1.362,
      "eval_steps_per_second": 0.085,
      "step": 494
    },
    {
      "epoch": 2.0012195121951217,
      "grad_norm": 2.0521011352539062,
      "learning_rate": 4.9909665763324305e-05,
      "loss": 0.2797,
      "step": 500
    },
    {
      "epoch": 2.0032520325203254,
      "grad_norm": 2.8973562717437744,
      "learning_rate": 4.9796747967479676e-05,
      "loss": 0.3618,
      "step": 510
    },
    {
      "epoch": 2.0052845528455285,
      "grad_norm": 7.602560043334961,
      "learning_rate": 4.9683830171635054e-05,
      "loss": 0.3044,
      "step": 520
    },
    {
      "epoch": 2.0073170731707317,
      "grad_norm": 11.021784782409668,
      "learning_rate": 4.957091237579043e-05,
      "loss": 0.2191,
      "step": 530
    },
    {
      "epoch": 2.009349593495935,
      "grad_norm": 4.367031097412109,
      "learning_rate": 4.9457994579945803e-05,
      "loss": 0.2548,
      "step": 540
    },
    {
      "epoch": 2.011382113821138,
      "grad_norm": 3.5690765380859375,
      "learning_rate": 4.9345076784101175e-05,
      "loss": 0.3221,
      "step": 550
    },
    {
      "epoch": 2.0134146341463413,
      "grad_norm": 1.8896007537841797,
      "learning_rate": 4.9232158988256546e-05,
      "loss": 0.2813,
      "step": 560
    },
    {
      "epoch": 2.015447154471545,
      "grad_norm": 0.9179109334945679,
      "learning_rate": 4.9119241192411924e-05,
      "loss": 0.3438,
      "step": 570
    },
    {
      "epoch": 2.017479674796748,
      "grad_norm": 2.7764482498168945,
      "learning_rate": 4.90063233965673e-05,
      "loss": 0.3114,
      "step": 580
    },
    {
      "epoch": 2.0195121951219512,
      "grad_norm": 7.815450191497803,
      "learning_rate": 4.8893405600722673e-05,
      "loss": 0.2304,
      "step": 590
    },
    {
      "epoch": 2.0215447154471544,
      "grad_norm": 1.2830098867416382,
      "learning_rate": 4.878048780487805e-05,
      "loss": 0.2185,
      "step": 600
    },
    {
      "epoch": 2.0235772357723576,
      "grad_norm": 3.990074872970581,
      "learning_rate": 4.866757000903342e-05,
      "loss": 0.1856,
      "step": 610
    },
    {
      "epoch": 2.0256097560975608,
      "grad_norm": 6.101552963256836,
      "learning_rate": 4.85546522131888e-05,
      "loss": 0.3284,
      "step": 620
    },
    {
      "epoch": 2.0276422764227644,
      "grad_norm": 2.2341134548187256,
      "learning_rate": 4.844173441734418e-05,
      "loss": 0.2448,
      "step": 630
    },
    {
      "epoch": 2.0296747967479676,
      "grad_norm": 0.1779327094554901,
      "learning_rate": 4.832881662149955e-05,
      "loss": 0.1343,
      "step": 640
    },
    {
      "epoch": 2.0317073170731708,
      "grad_norm": 9.588211059570312,
      "learning_rate": 4.821589882565493e-05,
      "loss": 0.3552,
      "step": 650
    },
    {
      "epoch": 2.033739837398374,
      "grad_norm": 1.7410920858383179,
      "learning_rate": 4.81029810298103e-05,
      "loss": 0.1702,
      "step": 660
    },
    {
      "epoch": 2.035772357723577,
      "grad_norm": 4.3470563888549805,
      "learning_rate": 4.799006323396568e-05,
      "loss": 0.3291,
      "step": 670
    },
    {
      "epoch": 2.0378048780487803,
      "grad_norm": 1.9769459962844849,
      "learning_rate": 4.787714543812105e-05,
      "loss": 0.2525,
      "step": 680
    },
    {
      "epoch": 2.039837398373984,
      "grad_norm": 5.235065460205078,
      "learning_rate": 4.776422764227643e-05,
      "loss": 0.1969,
      "step": 690
    },
    {
      "epoch": 2.041869918699187,
      "grad_norm": 3.3725054264068604,
      "learning_rate": 4.7651309846431805e-05,
      "loss": 0.1377,
      "step": 700
    },
    {
      "epoch": 2.0439024390243903,
      "grad_norm": 3.6034066677093506,
      "learning_rate": 4.7538392050587176e-05,
      "loss": 0.2816,
      "step": 710
    },
    {
      "epoch": 2.0459349593495935,
      "grad_norm": 4.112473964691162,
      "learning_rate": 4.7425474254742554e-05,
      "loss": 0.1848,
      "step": 720
    },
    {
      "epoch": 2.0479674796747966,
      "grad_norm": 1.5043107271194458,
      "learning_rate": 4.731255645889792e-05,
      "loss": 0.294,
      "step": 730
    },
    {
      "epoch": 2.05,
      "grad_norm": 3.0985732078552246,
      "learning_rate": 4.71996386630533e-05,
      "loss": 0.1638,
      "step": 740
    },
    {
      "epoch": 2.0502032520325204,
      "eval_accuracy": 0.6559749657735184,
      "eval_loss": 0.6743700504302979,
      "eval_runtime": 3677.0901,
      "eval_samples_per_second": 1.391,
      "eval_steps_per_second": 0.087,
      "step": 741
    },
    {
      "epoch": 3.001829268292683,
      "grad_norm": 12.086027145385742,
      "learning_rate": 4.7086720867208675e-05,
      "loss": 0.3547,
      "step": 750
    },
    {
      "epoch": 3.003861788617886,
      "grad_norm": 8.689488410949707,
      "learning_rate": 4.6973803071364046e-05,
      "loss": 0.257,
      "step": 760
    },
    {
      "epoch": 3.0058943089430894,
      "grad_norm": 2.267209529876709,
      "learning_rate": 4.6860885275519424e-05,
      "loss": 0.2171,
      "step": 770
    },
    {
      "epoch": 3.0079268292682926,
      "grad_norm": 4.9380784034729,
      "learning_rate": 4.6747967479674795e-05,
      "loss": 0.2579,
      "step": 780
    },
    {
      "epoch": 3.0099593495934958,
      "grad_norm": 3.0001747608184814,
      "learning_rate": 4.6635049683830173e-05,
      "loss": 0.2183,
      "step": 790
    },
    {
      "epoch": 3.0119918699186994,
      "grad_norm": 5.362173557281494,
      "learning_rate": 4.652213188798555e-05,
      "loss": 0.1774,
      "step": 800
    },
    {
      "epoch": 3.0140243902439026,
      "grad_norm": 5.573166847229004,
      "learning_rate": 4.640921409214092e-05,
      "loss": 0.1424,
      "step": 810
    },
    {
      "epoch": 3.0160569105691057,
      "grad_norm": 14.513651847839355,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.1088,
      "step": 820
    },
    {
      "epoch": 3.018089430894309,
      "grad_norm": 8.743453025817871,
      "learning_rate": 4.618337850045167e-05,
      "loss": 0.3518,
      "step": 830
    },
    {
      "epoch": 3.020121951219512,
      "grad_norm": 3.2687177658081055,
      "learning_rate": 4.607046070460705e-05,
      "loss": 0.2987,
      "step": 840
    },
    {
      "epoch": 3.0221544715447153,
      "grad_norm": 8.014973640441895,
      "learning_rate": 4.595754290876242e-05,
      "loss": 0.2568,
      "step": 850
    },
    {
      "epoch": 3.024186991869919,
      "grad_norm": 3.7590761184692383,
      "learning_rate": 4.58446251129178e-05,
      "loss": 0.1455,
      "step": 860
    },
    {
      "epoch": 3.026219512195122,
      "grad_norm": 5.427084922790527,
      "learning_rate": 4.573170731707318e-05,
      "loss": 0.1454,
      "step": 870
    },
    {
      "epoch": 3.0282520325203253,
      "grad_norm": 1.3013925552368164,
      "learning_rate": 4.561878952122855e-05,
      "loss": 0.1122,
      "step": 880
    },
    {
      "epoch": 3.0302845528455284,
      "grad_norm": 2.7502329349517822,
      "learning_rate": 4.550587172538393e-05,
      "loss": 0.1493,
      "step": 890
    },
    {
      "epoch": 3.0323170731707316,
      "grad_norm": 2.820256471633911,
      "learning_rate": 4.53929539295393e-05,
      "loss": 0.153,
      "step": 900
    },
    {
      "epoch": 3.034349593495935,
      "grad_norm": 1.7173501253128052,
      "learning_rate": 4.528003613369467e-05,
      "loss": 0.1442,
      "step": 910
    },
    {
      "epoch": 3.0363821138211384,
      "grad_norm": 0.7276012897491455,
      "learning_rate": 4.516711833785005e-05,
      "loss": 0.0876,
      "step": 920
    },
    {
      "epoch": 3.0384146341463416,
      "grad_norm": 3.6419296264648438,
      "learning_rate": 4.505420054200542e-05,
      "loss": 0.2348,
      "step": 930
    },
    {
      "epoch": 3.040447154471545,
      "grad_norm": 7.711927890777588,
      "learning_rate": 4.49412827461608e-05,
      "loss": 0.1251,
      "step": 940
    },
    {
      "epoch": 3.042479674796748,
      "grad_norm": 1.2444993257522583,
      "learning_rate": 4.482836495031617e-05,
      "loss": 0.1853,
      "step": 950
    },
    {
      "epoch": 3.044512195121951,
      "grad_norm": 0.5793710947036743,
      "learning_rate": 4.4715447154471546e-05,
      "loss": 0.1207,
      "step": 960
    },
    {
      "epoch": 3.0465447154471543,
      "grad_norm": 8.041537284851074,
      "learning_rate": 4.4602529358626924e-05,
      "loss": 0.1407,
      "step": 970
    },
    {
      "epoch": 3.0485772357723575,
      "grad_norm": 1.9752309322357178,
      "learning_rate": 4.4489611562782295e-05,
      "loss": 0.1194,
      "step": 980
    },
    {
      "epoch": 3.0502032520325204,
      "eval_accuracy": 0.8499902210052807,
      "eval_loss": 0.39140766859054565,
      "eval_runtime": 3736.0684,
      "eval_samples_per_second": 1.369,
      "eval_steps_per_second": 0.086,
      "step": 988
    },
    {
      "epoch": 4.00040650406504,
      "grad_norm": 5.747890949249268,
      "learning_rate": 4.4376693766937673e-05,
      "loss": 0.1136,
      "step": 990
    },
    {
      "epoch": 4.0024390243902435,
      "grad_norm": 3.7141377925872803,
      "learning_rate": 4.4263775971093045e-05,
      "loss": 0.0972,
      "step": 1000
    },
    {
      "epoch": 4.0044715447154475,
      "grad_norm": 5.8626837730407715,
      "learning_rate": 4.415085817524842e-05,
      "loss": 0.2382,
      "step": 1010
    },
    {
      "epoch": 4.006504065040651,
      "grad_norm": 2.206000328063965,
      "learning_rate": 4.4037940379403794e-05,
      "loss": 0.1387,
      "step": 1020
    },
    {
      "epoch": 4.008536585365854,
      "grad_norm": 10.47764778137207,
      "learning_rate": 4.392502258355917e-05,
      "loss": 0.1353,
      "step": 1030
    },
    {
      "epoch": 4.010569105691057,
      "grad_norm": 2.0999231338500977,
      "learning_rate": 4.381210478771455e-05,
      "loss": 0.1592,
      "step": 1040
    },
    {
      "epoch": 4.01260162601626,
      "grad_norm": 5.126746654510498,
      "learning_rate": 4.369918699186992e-05,
      "loss": 0.1646,
      "step": 1050
    },
    {
      "epoch": 4.014634146341463,
      "grad_norm": 2.038853645324707,
      "learning_rate": 4.35862691960253e-05,
      "loss": 0.1648,
      "step": 1060
    },
    {
      "epoch": 4.016666666666667,
      "grad_norm": 1.2086607217788696,
      "learning_rate": 4.347335140018067e-05,
      "loss": 0.1447,
      "step": 1070
    },
    {
      "epoch": 4.01869918699187,
      "grad_norm": 0.03065630793571472,
      "learning_rate": 4.336043360433605e-05,
      "loss": 0.1074,
      "step": 1080
    },
    {
      "epoch": 4.020731707317073,
      "grad_norm": 8.712028503417969,
      "learning_rate": 4.324751580849142e-05,
      "loss": 0.3328,
      "step": 1090
    },
    {
      "epoch": 4.022764227642276,
      "grad_norm": 0.872675895690918,
      "learning_rate": 4.313459801264679e-05,
      "loss": 0.1697,
      "step": 1100
    },
    {
      "epoch": 4.024796747967479,
      "grad_norm": 4.265038967132568,
      "learning_rate": 4.302168021680217e-05,
      "loss": 0.1516,
      "step": 1110
    },
    {
      "epoch": 4.0268292682926825,
      "grad_norm": 0.4007105231285095,
      "learning_rate": 4.290876242095754e-05,
      "loss": 0.1247,
      "step": 1120
    },
    {
      "epoch": 4.028861788617887,
      "grad_norm": 0.3496200740337372,
      "learning_rate": 4.279584462511292e-05,
      "loss": 0.1568,
      "step": 1130
    },
    {
      "epoch": 4.03089430894309,
      "grad_norm": 1.902456521987915,
      "learning_rate": 4.26829268292683e-05,
      "loss": 0.1519,
      "step": 1140
    },
    {
      "epoch": 4.032926829268293,
      "grad_norm": 1.0433646440505981,
      "learning_rate": 4.257000903342367e-05,
      "loss": 0.1261,
      "step": 1150
    },
    {
      "epoch": 4.034959349593496,
      "grad_norm": 5.954234600067139,
      "learning_rate": 4.2457091237579046e-05,
      "loss": 0.1753,
      "step": 1160
    },
    {
      "epoch": 4.036991869918699,
      "grad_norm": 1.3946830034255981,
      "learning_rate": 4.234417344173442e-05,
      "loss": 0.1974,
      "step": 1170
    },
    {
      "epoch": 4.0390243902439025,
      "grad_norm": 6.719627857208252,
      "learning_rate": 4.2231255645889795e-05,
      "loss": 0.1158,
      "step": 1180
    },
    {
      "epoch": 4.041056910569106,
      "grad_norm": 0.9860407114028931,
      "learning_rate": 4.211833785004517e-05,
      "loss": 0.0707,
      "step": 1190
    },
    {
      "epoch": 4.043089430894309,
      "grad_norm": 4.813327789306641,
      "learning_rate": 4.2005420054200545e-05,
      "loss": 0.1737,
      "step": 1200
    },
    {
      "epoch": 4.045121951219512,
      "grad_norm": 2.334260940551758,
      "learning_rate": 4.189250225835592e-05,
      "loss": 0.1816,
      "step": 1210
    },
    {
      "epoch": 4.047154471544715,
      "grad_norm": 9.71888542175293,
      "learning_rate": 4.1779584462511294e-05,
      "loss": 0.143,
      "step": 1220
    },
    {
      "epoch": 4.049186991869918,
      "grad_norm": 5.791767120361328,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.2124,
      "step": 1230
    },
    {
      "epoch": 4.05020325203252,
      "eval_accuracy": 0.7001760219049482,
      "eval_loss": 0.8540657162666321,
      "eval_runtime": 3722.1391,
      "eval_samples_per_second": 1.374,
      "eval_steps_per_second": 0.086,
      "step": 1235
    },
    {
      "epoch": 5.001016260162602,
      "grad_norm": 10.609312057495117,
      "learning_rate": 4.155374887082204e-05,
      "loss": 0.1071,
      "step": 1240
    },
    {
      "epoch": 5.003048780487805,
      "grad_norm": 3.1270151138305664,
      "learning_rate": 4.144083107497742e-05,
      "loss": 0.0715,
      "step": 1250
    },
    {
      "epoch": 5.005081300813008,
      "grad_norm": 0.3721328377723694,
      "learning_rate": 4.132791327913279e-05,
      "loss": 0.1614,
      "step": 1260
    },
    {
      "epoch": 5.007113821138211,
      "grad_norm": 1.483250379562378,
      "learning_rate": 4.1214995483288164e-05,
      "loss": 0.1187,
      "step": 1270
    },
    {
      "epoch": 5.009146341463414,
      "grad_norm": 0.02934689074754715,
      "learning_rate": 4.110207768744354e-05,
      "loss": 0.0382,
      "step": 1280
    },
    {
      "epoch": 5.0111788617886175,
      "grad_norm": 1.6646735668182373,
      "learning_rate": 4.098915989159891e-05,
      "loss": 0.1687,
      "step": 1290
    },
    {
      "epoch": 5.013211382113822,
      "grad_norm": 0.043164148926734924,
      "learning_rate": 4.087624209575429e-05,
      "loss": 0.0069,
      "step": 1300
    },
    {
      "epoch": 5.015243902439025,
      "grad_norm": 6.15376615524292,
      "learning_rate": 4.076332429990967e-05,
      "loss": 0.2118,
      "step": 1310
    },
    {
      "epoch": 5.017276422764228,
      "grad_norm": 1.8082466125488281,
      "learning_rate": 4.065040650406504e-05,
      "loss": 0.0623,
      "step": 1320
    },
    {
      "epoch": 5.019308943089431,
      "grad_norm": 0.04419846832752228,
      "learning_rate": 4.053748870822042e-05,
      "loss": 0.0499,
      "step": 1330
    },
    {
      "epoch": 5.021341463414634,
      "grad_norm": 0.39577415585517883,
      "learning_rate": 4.042457091237579e-05,
      "loss": 0.1324,
      "step": 1340
    },
    {
      "epoch": 5.0233739837398375,
      "grad_norm": 0.6317324638366699,
      "learning_rate": 4.031165311653117e-05,
      "loss": 0.1644,
      "step": 1350
    },
    {
      "epoch": 5.025406504065041,
      "grad_norm": 0.21281012892723083,
      "learning_rate": 4.019873532068654e-05,
      "loss": 0.1714,
      "step": 1360
    },
    {
      "epoch": 5.027439024390244,
      "grad_norm": 0.18493518233299255,
      "learning_rate": 4.008581752484192e-05,
      "loss": 0.0842,
      "step": 1370
    },
    {
      "epoch": 5.029471544715447,
      "grad_norm": 37.605682373046875,
      "learning_rate": 3.9972899728997295e-05,
      "loss": 0.114,
      "step": 1380
    },
    {
      "epoch": 5.03150406504065,
      "grad_norm": 3.824547052383423,
      "learning_rate": 3.985998193315267e-05,
      "loss": 0.0298,
      "step": 1390
    },
    {
      "epoch": 5.033536585365853,
      "grad_norm": 0.03947475552558899,
      "learning_rate": 3.9747064137308045e-05,
      "loss": 0.0648,
      "step": 1400
    },
    {
      "epoch": 5.0355691056910565,
      "grad_norm": 3.3510477542877197,
      "learning_rate": 3.9634146341463416e-05,
      "loss": 0.1212,
      "step": 1410
    },
    {
      "epoch": 5.037601626016261,
      "grad_norm": 1.0530486106872559,
      "learning_rate": 3.9521228545618794e-05,
      "loss": 0.0465,
      "step": 1420
    },
    {
      "epoch": 5.039634146341464,
      "grad_norm": 0.38464394211769104,
      "learning_rate": 3.940831074977417e-05,
      "loss": 0.063,
      "step": 1430
    },
    {
      "epoch": 5.041666666666667,
      "grad_norm": 0.1895415186882019,
      "learning_rate": 3.9295392953929537e-05,
      "loss": 0.1084,
      "step": 1440
    },
    {
      "epoch": 5.04369918699187,
      "grad_norm": 14.11963176727295,
      "learning_rate": 3.9182475158084915e-05,
      "loss": 0.0502,
      "step": 1450
    },
    {
      "epoch": 5.045731707317073,
      "grad_norm": 23.175413131713867,
      "learning_rate": 3.9069557362240286e-05,
      "loss": 0.0569,
      "step": 1460
    },
    {
      "epoch": 5.0477642276422765,
      "grad_norm": 2.6622891426086426,
      "learning_rate": 3.8956639566395664e-05,
      "loss": 0.0939,
      "step": 1470
    },
    {
      "epoch": 5.04979674796748,
      "grad_norm": 0.1387495994567871,
      "learning_rate": 3.884372177055104e-05,
      "loss": 0.0289,
      "step": 1480
    },
    {
      "epoch": 5.05020325203252,
      "eval_accuracy": 0.5087033053002151,
      "eval_loss": 2.8893117904663086,
      "eval_runtime": 3665.7363,
      "eval_samples_per_second": 1.395,
      "eval_steps_per_second": 0.087,
      "step": 1482
    },
    {
      "epoch": 6.001626016260163,
      "grad_norm": 0.06576679646968842,
      "learning_rate": 3.873080397470641e-05,
      "loss": 0.2386,
      "step": 1490
    },
    {
      "epoch": 6.003658536585366,
      "grad_norm": 0.03989925608038902,
      "learning_rate": 3.861788617886179e-05,
      "loss": 0.0548,
      "step": 1500
    },
    {
      "epoch": 6.005691056910569,
      "grad_norm": 0.2090814858675003,
      "learning_rate": 3.850496838301716e-05,
      "loss": 0.1741,
      "step": 1510
    },
    {
      "epoch": 6.007723577235772,
      "grad_norm": 0.0932309478521347,
      "learning_rate": 3.839205058717254e-05,
      "loss": 0.0184,
      "step": 1520
    },
    {
      "epoch": 6.009756097560976,
      "grad_norm": 5.776859283447266,
      "learning_rate": 3.827913279132791e-05,
      "loss": 0.1172,
      "step": 1530
    },
    {
      "epoch": 6.011788617886179,
      "grad_norm": 2.6321611404418945,
      "learning_rate": 3.816621499548329e-05,
      "loss": 0.1442,
      "step": 1540
    },
    {
      "epoch": 6.013821138211382,
      "grad_norm": 4.031448841094971,
      "learning_rate": 3.805329719963867e-05,
      "loss": 0.0846,
      "step": 1550
    },
    {
      "epoch": 6.015853658536585,
      "grad_norm": 2.18776535987854,
      "learning_rate": 3.794037940379404e-05,
      "loss": 0.027,
      "step": 1560
    },
    {
      "epoch": 6.017886178861788,
      "grad_norm": 0.1803622543811798,
      "learning_rate": 3.782746160794942e-05,
      "loss": 0.0699,
      "step": 1570
    },
    {
      "epoch": 6.0199186991869915,
      "grad_norm": 0.05358570069074631,
      "learning_rate": 3.771454381210479e-05,
      "loss": 0.1084,
      "step": 1580
    },
    {
      "epoch": 6.021951219512195,
      "grad_norm": 0.20270071923732758,
      "learning_rate": 3.760162601626017e-05,
      "loss": 0.0367,
      "step": 1590
    },
    {
      "epoch": 6.023983739837399,
      "grad_norm": 0.048851825296878815,
      "learning_rate": 3.7488708220415545e-05,
      "loss": 0.1415,
      "step": 1600
    },
    {
      "epoch": 6.026016260162602,
      "grad_norm": 0.0154234878718853,
      "learning_rate": 3.7375790424570916e-05,
      "loss": 0.0221,
      "step": 1610
    },
    {
      "epoch": 6.028048780487805,
      "grad_norm": 2.328122615814209,
      "learning_rate": 3.726287262872629e-05,
      "loss": 0.1567,
      "step": 1620
    },
    {
      "epoch": 6.030081300813008,
      "grad_norm": 5.369495391845703,
      "learning_rate": 3.714995483288166e-05,
      "loss": 0.1395,
      "step": 1630
    },
    {
      "epoch": 6.0321138211382115,
      "grad_norm": 14.819968223571777,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2039,
      "step": 1640
    },
    {
      "epoch": 6.034146341463415,
      "grad_norm": 4.830562591552734,
      "learning_rate": 3.6924119241192415e-05,
      "loss": 0.143,
      "step": 1650
    },
    {
      "epoch": 6.036178861788618,
      "grad_norm": 7.940253257751465,
      "learning_rate": 3.6811201445347786e-05,
      "loss": 0.0578,
      "step": 1660
    },
    {
      "epoch": 6.038211382113821,
      "grad_norm": 1.155714511871338,
      "learning_rate": 3.6698283649503164e-05,
      "loss": 0.1233,
      "step": 1670
    },
    {
      "epoch": 6.040243902439024,
      "grad_norm": 0.5576316714286804,
      "learning_rate": 3.6585365853658535e-05,
      "loss": 0.1033,
      "step": 1680
    },
    {
      "epoch": 6.042276422764227,
      "grad_norm": 0.2718254327774048,
      "learning_rate": 3.647244805781391e-05,
      "loss": 0.1221,
      "step": 1690
    },
    {
      "epoch": 6.044308943089431,
      "grad_norm": 0.664254367351532,
      "learning_rate": 3.635953026196929e-05,
      "loss": 0.054,
      "step": 1700
    },
    {
      "epoch": 6.046341463414634,
      "grad_norm": 10.992645263671875,
      "learning_rate": 3.624661246612466e-05,
      "loss": 0.0684,
      "step": 1710
    },
    {
      "epoch": 6.048373983739838,
      "grad_norm": 0.9990622997283936,
      "learning_rate": 3.613369467028004e-05,
      "loss": 0.0909,
      "step": 1720
    },
    {
      "epoch": 6.05020325203252,
      "eval_accuracy": 0.6821826716213574,
      "eval_loss": 1.3700344562530518,
      "eval_runtime": 3801.0542,
      "eval_samples_per_second": 1.345,
      "eval_steps_per_second": 0.084,
      "step": 1729
    },
    {
      "epoch": 7.00020325203252,
      "grad_norm": 0.10866906493902206,
      "learning_rate": 3.602077687443541e-05,
      "loss": 0.0921,
      "step": 1730
    },
    {
      "epoch": 7.002235772357723,
      "grad_norm": 0.4908745288848877,
      "learning_rate": 3.590785907859079e-05,
      "loss": 0.0669,
      "step": 1740
    },
    {
      "epoch": 7.0042682926829265,
      "grad_norm": 0.3369530439376831,
      "learning_rate": 3.579494128274616e-05,
      "loss": 0.0676,
      "step": 1750
    },
    {
      "epoch": 7.00630081300813,
      "grad_norm": 0.08251447230577469,
      "learning_rate": 3.568202348690154e-05,
      "loss": 0.0729,
      "step": 1760
    },
    {
      "epoch": 7.008333333333334,
      "grad_norm": 0.05544485151767731,
      "learning_rate": 3.556910569105692e-05,
      "loss": 0.051,
      "step": 1770
    },
    {
      "epoch": 7.010365853658537,
      "grad_norm": 0.04277024790644646,
      "learning_rate": 3.545618789521229e-05,
      "loss": 0.0047,
      "step": 1780
    },
    {
      "epoch": 7.01239837398374,
      "grad_norm": 0.024971861392259598,
      "learning_rate": 3.534327009936767e-05,
      "loss": 0.0744,
      "step": 1790
    },
    {
      "epoch": 7.014430894308943,
      "grad_norm": 0.034789636731147766,
      "learning_rate": 3.523035230352303e-05,
      "loss": 0.1223,
      "step": 1800
    },
    {
      "epoch": 7.0164634146341465,
      "grad_norm": 0.06158401072025299,
      "learning_rate": 3.511743450767841e-05,
      "loss": 0.0142,
      "step": 1810
    },
    {
      "epoch": 7.01849593495935,
      "grad_norm": 15.455686569213867,
      "learning_rate": 3.500451671183379e-05,
      "loss": 0.132,
      "step": 1820
    },
    {
      "epoch": 7.020528455284553,
      "grad_norm": 0.04913310706615448,
      "learning_rate": 3.489159891598916e-05,
      "loss": 0.1211,
      "step": 1830
    },
    {
      "epoch": 7.022560975609756,
      "grad_norm": 17.346275329589844,
      "learning_rate": 3.4778681120144537e-05,
      "loss": 0.0489,
      "step": 1840
    },
    {
      "epoch": 7.024593495934959,
      "grad_norm": 7.290595531463623,
      "learning_rate": 3.466576332429991e-05,
      "loss": 0.0612,
      "step": 1850
    },
    {
      "epoch": 7.026626016260162,
      "grad_norm": 0.7925562262535095,
      "learning_rate": 3.4552845528455286e-05,
      "loss": 0.0708,
      "step": 1860
    },
    {
      "epoch": 7.0286585365853655,
      "grad_norm": 0.03101346269249916,
      "learning_rate": 3.4439927732610664e-05,
      "loss": 0.0393,
      "step": 1870
    },
    {
      "epoch": 7.030691056910569,
      "grad_norm": 0.07088115811347961,
      "learning_rate": 3.4327009936766035e-05,
      "loss": 0.0056,
      "step": 1880
    },
    {
      "epoch": 7.032723577235773,
      "grad_norm": 0.6128086447715759,
      "learning_rate": 3.421409214092141e-05,
      "loss": 0.0651,
      "step": 1890
    },
    {
      "epoch": 7.034756097560976,
      "grad_norm": 3.2570462226867676,
      "learning_rate": 3.4101174345076785e-05,
      "loss": 0.1438,
      "step": 1900
    },
    {
      "epoch": 7.036788617886179,
      "grad_norm": 9.589098930358887,
      "learning_rate": 3.398825654923216e-05,
      "loss": 0.1255,
      "step": 1910
    },
    {
      "epoch": 7.038821138211382,
      "grad_norm": 0.026409929618239403,
      "learning_rate": 3.3875338753387534e-05,
      "loss": 0.0259,
      "step": 1920
    },
    {
      "epoch": 7.0408536585365855,
      "grad_norm": 8.720446586608887,
      "learning_rate": 3.376242095754291e-05,
      "loss": 0.0683,
      "step": 1930
    },
    {
      "epoch": 7.042886178861789,
      "grad_norm": 0.02734554372727871,
      "learning_rate": 3.364950316169829e-05,
      "loss": 0.2492,
      "step": 1940
    },
    {
      "epoch": 7.044918699186992,
      "grad_norm": 0.02654699981212616,
      "learning_rate": 3.353658536585366e-05,
      "loss": 0.0342,
      "step": 1950
    },
    {
      "epoch": 7.046951219512195,
      "grad_norm": 0.04390241950750351,
      "learning_rate": 3.342366757000904e-05,
      "loss": 0.0928,
      "step": 1960
    },
    {
      "epoch": 7.048983739837398,
      "grad_norm": 0.41492825746536255,
      "learning_rate": 3.331074977416441e-05,
      "loss": 0.0729,
      "step": 1970
    },
    {
      "epoch": 7.05020325203252,
      "eval_accuracy": 0.6671230197535694,
      "eval_loss": 1.4458717107772827,
      "eval_runtime": 3662.3026,
      "eval_samples_per_second": 1.396,
      "eval_steps_per_second": 0.087,
      "step": 1976
    },
    {
      "epoch": 8.00081300813008,
      "grad_norm": 0.02490147575736046,
      "learning_rate": 3.319783197831978e-05,
      "loss": 0.0047,
      "step": 1980
    },
    {
      "epoch": 8.002845528455284,
      "grad_norm": 4.008778095245361,
      "learning_rate": 3.308491418247516e-05,
      "loss": 0.0935,
      "step": 1990
    },
    {
      "epoch": 8.004878048780487,
      "grad_norm": 11.095658302307129,
      "learning_rate": 3.297199638663053e-05,
      "loss": 0.0738,
      "step": 2000
    },
    {
      "epoch": 8.006910569105692,
      "grad_norm": 0.5240073800086975,
      "learning_rate": 3.285907859078591e-05,
      "loss": 0.0266,
      "step": 2010
    },
    {
      "epoch": 8.008943089430895,
      "grad_norm": 7.145389080047607,
      "learning_rate": 3.274616079494128e-05,
      "loss": 0.0837,
      "step": 2020
    },
    {
      "epoch": 8.010975609756098,
      "grad_norm": 21.071447372436523,
      "learning_rate": 3.263324299909666e-05,
      "loss": 0.0519,
      "step": 2030
    },
    {
      "epoch": 8.013008130081301,
      "grad_norm": 7.303457260131836,
      "learning_rate": 3.2520325203252037e-05,
      "loss": 0.0518,
      "step": 2040
    },
    {
      "epoch": 8.015040650406505,
      "grad_norm": 0.10021180659532547,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.0728,
      "step": 2050
    },
    {
      "epoch": 8.017073170731708,
      "grad_norm": 0.02886616811156273,
      "learning_rate": 3.2294489611562786e-05,
      "loss": 0.101,
      "step": 2060
    },
    {
      "epoch": 8.019105691056911,
      "grad_norm": 1.6222566366195679,
      "learning_rate": 3.218157181571816e-05,
      "loss": 0.0473,
      "step": 2070
    },
    {
      "epoch": 8.021138211382114,
      "grad_norm": 0.023972176015377045,
      "learning_rate": 3.2068654019873535e-05,
      "loss": 0.0425,
      "step": 2080
    },
    {
      "epoch": 8.023170731707317,
      "grad_norm": 21.911823272705078,
      "learning_rate": 3.1955736224028906e-05,
      "loss": 0.0675,
      "step": 2090
    },
    {
      "epoch": 8.02520325203252,
      "grad_norm": 0.04748200625181198,
      "learning_rate": 3.1842818428184285e-05,
      "loss": 0.003,
      "step": 2100
    },
    {
      "epoch": 8.027235772357724,
      "grad_norm": 0.008865147829055786,
      "learning_rate": 3.172990063233966e-05,
      "loss": 0.027,
      "step": 2110
    },
    {
      "epoch": 8.029268292682927,
      "grad_norm": 0.03952307626605034,
      "learning_rate": 3.1616982836495034e-05,
      "loss": 0.0297,
      "step": 2120
    },
    {
      "epoch": 8.03130081300813,
      "grad_norm": 0.0076277609914541245,
      "learning_rate": 3.150406504065041e-05,
      "loss": 0.0377,
      "step": 2130
    },
    {
      "epoch": 8.033333333333333,
      "grad_norm": 0.008851220831274986,
      "learning_rate": 3.139114724480578e-05,
      "loss": 0.0438,
      "step": 2140
    },
    {
      "epoch": 8.035365853658536,
      "grad_norm": 0.011797068640589714,
      "learning_rate": 3.127822944896116e-05,
      "loss": 0.1529,
      "step": 2150
    },
    {
      "epoch": 8.03739837398374,
      "grad_norm": 2.0825767517089844,
      "learning_rate": 3.116531165311653e-05,
      "loss": 0.0888,
      "step": 2160
    },
    {
      "epoch": 8.039430894308943,
      "grad_norm": 0.07822691649198532,
      "learning_rate": 3.1052393857271904e-05,
      "loss": 0.034,
      "step": 2170
    },
    {
      "epoch": 8.041463414634146,
      "grad_norm": 0.02826240472495556,
      "learning_rate": 3.093947606142728e-05,
      "loss": 0.094,
      "step": 2180
    },
    {
      "epoch": 8.04349593495935,
      "grad_norm": 9.466514587402344,
      "learning_rate": 3.082655826558265e-05,
      "loss": 0.0636,
      "step": 2190
    },
    {
      "epoch": 8.045528455284552,
      "grad_norm": 9.791631698608398,
      "learning_rate": 3.071364046973803e-05,
      "loss": 0.0792,
      "step": 2200
    },
    {
      "epoch": 8.047560975609755,
      "grad_norm": 0.750911295413971,
      "learning_rate": 3.060072267389341e-05,
      "loss": 0.0752,
      "step": 2210
    },
    {
      "epoch": 8.049593495934959,
      "grad_norm": 0.14496035873889923,
      "learning_rate": 3.048780487804878e-05,
      "loss": 0.0122,
      "step": 2220
    },
    {
      "epoch": 8.05020325203252,
      "eval_accuracy": 0.6381771953843145,
      "eval_loss": 1.9108223915100098,
      "eval_runtime": 7236.2994,
      "eval_samples_per_second": 0.707,
      "eval_steps_per_second": 0.044,
      "step": 2223
    },
    {
      "epoch": 9.001422764227643,
      "grad_norm": 0.05617127940058708,
      "learning_rate": 3.0374887082204155e-05,
      "loss": 0.0025,
      "step": 2230
    },
    {
      "epoch": 9.003455284552846,
      "grad_norm": 0.026429414749145508,
      "learning_rate": 3.0261969286359533e-05,
      "loss": 0.0061,
      "step": 2240
    },
    {
      "epoch": 9.00548780487805,
      "grad_norm": 0.10820137709379196,
      "learning_rate": 3.0149051490514908e-05,
      "loss": 0.0242,
      "step": 2250
    },
    {
      "epoch": 9.007520325203252,
      "grad_norm": 0.011124173179268837,
      "learning_rate": 3.0036133694670282e-05,
      "loss": 0.0155,
      "step": 2260
    },
    {
      "epoch": 9.009552845528455,
      "grad_norm": 0.0096007464453578,
      "learning_rate": 2.9923215898825657e-05,
      "loss": 0.0031,
      "step": 2270
    },
    {
      "epoch": 9.011585365853659,
      "grad_norm": 0.009145950898528099,
      "learning_rate": 2.9810298102981032e-05,
      "loss": 0.0047,
      "step": 2280
    },
    {
      "epoch": 9.013617886178862,
      "grad_norm": 0.018553193658590317,
      "learning_rate": 2.9697380307136406e-05,
      "loss": 0.078,
      "step": 2290
    },
    {
      "epoch": 9.015650406504065,
      "grad_norm": 0.8615335822105408,
      "learning_rate": 2.958446251129178e-05,
      "loss": 0.0169,
      "step": 2300
    },
    {
      "epoch": 9.017682926829268,
      "grad_norm": 4.962747097015381,
      "learning_rate": 2.947154471544716e-05,
      "loss": 0.0871,
      "step": 2310
    },
    {
      "epoch": 9.019715447154471,
      "grad_norm": 0.11099843680858612,
      "learning_rate": 2.9358626919602534e-05,
      "loss": 0.0257,
      "step": 2320
    },
    {
      "epoch": 9.021747967479675,
      "grad_norm": 6.154956817626953,
      "learning_rate": 2.924570912375791e-05,
      "loss": 0.094,
      "step": 2330
    },
    {
      "epoch": 9.023780487804878,
      "grad_norm": 0.04944484308362007,
      "learning_rate": 2.9132791327913276e-05,
      "loss": 0.003,
      "step": 2340
    },
    {
      "epoch": 9.025813008130081,
      "grad_norm": 0.019102057442069054,
      "learning_rate": 2.9019873532068654e-05,
      "loss": 0.0053,
      "step": 2350
    },
    {
      "epoch": 9.027845528455284,
      "grad_norm": 0.05737742409110069,
      "learning_rate": 2.890695573622403e-05,
      "loss": 0.0093,
      "step": 2360
    },
    {
      "epoch": 9.029878048780487,
      "grad_norm": 0.012348340824246407,
      "learning_rate": 2.8794037940379404e-05,
      "loss": 0.0971,
      "step": 2370
    },
    {
      "epoch": 9.03191056910569,
      "grad_norm": 0.017456533387303352,
      "learning_rate": 2.868112014453478e-05,
      "loss": 0.0431,
      "step": 2380
    },
    {
      "epoch": 9.033943089430894,
      "grad_norm": 0.0686996802687645,
      "learning_rate": 2.8568202348690153e-05,
      "loss": 0.0153,
      "step": 2390
    },
    {
      "epoch": 9.035975609756097,
      "grad_norm": 7.5223798751831055,
      "learning_rate": 2.8455284552845528e-05,
      "loss": 0.0555,
      "step": 2400
    },
    {
      "epoch": 9.0380081300813,
      "grad_norm": 8.943709373474121,
      "learning_rate": 2.8342366757000906e-05,
      "loss": 0.0577,
      "step": 2410
    },
    {
      "epoch": 9.040040650406503,
      "grad_norm": 0.04952160641551018,
      "learning_rate": 2.822944896115628e-05,
      "loss": 0.0483,
      "step": 2420
    },
    {
      "epoch": 9.042073170731708,
      "grad_norm": 0.05472104996442795,
      "learning_rate": 2.8116531165311655e-05,
      "loss": 0.0447,
      "step": 2430
    },
    {
      "epoch": 9.044105691056911,
      "grad_norm": 0.03964385762810707,
      "learning_rate": 2.800361336946703e-05,
      "loss": 0.0798,
      "step": 2440
    },
    {
      "epoch": 9.046138211382114,
      "grad_norm": 0.025919150561094284,
      "learning_rate": 2.7890695573622404e-05,
      "loss": 0.0527,
      "step": 2450
    },
    {
      "epoch": 9.048170731707318,
      "grad_norm": 11.724416732788086,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.0328,
      "step": 2460
    },
    {
      "epoch": 9.05020325203252,
      "grad_norm": 0.016226714476943016,
      "learning_rate": 2.7664859981933157e-05,
      "loss": 0.0269,
      "step": 2470
    },
    {
      "epoch": 9.05020325203252,
      "eval_accuracy": 0.5734402503422649,
      "eval_loss": 2.4834823608398438,
      "eval_runtime": 3560.2246,
      "eval_samples_per_second": 1.436,
      "eval_steps_per_second": 0.09,
      "step": 2470
    },
    {
      "epoch": 10.002032520325203,
      "grad_norm": 0.00729691656306386,
      "learning_rate": 2.7551942186088532e-05,
      "loss": 0.0393,
      "step": 2480
    },
    {
      "epoch": 10.004065040650406,
      "grad_norm": 0.01799885556101799,
      "learning_rate": 2.7439024390243906e-05,
      "loss": 0.0192,
      "step": 2490
    },
    {
      "epoch": 10.00609756097561,
      "grad_norm": 0.045552246272563934,
      "learning_rate": 2.732610659439928e-05,
      "loss": 0.0671,
      "step": 2500
    },
    {
      "epoch": 10.008130081300813,
      "grad_norm": 0.060591671615839005,
      "learning_rate": 2.7213188798554652e-05,
      "loss": 0.0028,
      "step": 2510
    },
    {
      "epoch": 10.010162601626016,
      "grad_norm": 0.06287448108196259,
      "learning_rate": 2.7100271002710027e-05,
      "loss": 0.0052,
      "step": 2520
    },
    {
      "epoch": 10.012195121951219,
      "grad_norm": 0.01657436043024063,
      "learning_rate": 2.6987353206865402e-05,
      "loss": 0.0029,
      "step": 2530
    },
    {
      "epoch": 10.014227642276422,
      "grad_norm": 0.010985158383846283,
      "learning_rate": 2.6874435411020776e-05,
      "loss": 0.0249,
      "step": 2540
    },
    {
      "epoch": 10.016260162601625,
      "grad_norm": 20.755332946777344,
      "learning_rate": 2.676151761517615e-05,
      "loss": 0.0781,
      "step": 2550
    },
    {
      "epoch": 10.018292682926829,
      "grad_norm": 0.01775680109858513,
      "learning_rate": 2.6648599819331526e-05,
      "loss": 0.0997,
      "step": 2560
    },
    {
      "epoch": 10.020325203252032,
      "grad_norm": 0.007071224506944418,
      "learning_rate": 2.65356820234869e-05,
      "loss": 0.0359,
      "step": 2570
    },
    {
      "epoch": 10.022357723577235,
      "grad_norm": 0.01726052351295948,
      "learning_rate": 2.642276422764228e-05,
      "loss": 0.0365,
      "step": 2580
    },
    {
      "epoch": 10.024390243902438,
      "grad_norm": 2.923809766769409,
      "learning_rate": 2.6309846431797653e-05,
      "loss": 0.0093,
      "step": 2590
    },
    {
      "epoch": 10.026422764227643,
      "grad_norm": 0.07915239036083221,
      "learning_rate": 2.6196928635953028e-05,
      "loss": 0.0092,
      "step": 2600
    },
    {
      "epoch": 10.028455284552846,
      "grad_norm": 0.012806827202439308,
      "learning_rate": 2.6084010840108402e-05,
      "loss": 0.0265,
      "step": 2610
    },
    {
      "epoch": 10.03048780487805,
      "grad_norm": 12.730127334594727,
      "learning_rate": 2.5971093044263777e-05,
      "loss": 0.076,
      "step": 2620
    },
    {
      "epoch": 10.032520325203253,
      "grad_norm": 0.03381958231329918,
      "learning_rate": 2.5858175248419152e-05,
      "loss": 0.044,
      "step": 2630
    },
    {
      "epoch": 10.034552845528456,
      "grad_norm": 11.684189796447754,
      "learning_rate": 2.574525745257453e-05,
      "loss": 0.08,
      "step": 2640
    },
    {
      "epoch": 10.036585365853659,
      "grad_norm": 0.16794830560684204,
      "learning_rate": 2.5632339656729904e-05,
      "loss": 0.0144,
      "step": 2650
    },
    {
      "epoch": 10.038617886178862,
      "grad_norm": 0.02607915922999382,
      "learning_rate": 2.551942186088528e-05,
      "loss": 0.0173,
      "step": 2660
    },
    {
      "epoch": 10.040650406504065,
      "grad_norm": 0.010366716422140598,
      "learning_rate": 2.5406504065040654e-05,
      "loss": 0.0496,
      "step": 2670
    },
    {
      "epoch": 10.042682926829269,
      "grad_norm": 0.024410279467701912,
      "learning_rate": 2.529358626919603e-05,
      "loss": 0.0741,
      "step": 2680
    },
    {
      "epoch": 10.044715447154472,
      "grad_norm": 0.012587165459990501,
      "learning_rate": 2.51806684733514e-05,
      "loss": 0.0014,
      "step": 2690
    },
    {
      "epoch": 10.046747967479675,
      "grad_norm": 0.03345547243952751,
      "learning_rate": 2.5067750677506774e-05,
      "loss": 0.0857,
      "step": 2700
    },
    {
      "epoch": 10.048780487804878,
      "grad_norm": 0.0314403772354126,
      "learning_rate": 2.4954832881662152e-05,
      "loss": 0.0659,
      "step": 2710
    },
    {
      "epoch": 10.05020325203252,
      "eval_accuracy": 0.6111871699589282,
      "eval_loss": 2.8008034229278564,
      "eval_runtime": 3817.3937,
      "eval_samples_per_second": 1.339,
      "eval_steps_per_second": 0.084,
      "step": 2717
    },
    {
      "epoch": 11.00060975609756,
      "grad_norm": 0.32117655873298645,
      "learning_rate": 2.4841915085817527e-05,
      "loss": 0.05,
      "step": 2720
    },
    {
      "epoch": 11.002642276422764,
      "grad_norm": 0.024879859760403633,
      "learning_rate": 2.4728997289972902e-05,
      "loss": 0.0628,
      "step": 2730
    },
    {
      "epoch": 11.004674796747967,
      "grad_norm": 5.877398490905762,
      "learning_rate": 2.4616079494128273e-05,
      "loss": 0.0366,
      "step": 2740
    },
    {
      "epoch": 11.00670731707317,
      "grad_norm": 0.012381893582642078,
      "learning_rate": 2.450316169828365e-05,
      "loss": 0.0016,
      "step": 2750
    },
    {
      "epoch": 11.008739837398373,
      "grad_norm": 1.0652272701263428,
      "learning_rate": 2.4390243902439026e-05,
      "loss": 0.0015,
      "step": 2760
    },
    {
      "epoch": 11.010772357723578,
      "grad_norm": 0.6546475291252136,
      "learning_rate": 2.42773261065944e-05,
      "loss": 0.019,
      "step": 2770
    },
    {
      "epoch": 11.012804878048781,
      "grad_norm": 0.015929026529192924,
      "learning_rate": 2.4164408310749775e-05,
      "loss": 0.0412,
      "step": 2780
    },
    {
      "epoch": 11.014837398373984,
      "grad_norm": 11.400144577026367,
      "learning_rate": 2.405149051490515e-05,
      "loss": 0.0207,
      "step": 2790
    },
    {
      "epoch": 11.016869918699188,
      "grad_norm": 18.641632080078125,
      "learning_rate": 2.3938572719060524e-05,
      "loss": 0.1107,
      "step": 2800
    },
    {
      "epoch": 11.01890243902439,
      "grad_norm": 0.07635531574487686,
      "learning_rate": 2.3825654923215902e-05,
      "loss": 0.0365,
      "step": 2810
    },
    {
      "epoch": 11.020934959349594,
      "grad_norm": 0.02457919530570507,
      "learning_rate": 2.3712737127371277e-05,
      "loss": 0.1156,
      "step": 2820
    },
    {
      "epoch": 11.022967479674797,
      "grad_norm": 6.371166229248047,
      "learning_rate": 2.359981933152665e-05,
      "loss": 0.0452,
      "step": 2830
    },
    {
      "epoch": 11.025,
      "grad_norm": 0.033828433603048325,
      "learning_rate": 2.3486901535682023e-05,
      "loss": 0.0583,
      "step": 2840
    },
    {
      "epoch": 11.027032520325204,
      "grad_norm": 0.012535388581454754,
      "learning_rate": 2.3373983739837398e-05,
      "loss": 0.0543,
      "step": 2850
    },
    {
      "epoch": 11.029065040650407,
      "grad_norm": 0.038947369903326035,
      "learning_rate": 2.3261065943992776e-05,
      "loss": 0.0453,
      "step": 2860
    },
    {
      "epoch": 11.03109756097561,
      "grad_norm": 13.507746696472168,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.0106,
      "step": 2870
    },
    {
      "epoch": 11.033130081300813,
      "grad_norm": 2.2146706581115723,
      "learning_rate": 2.3035230352303525e-05,
      "loss": 0.1362,
      "step": 2880
    },
    {
      "epoch": 11.035162601626016,
      "grad_norm": 0.06082390248775482,
      "learning_rate": 2.29223125564589e-05,
      "loss": 0.0256,
      "step": 2890
    },
    {
      "epoch": 11.03719512195122,
      "grad_norm": 0.08451150357723236,
      "learning_rate": 2.2809394760614274e-05,
      "loss": 0.0192,
      "step": 2900
    },
    {
      "epoch": 11.039227642276423,
      "grad_norm": 0.2093670517206192,
      "learning_rate": 2.269647696476965e-05,
      "loss": 0.0102,
      "step": 2910
    },
    {
      "epoch": 11.041260162601626,
      "grad_norm": 3.509871244430542,
      "learning_rate": 2.2583559168925024e-05,
      "loss": 0.0531,
      "step": 2920
    },
    {
      "epoch": 11.043292682926829,
      "grad_norm": 0.010888086631894112,
      "learning_rate": 2.24706413730804e-05,
      "loss": 0.0313,
      "step": 2930
    },
    {
      "epoch": 11.045325203252032,
      "grad_norm": 0.0575355663895607,
      "learning_rate": 2.2357723577235773e-05,
      "loss": 0.002,
      "step": 2940
    },
    {
      "epoch": 11.047357723577235,
      "grad_norm": 0.012483473867177963,
      "learning_rate": 2.2244805781391148e-05,
      "loss": 0.0423,
      "step": 2950
    },
    {
      "epoch": 11.049390243902439,
      "grad_norm": 0.05652502179145813,
      "learning_rate": 2.2131887985546522e-05,
      "loss": 0.0371,
      "step": 2960
    },
    {
      "epoch": 11.05020325203252,
      "eval_accuracy": 0.5996479561901037,
      "eval_loss": 2.7996442317962646,
      "eval_runtime": 3510.4284,
      "eval_samples_per_second": 1.457,
      "eval_steps_per_second": 0.091,
      "step": 2964
    },
    {
      "epoch": 12.001219512195123,
      "grad_norm": 13.35235595703125,
      "learning_rate": 2.2018970189701897e-05,
      "loss": 0.008,
      "step": 2970
    },
    {
      "epoch": 12.003252032520326,
      "grad_norm": 5.779291152954102,
      "learning_rate": 2.1906052393857275e-05,
      "loss": 0.0036,
      "step": 2980
    },
    {
      "epoch": 12.005284552845529,
      "grad_norm": 17.436155319213867,
      "learning_rate": 2.179313459801265e-05,
      "loss": 0.1074,
      "step": 2990
    },
    {
      "epoch": 12.007317073170732,
      "grad_norm": 0.018812980502843857,
      "learning_rate": 2.1680216802168024e-05,
      "loss": 0.0104,
      "step": 3000
    },
    {
      "epoch": 12.009349593495935,
      "grad_norm": 0.017718536779284477,
      "learning_rate": 2.1567299006323396e-05,
      "loss": 0.0439,
      "step": 3010
    },
    {
      "epoch": 12.011382113821139,
      "grad_norm": 0.021905813366174698,
      "learning_rate": 2.145438121047877e-05,
      "loss": 0.0298,
      "step": 3020
    },
    {
      "epoch": 12.013414634146342,
      "grad_norm": 0.006349028553813696,
      "learning_rate": 2.134146341463415e-05,
      "loss": 0.0009,
      "step": 3030
    },
    {
      "epoch": 12.015447154471545,
      "grad_norm": 0.010411541908979416,
      "learning_rate": 2.1228545618789523e-05,
      "loss": 0.0193,
      "step": 3040
    },
    {
      "epoch": 12.017479674796748,
      "grad_norm": 0.02618451602756977,
      "learning_rate": 2.1115627822944898e-05,
      "loss": 0.0032,
      "step": 3050
    },
    {
      "epoch": 12.019512195121951,
      "grad_norm": 0.004920827690511942,
      "learning_rate": 2.1002710027100272e-05,
      "loss": 0.0727,
      "step": 3060
    },
    {
      "epoch": 12.021544715447154,
      "grad_norm": 0.010978104546666145,
      "learning_rate": 2.0889792231255647e-05,
      "loss": 0.0049,
      "step": 3070
    },
    {
      "epoch": 12.023577235772358,
      "grad_norm": 0.006544812116771936,
      "learning_rate": 2.077687443541102e-05,
      "loss": 0.0479,
      "step": 3080
    },
    {
      "epoch": 12.02560975609756,
      "grad_norm": 0.17064248025417328,
      "learning_rate": 2.0663956639566396e-05,
      "loss": 0.0131,
      "step": 3090
    },
    {
      "epoch": 12.027642276422764,
      "grad_norm": 0.021975887939333916,
      "learning_rate": 2.055103884372177e-05,
      "loss": 0.0031,
      "step": 3100
    },
    {
      "epoch": 12.029674796747967,
      "grad_norm": 0.010594031773507595,
      "learning_rate": 2.0438121047877146e-05,
      "loss": 0.0009,
      "step": 3110
    },
    {
      "epoch": 12.03170731707317,
      "grad_norm": 0.006409040652215481,
      "learning_rate": 2.032520325203252e-05,
      "loss": 0.0399,
      "step": 3120
    },
    {
      "epoch": 12.033739837398373,
      "grad_norm": 0.013711044564843178,
      "learning_rate": 2.0212285456187895e-05,
      "loss": 0.0207,
      "step": 3130
    },
    {
      "epoch": 12.035772357723577,
      "grad_norm": 0.16065192222595215,
      "learning_rate": 2.009936766034327e-05,
      "loss": 0.0487,
      "step": 3140
    },
    {
      "epoch": 12.03780487804878,
      "grad_norm": 0.04054383561015129,
      "learning_rate": 1.9986449864498648e-05,
      "loss": 0.005,
      "step": 3150
    },
    {
      "epoch": 12.039837398373983,
      "grad_norm": 2.773172616958618,
      "learning_rate": 1.9873532068654022e-05,
      "loss": 0.0627,
      "step": 3160
    },
    {
      "epoch": 12.041869918699186,
      "grad_norm": 0.02293955534696579,
      "learning_rate": 1.9760614272809397e-05,
      "loss": 0.0033,
      "step": 3170
    },
    {
      "epoch": 12.04390243902439,
      "grad_norm": 0.032223694026470184,
      "learning_rate": 1.9647696476964768e-05,
      "loss": 0.0296,
      "step": 3180
    },
    {
      "epoch": 12.045934959349594,
      "grad_norm": 11.472624778747559,
      "learning_rate": 1.9534778681120143e-05,
      "loss": 0.0209,
      "step": 3190
    },
    {
      "epoch": 12.047967479674798,
      "grad_norm": 0.014034503139555454,
      "learning_rate": 1.942186088527552e-05,
      "loss": 0.0013,
      "step": 3200
    },
    {
      "epoch": 12.05,
      "grad_norm": 0.4106840193271637,
      "learning_rate": 1.9308943089430896e-05,
      "loss": 0.1188,
      "step": 3210
    },
    {
      "epoch": 12.05020325203252,
      "eval_accuracy": 0.6062976725992568,
      "eval_loss": 2.8353450298309326,
      "eval_runtime": 10249.8185,
      "eval_samples_per_second": 0.499,
      "eval_steps_per_second": 0.031,
      "step": 3211
    },
    {
      "epoch": 13.001829268292683,
      "grad_norm": 0.014028260484337807,
      "learning_rate": 1.919602529358627e-05,
      "loss": 0.001,
      "step": 3220
    },
    {
      "epoch": 13.003861788617886,
      "grad_norm": 0.03284154087305069,
      "learning_rate": 1.9083107497741645e-05,
      "loss": 0.0009,
      "step": 3230
    },
    {
      "epoch": 13.00589430894309,
      "grad_norm": 0.5965408086776733,
      "learning_rate": 1.897018970189702e-05,
      "loss": 0.0008,
      "step": 3240
    },
    {
      "epoch": 13.007926829268293,
      "grad_norm": 0.006544598378241062,
      "learning_rate": 1.8857271906052394e-05,
      "loss": 0.0004,
      "step": 3250
    },
    {
      "epoch": 13.009959349593496,
      "grad_norm": 0.023756476119160652,
      "learning_rate": 1.8744354110207772e-05,
      "loss": 0.032,
      "step": 3260
    },
    {
      "epoch": 13.011991869918699,
      "grad_norm": 0.006157791242003441,
      "learning_rate": 1.8631436314363144e-05,
      "loss": 0.0031,
      "step": 3270
    },
    {
      "epoch": 13.014024390243902,
      "grad_norm": 0.04071442410349846,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.0262,
      "step": 3280
    },
    {
      "epoch": 13.016056910569105,
      "grad_norm": 0.007779354229569435,
      "learning_rate": 1.8405600722673893e-05,
      "loss": 0.0646,
      "step": 3290
    },
    {
      "epoch": 13.018089430894308,
      "grad_norm": 0.034714922308921814,
      "learning_rate": 1.8292682926829268e-05,
      "loss": 0.0176,
      "step": 3300
    },
    {
      "epoch": 13.020121951219512,
      "grad_norm": 0.014596055261790752,
      "learning_rate": 1.8179765130984646e-05,
      "loss": 0.0064,
      "step": 3310
    },
    {
      "epoch": 13.022154471544715,
      "grad_norm": 0.009341662749648094,
      "learning_rate": 1.806684733514002e-05,
      "loss": 0.0315,
      "step": 3320
    },
    {
      "epoch": 13.024186991869918,
      "grad_norm": 25.028392791748047,
      "learning_rate": 1.7953929539295395e-05,
      "loss": 0.0523,
      "step": 3330
    },
    {
      "epoch": 13.026219512195121,
      "grad_norm": 2.333561658859253,
      "learning_rate": 1.784101174345077e-05,
      "loss": 0.0423,
      "step": 3340
    },
    {
      "epoch": 13.028252032520324,
      "grad_norm": 0.0088898791000247,
      "learning_rate": 1.7728093947606144e-05,
      "loss": 0.0161,
      "step": 3350
    },
    {
      "epoch": 13.03028455284553,
      "grad_norm": 0.23925018310546875,
      "learning_rate": 1.7615176151761516e-05,
      "loss": 0.0012,
      "step": 3360
    },
    {
      "epoch": 13.032317073170733,
      "grad_norm": 0.09391099959611893,
      "learning_rate": 1.7502258355916894e-05,
      "loss": 0.0011,
      "step": 3370
    },
    {
      "epoch": 13.034349593495936,
      "grad_norm": 0.010089404881000519,
      "learning_rate": 1.7389340560072268e-05,
      "loss": 0.0025,
      "step": 3380
    },
    {
      "epoch": 13.036382113821139,
      "grad_norm": 0.014005468226969242,
      "learning_rate": 1.7276422764227643e-05,
      "loss": 0.0007,
      "step": 3390
    },
    {
      "epoch": 13.038414634146342,
      "grad_norm": 0.02113616280257702,
      "learning_rate": 1.7163504968383018e-05,
      "loss": 0.0745,
      "step": 3400
    },
    {
      "epoch": 13.040447154471545,
      "grad_norm": 0.01482979767024517,
      "learning_rate": 1.7050587172538392e-05,
      "loss": 0.0107,
      "step": 3410
    },
    {
      "epoch": 13.042479674796748,
      "grad_norm": 0.013184974901378155,
      "learning_rate": 1.6937669376693767e-05,
      "loss": 0.0396,
      "step": 3420
    },
    {
      "epoch": 13.044512195121952,
      "grad_norm": 2.344749689102173,
      "learning_rate": 1.6824751580849145e-05,
      "loss": 0.1209,
      "step": 3430
    },
    {
      "epoch": 13.046544715447155,
      "grad_norm": 0.31985217332839966,
      "learning_rate": 1.671183378500452e-05,
      "loss": 0.0075,
      "step": 3440
    },
    {
      "epoch": 13.048577235772358,
      "grad_norm": 0.05049590393900871,
      "learning_rate": 1.659891598915989e-05,
      "loss": 0.0203,
      "step": 3450
    },
    {
      "epoch": 13.05020325203252,
      "eval_accuracy": 0.6544103266184236,
      "eval_loss": 2.128178834915161,
      "eval_runtime": 3814.0587,
      "eval_samples_per_second": 1.341,
      "eval_steps_per_second": 0.084,
      "step": 3458
    },
    {
      "epoch": 14.00040650406504,
      "grad_norm": 0.027489762753248215,
      "learning_rate": 1.6485998193315266e-05,
      "loss": 0.0011,
      "step": 3460
    },
    {
      "epoch": 14.002439024390243,
      "grad_norm": 18.3277587890625,
      "learning_rate": 1.637308039747064e-05,
      "loss": 0.0312,
      "step": 3470
    },
    {
      "epoch": 14.004471544715447,
      "grad_norm": 0.012971523217856884,
      "learning_rate": 1.6260162601626018e-05,
      "loss": 0.0394,
      "step": 3480
    },
    {
      "epoch": 14.00650406504065,
      "grad_norm": 0.03960908204317093,
      "learning_rate": 1.6147244805781393e-05,
      "loss": 0.0541,
      "step": 3490
    },
    {
      "epoch": 14.008536585365853,
      "grad_norm": 0.046985089778900146,
      "learning_rate": 1.6034327009936768e-05,
      "loss": 0.0013,
      "step": 3500
    },
    {
      "epoch": 14.010569105691056,
      "grad_norm": 0.00806176383048296,
      "learning_rate": 1.5921409214092142e-05,
      "loss": 0.001,
      "step": 3510
    },
    {
      "epoch": 14.01260162601626,
      "grad_norm": 0.7922822833061218,
      "learning_rate": 1.5808491418247517e-05,
      "loss": 0.0011,
      "step": 3520
    },
    {
      "epoch": 14.014634146341463,
      "grad_norm": 0.05365770310163498,
      "learning_rate": 1.569557362240289e-05,
      "loss": 0.0008,
      "step": 3530
    },
    {
      "epoch": 14.016666666666667,
      "grad_norm": 0.02195759490132332,
      "learning_rate": 1.5582655826558266e-05,
      "loss": 0.0398,
      "step": 3540
    },
    {
      "epoch": 14.01869918699187,
      "grad_norm": 0.013551034033298492,
      "learning_rate": 1.546973803071364e-05,
      "loss": 0.0125,
      "step": 3550
    },
    {
      "epoch": 14.020731707317074,
      "grad_norm": 0.010373839177191257,
      "learning_rate": 1.5356820234869016e-05,
      "loss": 0.0016,
      "step": 3560
    },
    {
      "epoch": 14.022764227642277,
      "grad_norm": 0.007704328745603561,
      "learning_rate": 1.524390243902439e-05,
      "loss": 0.0046,
      "step": 3570
    },
    {
      "epoch": 14.02479674796748,
      "grad_norm": 0.021973930299282074,
      "learning_rate": 1.5130984643179767e-05,
      "loss": 0.0332,
      "step": 3580
    },
    {
      "epoch": 14.026829268292683,
      "grad_norm": 0.020018063485622406,
      "learning_rate": 1.5018066847335141e-05,
      "loss": 0.0639,
      "step": 3590
    },
    {
      "epoch": 14.028861788617887,
      "grad_norm": 0.022794604301452637,
      "learning_rate": 1.4905149051490516e-05,
      "loss": 0.0011,
      "step": 3600
    },
    {
      "epoch": 14.03089430894309,
      "grad_norm": 0.006530186627060175,
      "learning_rate": 1.479223125564589e-05,
      "loss": 0.0305,
      "step": 3610
    },
    {
      "epoch": 14.032926829268293,
      "grad_norm": 0.02438746951520443,
      "learning_rate": 1.4679313459801267e-05,
      "loss": 0.024,
      "step": 3620
    },
    {
      "epoch": 14.034959349593496,
      "grad_norm": 0.014021486043930054,
      "learning_rate": 1.4566395663956638e-05,
      "loss": 0.0006,
      "step": 3630
    },
    {
      "epoch": 14.0369918699187,
      "grad_norm": 6.659039497375488,
      "learning_rate": 1.4453477868112015e-05,
      "loss": 0.0022,
      "step": 3640
    },
    {
      "epoch": 14.039024390243902,
      "grad_norm": 0.010285748168826103,
      "learning_rate": 1.434056007226739e-05,
      "loss": 0.0206,
      "step": 3650
    },
    {
      "epoch": 14.041056910569106,
      "grad_norm": 0.01677355356514454,
      "learning_rate": 1.4227642276422764e-05,
      "loss": 0.057,
      "step": 3660
    },
    {
      "epoch": 14.043089430894309,
      "grad_norm": 0.22299115359783173,
      "learning_rate": 1.411472448057814e-05,
      "loss": 0.0016,
      "step": 3670
    },
    {
      "epoch": 14.045121951219512,
      "grad_norm": 0.04189331457018852,
      "learning_rate": 1.4001806684733515e-05,
      "loss": 0.0011,
      "step": 3680
    },
    {
      "epoch": 14.047154471544715,
      "grad_norm": 0.0389564223587513,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.037,
      "step": 3690
    },
    {
      "epoch": 14.049186991869918,
      "grad_norm": 0.05326010286808014,
      "learning_rate": 1.3775971093044266e-05,
      "loss": 0.053,
      "step": 3700
    },
    {
      "epoch": 14.05020325203252,
      "eval_accuracy": 0.6430666927439859,
      "eval_loss": 2.546066999435425,
      "eval_runtime": 3626.643,
      "eval_samples_per_second": 1.41,
      "eval_steps_per_second": 0.088,
      "step": 3705
    },
    {
      "epoch": 15.001016260162602,
      "grad_norm": 0.05411051586270332,
      "learning_rate": 1.366305329719964e-05,
      "loss": 0.0062,
      "step": 3710
    },
    {
      "epoch": 15.003048780487806,
      "grad_norm": 2.4724347591400146,
      "learning_rate": 1.3550135501355014e-05,
      "loss": 0.0236,
      "step": 3720
    },
    {
      "epoch": 15.005081300813009,
      "grad_norm": 0.10434508323669434,
      "learning_rate": 1.3437217705510388e-05,
      "loss": 0.0373,
      "step": 3730
    },
    {
      "epoch": 15.007113821138212,
      "grad_norm": 0.1086881011724472,
      "learning_rate": 1.3324299909665763e-05,
      "loss": 0.0024,
      "step": 3740
    },
    {
      "epoch": 15.009146341463415,
      "grad_norm": 0.010930895805358887,
      "learning_rate": 1.321138211382114e-05,
      "loss": 0.002,
      "step": 3750
    },
    {
      "epoch": 15.011178861788618,
      "grad_norm": 9.614416122436523,
      "learning_rate": 1.3098464317976514e-05,
      "loss": 0.0462,
      "step": 3760
    },
    {
      "epoch": 15.013211382113822,
      "grad_norm": 0.2737988233566284,
      "learning_rate": 1.2985546522131889e-05,
      "loss": 0.0259,
      "step": 3770
    },
    {
      "epoch": 15.015243902439025,
      "grad_norm": 0.005054129753261805,
      "learning_rate": 1.2872628726287265e-05,
      "loss": 0.0028,
      "step": 3780
    },
    {
      "epoch": 15.017276422764228,
      "grad_norm": 27.949443817138672,
      "learning_rate": 1.275971093044264e-05,
      "loss": 0.0219,
      "step": 3790
    },
    {
      "epoch": 15.019308943089431,
      "grad_norm": 0.00426591606810689,
      "learning_rate": 1.2646793134598014e-05,
      "loss": 0.0053,
      "step": 3800
    },
    {
      "epoch": 15.021341463414634,
      "grad_norm": 0.003677231492474675,
      "learning_rate": 1.2533875338753387e-05,
      "loss": 0.0003,
      "step": 3810
    },
    {
      "epoch": 15.023373983739837,
      "grad_norm": 0.003065867815166712,
      "learning_rate": 1.2420957542908764e-05,
      "loss": 0.0259,
      "step": 3820
    },
    {
      "epoch": 15.02540650406504,
      "grad_norm": 19.211761474609375,
      "learning_rate": 1.2308039747064137e-05,
      "loss": 0.0178,
      "step": 3830
    },
    {
      "epoch": 15.027439024390244,
      "grad_norm": 0.005159091204404831,
      "learning_rate": 1.2195121951219513e-05,
      "loss": 0.0005,
      "step": 3840
    },
    {
      "epoch": 15.029471544715447,
      "grad_norm": 0.011434073559939861,
      "learning_rate": 1.2082204155374888e-05,
      "loss": 0.0005,
      "step": 3850
    },
    {
      "epoch": 15.03150406504065,
      "grad_norm": 0.007803434506058693,
      "learning_rate": 1.1969286359530262e-05,
      "loss": 0.0078,
      "step": 3860
    },
    {
      "epoch": 15.033536585365853,
      "grad_norm": 0.0044273491948843,
      "learning_rate": 1.1856368563685639e-05,
      "loss": 0.0439,
      "step": 3870
    },
    {
      "epoch": 15.035569105691057,
      "grad_norm": 0.006281462963670492,
      "learning_rate": 1.1743450767841012e-05,
      "loss": 0.0169,
      "step": 3880
    },
    {
      "epoch": 15.03760162601626,
      "grad_norm": 0.008350726217031479,
      "learning_rate": 1.1630532971996388e-05,
      "loss": 0.0005,
      "step": 3890
    },
    {
      "epoch": 15.039634146341463,
      "grad_norm": 0.005144812166690826,
      "learning_rate": 1.1517615176151763e-05,
      "loss": 0.0008,
      "step": 3900
    },
    {
      "epoch": 15.041666666666666,
      "grad_norm": 2.5425307750701904,
      "learning_rate": 1.1404697380307137e-05,
      "loss": 0.0549,
      "step": 3910
    },
    {
      "epoch": 15.04369918699187,
      "grad_norm": 0.08249751478433609,
      "learning_rate": 1.1291779584462512e-05,
      "loss": 0.0028,
      "step": 3920
    },
    {
      "epoch": 15.045731707317072,
      "grad_norm": 0.09613699465990067,
      "learning_rate": 1.1178861788617887e-05,
      "loss": 0.0232,
      "step": 3930
    },
    {
      "epoch": 15.047764227642276,
      "grad_norm": 1.8868670463562012,
      "learning_rate": 1.1065943992773261e-05,
      "loss": 0.0029,
      "step": 3940
    },
    {
      "epoch": 15.049796747967479,
      "grad_norm": 0.00344731449149549,
      "learning_rate": 1.0953026196928638e-05,
      "loss": 0.0006,
      "step": 3950
    },
    {
      "epoch": 15.05020325203252,
      "eval_accuracy": 0.6256600821435556,
      "eval_loss": 2.9019858837127686,
      "eval_runtime": 3521.9337,
      "eval_samples_per_second": 1.452,
      "eval_steps_per_second": 0.091,
      "step": 3952
    },
    {
      "epoch": 16.00162601626016,
      "grad_norm": 0.004141269251704216,
      "learning_rate": 1.0840108401084012e-05,
      "loss": 0.0003,
      "step": 3960
    },
    {
      "epoch": 16.003658536585366,
      "grad_norm": 0.0039109280332922935,
      "learning_rate": 1.0727190605239385e-05,
      "loss": 0.0003,
      "step": 3970
    },
    {
      "epoch": 16.005691056910567,
      "grad_norm": 0.8462672829627991,
      "learning_rate": 1.0614272809394762e-05,
      "loss": 0.0005,
      "step": 3980
    },
    {
      "epoch": 16.007723577235772,
      "grad_norm": 0.004943956155329943,
      "learning_rate": 1.0501355013550136e-05,
      "loss": 0.0004,
      "step": 3990
    },
    {
      "epoch": 16.009756097560974,
      "grad_norm": 0.0036170475650578737,
      "learning_rate": 1.038843721770551e-05,
      "loss": 0.0003,
      "step": 4000
    },
    {
      "epoch": 16.01178861788618,
      "grad_norm": 6.780259609222412,
      "learning_rate": 1.0275519421860885e-05,
      "loss": 0.0276,
      "step": 4010
    },
    {
      "epoch": 16.013821138211384,
      "grad_norm": 0.002631395123898983,
      "learning_rate": 1.016260162601626e-05,
      "loss": 0.0004,
      "step": 4020
    },
    {
      "epoch": 16.015853658536585,
      "grad_norm": 7.172311305999756,
      "learning_rate": 1.0049683830171635e-05,
      "loss": 0.048,
      "step": 4030
    },
    {
      "epoch": 16.01788617886179,
      "grad_norm": 0.0053691561333835125,
      "learning_rate": 9.936766034327011e-06,
      "loss": 0.0248,
      "step": 4040
    },
    {
      "epoch": 16.01991869918699,
      "grad_norm": 0.009503933601081371,
      "learning_rate": 9.823848238482384e-06,
      "loss": 0.001,
      "step": 4050
    },
    {
      "epoch": 16.021951219512196,
      "grad_norm": 0.003945863340049982,
      "learning_rate": 9.71093044263776e-06,
      "loss": 0.0109,
      "step": 4060
    },
    {
      "epoch": 16.023983739837398,
      "grad_norm": 0.0034280980471521616,
      "learning_rate": 9.598012646793135e-06,
      "loss": 0.0055,
      "step": 4070
    },
    {
      "epoch": 16.026016260162603,
      "grad_norm": 0.008122052997350693,
      "learning_rate": 9.48509485094851e-06,
      "loss": 0.0329,
      "step": 4080
    },
    {
      "epoch": 16.028048780487804,
      "grad_norm": 0.01129022054374218,
      "learning_rate": 9.372177055103886e-06,
      "loss": 0.0461,
      "step": 4090
    },
    {
      "epoch": 16.03008130081301,
      "grad_norm": 13.332000732421875,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.0404,
      "step": 4100
    },
    {
      "epoch": 16.03211382113821,
      "grad_norm": 0.035617753863334656,
      "learning_rate": 9.146341463414634e-06,
      "loss": 0.0943,
      "step": 4110
    },
    {
      "epoch": 16.034146341463416,
      "grad_norm": 0.2601952850818634,
      "learning_rate": 9.03342366757001e-06,
      "loss": 0.0297,
      "step": 4120
    },
    {
      "epoch": 16.036178861788617,
      "grad_norm": 0.006036252249032259,
      "learning_rate": 8.920505871725385e-06,
      "loss": 0.0006,
      "step": 4130
    },
    {
      "epoch": 16.038211382113822,
      "grad_norm": 0.005205197259783745,
      "learning_rate": 8.807588075880758e-06,
      "loss": 0.0589,
      "step": 4140
    },
    {
      "epoch": 16.040243902439023,
      "grad_norm": 21.53049087524414,
      "learning_rate": 8.694670280036134e-06,
      "loss": 0.0438,
      "step": 4150
    },
    {
      "epoch": 16.04227642276423,
      "grad_norm": 0.014143107458949089,
      "learning_rate": 8.581752484191509e-06,
      "loss": 0.035,
      "step": 4160
    },
    {
      "epoch": 16.04430894308943,
      "grad_norm": 0.0035730754025280476,
      "learning_rate": 8.468834688346883e-06,
      "loss": 0.0009,
      "step": 4170
    },
    {
      "epoch": 16.046341463414635,
      "grad_norm": 0.01344846747815609,
      "learning_rate": 8.35591689250226e-06,
      "loss": 0.0007,
      "step": 4180
    },
    {
      "epoch": 16.048373983739836,
      "grad_norm": 8.768914222717285,
      "learning_rate": 8.242999096657633e-06,
      "loss": 0.0365,
      "step": 4190
    },
    {
      "epoch": 16.05020325203252,
      "eval_accuracy": 0.5914336006258557,
      "eval_loss": 3.0215682983398438,
      "eval_runtime": 3567.0131,
      "eval_samples_per_second": 1.433,
      "eval_steps_per_second": 0.09,
      "step": 4199
    },
    {
      "epoch": 17.00020325203252,
      "grad_norm": 0.003917406778782606,
      "learning_rate": 8.130081300813009e-06,
      "loss": 0.0003,
      "step": 4200
    },
    {
      "epoch": 17.002235772357725,
      "grad_norm": 0.00643058679997921,
      "learning_rate": 8.017163504968384e-06,
      "loss": 0.0346,
      "step": 4210
    },
    {
      "epoch": 17.004268292682926,
      "grad_norm": 0.006049501709640026,
      "learning_rate": 7.904245709123758e-06,
      "loss": 0.0005,
      "step": 4220
    },
    {
      "epoch": 17.00630081300813,
      "grad_norm": 0.032857026904821396,
      "learning_rate": 7.791327913279133e-06,
      "loss": 0.0039,
      "step": 4230
    },
    {
      "epoch": 17.008333333333333,
      "grad_norm": 0.012625771574676037,
      "learning_rate": 7.678410117434508e-06,
      "loss": 0.0013,
      "step": 4240
    },
    {
      "epoch": 17.010365853658538,
      "grad_norm": 0.008268596604466438,
      "learning_rate": 7.565492321589883e-06,
      "loss": 0.0318,
      "step": 4250
    },
    {
      "epoch": 17.01239837398374,
      "grad_norm": 0.008699605241417885,
      "learning_rate": 7.452574525745258e-06,
      "loss": 0.0017,
      "step": 4260
    },
    {
      "epoch": 17.014430894308944,
      "grad_norm": 0.0037073201965540648,
      "learning_rate": 7.3396567299006335e-06,
      "loss": 0.0004,
      "step": 4270
    },
    {
      "epoch": 17.016463414634146,
      "grad_norm": 0.012861008755862713,
      "learning_rate": 7.226738934056007e-06,
      "loss": 0.0006,
      "step": 4280
    },
    {
      "epoch": 17.01849593495935,
      "grad_norm": 0.009572568349540234,
      "learning_rate": 7.113821138211382e-06,
      "loss": 0.0004,
      "step": 4290
    },
    {
      "epoch": 17.020528455284552,
      "grad_norm": 0.05249549075961113,
      "learning_rate": 7.0009033423667574e-06,
      "loss": 0.0003,
      "step": 4300
    },
    {
      "epoch": 17.022560975609757,
      "grad_norm": 0.00690675200894475,
      "learning_rate": 6.887985546522133e-06,
      "loss": 0.0222,
      "step": 4310
    },
    {
      "epoch": 17.02459349593496,
      "grad_norm": 0.14723120629787445,
      "learning_rate": 6.775067750677507e-06,
      "loss": 0.0447,
      "step": 4320
    },
    {
      "epoch": 17.026626016260163,
      "grad_norm": 0.006879771128296852,
      "learning_rate": 6.6621499548328814e-06,
      "loss": 0.0004,
      "step": 4330
    },
    {
      "epoch": 17.028658536585365,
      "grad_norm": 0.010147513821721077,
      "learning_rate": 6.549232158988257e-06,
      "loss": 0.0006,
      "step": 4340
    },
    {
      "epoch": 17.03069105691057,
      "grad_norm": 0.008599472232162952,
      "learning_rate": 6.4363143631436324e-06,
      "loss": 0.0004,
      "step": 4350
    },
    {
      "epoch": 17.03272357723577,
      "grad_norm": 0.01941720023751259,
      "learning_rate": 6.323396567299007e-06,
      "loss": 0.0137,
      "step": 4360
    },
    {
      "epoch": 17.034756097560976,
      "grad_norm": 0.09482240676879883,
      "learning_rate": 6.210478771454382e-06,
      "loss": 0.0097,
      "step": 4370
    },
    {
      "epoch": 17.036788617886177,
      "grad_norm": 0.008161032572388649,
      "learning_rate": 6.0975609756097564e-06,
      "loss": 0.0015,
      "step": 4380
    },
    {
      "epoch": 17.038821138211382,
      "grad_norm": 0.007093061227351427,
      "learning_rate": 5.984643179765131e-06,
      "loss": 0.0158,
      "step": 4390
    },
    {
      "epoch": 17.040853658536584,
      "grad_norm": 0.005730860400944948,
      "learning_rate": 5.871725383920506e-06,
      "loss": 0.0004,
      "step": 4400
    },
    {
      "epoch": 17.04288617886179,
      "grad_norm": 0.03012336790561676,
      "learning_rate": 5.758807588075881e-06,
      "loss": 0.0028,
      "step": 4410
    },
    {
      "epoch": 17.044918699186994,
      "grad_norm": 0.009378884918987751,
      "learning_rate": 5.645889792231256e-06,
      "loss": 0.0004,
      "step": 4420
    },
    {
      "epoch": 17.046951219512195,
      "grad_norm": 0.07404131442308426,
      "learning_rate": 5.532971996386631e-06,
      "loss": 0.0791,
      "step": 4430
    },
    {
      "epoch": 17.0489837398374,
      "grad_norm": 0.007493541110306978,
      "learning_rate": 5.420054200542006e-06,
      "loss": 0.0005,
      "step": 4440
    },
    {
      "epoch": 17.05020325203252,
      "eval_accuracy": 0.6035595540778408,
      "eval_loss": 2.8912858963012695,
      "eval_runtime": 3507.2573,
      "eval_samples_per_second": 1.458,
      "eval_steps_per_second": 0.091,
      "step": 4446
    },
    {
      "epoch": 18.00081300813008,
      "grad_norm": 0.032766859978437424,
      "learning_rate": 5.307136404697381e-06,
      "loss": 0.0023,
      "step": 4450
    },
    {
      "epoch": 18.002845528455286,
      "grad_norm": 0.005975415464490652,
      "learning_rate": 5.194218608852755e-06,
      "loss": 0.0007,
      "step": 4460
    },
    {
      "epoch": 18.004878048780487,
      "grad_norm": 5.088438510894775,
      "learning_rate": 5.08130081300813e-06,
      "loss": 0.0187,
      "step": 4470
    },
    {
      "epoch": 18.006910569105692,
      "grad_norm": 0.016140466555953026,
      "learning_rate": 4.968383017163506e-06,
      "loss": 0.0043,
      "step": 4480
    },
    {
      "epoch": 18.008943089430893,
      "grad_norm": 0.004997374024242163,
      "learning_rate": 4.85546522131888e-06,
      "loss": 0.0468,
      "step": 4490
    },
    {
      "epoch": 18.0109756097561,
      "grad_norm": 0.017591752111911774,
      "learning_rate": 4.742547425474255e-06,
      "loss": 0.0018,
      "step": 4500
    },
    {
      "epoch": 18.0130081300813,
      "grad_norm": 0.015785448253154755,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.0057,
      "step": 4510
    },
    {
      "epoch": 18.015040650406505,
      "grad_norm": 0.013328205794095993,
      "learning_rate": 4.516711833785005e-06,
      "loss": 0.0004,
      "step": 4520
    },
    {
      "epoch": 18.017073170731706,
      "grad_norm": 0.004184463527053595,
      "learning_rate": 4.403794037940379e-06,
      "loss": 0.0042,
      "step": 4530
    },
    {
      "epoch": 18.01910569105691,
      "grad_norm": 17.984004974365234,
      "learning_rate": 4.290876242095754e-06,
      "loss": 0.0274,
      "step": 4540
    },
    {
      "epoch": 18.021138211382112,
      "grad_norm": 0.031190020963549614,
      "learning_rate": 4.17795844625113e-06,
      "loss": 0.001,
      "step": 4550
    },
    {
      "epoch": 18.023170731707317,
      "grad_norm": 0.008906004950404167,
      "learning_rate": 4.0650406504065046e-06,
      "loss": 0.0023,
      "step": 4560
    },
    {
      "epoch": 18.02520325203252,
      "grad_norm": 0.0037233394104987383,
      "learning_rate": 3.952122854561879e-06,
      "loss": 0.0416,
      "step": 4570
    },
    {
      "epoch": 18.027235772357724,
      "grad_norm": 0.002985101193189621,
      "learning_rate": 3.839205058717254e-06,
      "loss": 0.0003,
      "step": 4580
    },
    {
      "epoch": 18.029268292682925,
      "grad_norm": 0.018991755321621895,
      "learning_rate": 3.726287262872629e-06,
      "loss": 0.0005,
      "step": 4590
    },
    {
      "epoch": 18.03130081300813,
      "grad_norm": 0.002975381212309003,
      "learning_rate": 3.6133694670280036e-06,
      "loss": 0.0009,
      "step": 4600
    },
    {
      "epoch": 18.033333333333335,
      "grad_norm": 0.0034014638513326645,
      "learning_rate": 3.5004516711833787e-06,
      "loss": 0.0004,
      "step": 4610
    },
    {
      "epoch": 18.035365853658536,
      "grad_norm": 0.23700159788131714,
      "learning_rate": 3.3875338753387534e-06,
      "loss": 0.0003,
      "step": 4620
    },
    {
      "epoch": 18.03739837398374,
      "grad_norm": 31.48465919494629,
      "learning_rate": 3.2746160794941285e-06,
      "loss": 0.0193,
      "step": 4630
    },
    {
      "epoch": 18.039430894308943,
      "grad_norm": 3.6522305011749268,
      "learning_rate": 3.1616982836495036e-06,
      "loss": 0.0449,
      "step": 4640
    },
    {
      "epoch": 18.041463414634148,
      "grad_norm": 0.007250097580254078,
      "learning_rate": 3.0487804878048782e-06,
      "loss": 0.0132,
      "step": 4650
    },
    {
      "epoch": 18.04349593495935,
      "grad_norm": 0.008960702456533909,
      "learning_rate": 2.935862691960253e-06,
      "loss": 0.0037,
      "step": 4660
    },
    {
      "epoch": 18.045528455284554,
      "grad_norm": 42.57283401489258,
      "learning_rate": 2.822944896115628e-06,
      "loss": 0.0076,
      "step": 4670
    },
    {
      "epoch": 18.047560975609755,
      "grad_norm": 0.003039554925635457,
      "learning_rate": 2.710027100271003e-06,
      "loss": 0.0003,
      "step": 4680
    },
    {
      "epoch": 18.04959349593496,
      "grad_norm": 0.011103064753115177,
      "learning_rate": 2.5971093044263777e-06,
      "loss": 0.0003,
      "step": 4690
    },
    {
      "epoch": 18.05020325203252,
      "eval_accuracy": 0.6098181106982202,
      "eval_loss": 2.9040656089782715,
      "eval_runtime": 3751.9089,
      "eval_samples_per_second": 1.363,
      "eval_steps_per_second": 0.085,
      "step": 4693
    },
    {
      "epoch": 19.00142276422764,
      "grad_norm": 0.019083699211478233,
      "learning_rate": 2.484191508581753e-06,
      "loss": 0.0003,
      "step": 4700
    },
    {
      "epoch": 19.003455284552846,
      "grad_norm": 10.416364669799805,
      "learning_rate": 2.3712737127371275e-06,
      "loss": 0.0531,
      "step": 4710
    },
    {
      "epoch": 19.005487804878047,
      "grad_norm": 0.003458574879914522,
      "learning_rate": 2.2583559168925025e-06,
      "loss": 0.0236,
      "step": 4720
    },
    {
      "epoch": 19.007520325203252,
      "grad_norm": 0.004417909309267998,
      "learning_rate": 2.145438121047877e-06,
      "loss": 0.0002,
      "step": 4730
    },
    {
      "epoch": 19.009552845528454,
      "grad_norm": 0.0028191388119012117,
      "learning_rate": 2.0325203252032523e-06,
      "loss": 0.001,
      "step": 4740
    },
    {
      "epoch": 19.01158536585366,
      "grad_norm": 0.00929015688598156,
      "learning_rate": 1.919602529358627e-06,
      "loss": 0.0334,
      "step": 4750
    },
    {
      "epoch": 19.01361788617886,
      "grad_norm": 0.0023010026197880507,
      "learning_rate": 1.8066847335140018e-06,
      "loss": 0.0418,
      "step": 4760
    },
    {
      "epoch": 19.015650406504065,
      "grad_norm": 0.0030450611375272274,
      "learning_rate": 1.6937669376693767e-06,
      "loss": 0.0003,
      "step": 4770
    },
    {
      "epoch": 19.01768292682927,
      "grad_norm": 0.4362029731273651,
      "learning_rate": 1.5808491418247518e-06,
      "loss": 0.0007,
      "step": 4780
    },
    {
      "epoch": 19.01971544715447,
      "grad_norm": 0.005422713700681925,
      "learning_rate": 1.4679313459801264e-06,
      "loss": 0.0003,
      "step": 4790
    },
    {
      "epoch": 19.021747967479676,
      "grad_norm": 0.0024971056263893843,
      "learning_rate": 1.3550135501355015e-06,
      "loss": 0.0003,
      "step": 4800
    },
    {
      "epoch": 19.023780487804878,
      "grad_norm": 0.0028703399002552032,
      "learning_rate": 1.2420957542908764e-06,
      "loss": 0.0003,
      "step": 4810
    },
    {
      "epoch": 19.025813008130083,
      "grad_norm": 0.21473000943660736,
      "learning_rate": 1.1291779584462513e-06,
      "loss": 0.0004,
      "step": 4820
    },
    {
      "epoch": 19.027845528455284,
      "grad_norm": 0.003547381144016981,
      "learning_rate": 1.0162601626016261e-06,
      "loss": 0.0002,
      "step": 4830
    },
    {
      "epoch": 19.02987804878049,
      "grad_norm": 0.004586045630276203,
      "learning_rate": 9.033423667570009e-07,
      "loss": 0.0003,
      "step": 4840
    },
    {
      "epoch": 19.03191056910569,
      "grad_norm": 0.0027067726477980614,
      "learning_rate": 7.904245709123759e-07,
      "loss": 0.0004,
      "step": 4850
    },
    {
      "epoch": 19.033943089430895,
      "grad_norm": 0.00425006914883852,
      "learning_rate": 6.775067750677508e-07,
      "loss": 0.0003,
      "step": 4860
    },
    {
      "epoch": 19.035975609756097,
      "grad_norm": 0.018539980053901672,
      "learning_rate": 5.645889792231256e-07,
      "loss": 0.0285,
      "step": 4870
    },
    {
      "epoch": 19.0380081300813,
      "grad_norm": 0.0027310731820762157,
      "learning_rate": 4.5167118337850045e-07,
      "loss": 0.0003,
      "step": 4880
    },
    {
      "epoch": 19.040040650406503,
      "grad_norm": 0.0035171646159142256,
      "learning_rate": 3.387533875338754e-07,
      "loss": 0.0004,
      "step": 4890
    },
    {
      "epoch": 19.042073170731708,
      "grad_norm": 0.002373168943449855,
      "learning_rate": 2.2583559168925023e-07,
      "loss": 0.0002,
      "step": 4900
    },
    {
      "epoch": 19.04410569105691,
      "grad_norm": 0.004842995200306177,
      "learning_rate": 1.1291779584462511e-07,
      "loss": 0.0003,
      "step": 4910
    },
    {
      "epoch": 19.046138211382114,
      "grad_norm": 0.0033168108202517033,
      "learning_rate": 0.0,
      "loss": 0.0003,
      "step": 4920
    },
    {
      "epoch": 19.046138211382114,
      "eval_accuracy": 0.6027772345002934,
      "eval_loss": 2.982931613922119,
      "eval_runtime": 3702.8397,
      "eval_samples_per_second": 1.381,
      "eval_steps_per_second": 0.086,
      "step": 4920
    },
    {
      "epoch": 19.046138211382114,
      "step": 4920,
      "total_flos": 9.778329856703368e+19,
      "train_loss": 0.10708648971929328,
      "train_runtime": 143360.3413,
      "train_samples_per_second": 0.549,
      "train_steps_per_second": 0.034
    },
    {
      "epoch": 19.046138211382114,
      "eval_accuracy": 0.5411836485661989,
      "eval_loss": 1.1800525188446045,
      "eval_runtime": 901.1748,
      "eval_samples_per_second": 1.819,
      "eval_steps_per_second": 0.114,
      "step": 4920
    },
    {
      "epoch": 19.046138211382114,
      "eval_accuracy": 0.5411836485661989,
      "eval_loss": 1.180052399635315,
      "eval_runtime": 907.9807,
      "eval_samples_per_second": 1.805,
      "eval_steps_per_second": 0.113,
      "step": 4920
    }
  ],
  "logging_steps": 10,
  "max_steps": 4920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.778329856703368e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}