{
  "best_metric": 0.9597899913787842,
  "best_model_checkpoint": "./kd_results/apple/mobilevit-x-small_alpha0.7_temp3.0_t3/checkpoint-2132",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 3280,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1524390243902439,
      "grad_norm": 3.7345454692840576,
      "learning_rate": 3.8109756097560976e-06,
      "loss": 0.6279,
      "step": 25
    },
    {
      "epoch": 0.3048780487804878,
      "grad_norm": 1.898392915725708,
      "learning_rate": 7.621951219512195e-06,
      "loss": 0.5665,
      "step": 50
    },
    {
      "epoch": 0.4573170731707317,
      "grad_norm": 2.2714648246765137,
      "learning_rate": 1.1432926829268294e-05,
      "loss": 0.5694,
      "step": 75
    },
    {
      "epoch": 0.6097560975609756,
      "grad_norm": 1.6978486776351929,
      "learning_rate": 1.524390243902439e-05,
      "loss": 0.5615,
      "step": 100
    },
    {
      "epoch": 0.7621951219512195,
      "grad_norm": 1.8619637489318848,
      "learning_rate": 1.9054878048780488e-05,
      "loss": 0.5597,
      "step": 125
    },
    {
      "epoch": 0.9146341463414634,
      "grad_norm": 1.9784170389175415,
      "learning_rate": 2.286585365853659e-05,
      "loss": 0.5532,
      "step": 150
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2765726681127983,
      "eval_loss": 1.3949692249298096,
      "eval_runtime": 40.477,
      "eval_samples_per_second": 22.778,
      "eval_steps_per_second": 0.716,
      "step": 164
    },
    {
      "epoch": 1.0670731707317074,
      "grad_norm": 1.3847393989562988,
      "learning_rate": 2.6676829268292686e-05,
      "loss": 0.5627,
      "step": 175
    },
    {
      "epoch": 1.2195121951219512,
      "grad_norm": 1.3784176111221313,
      "learning_rate": 3.048780487804878e-05,
      "loss": 0.5582,
      "step": 200
    },
    {
      "epoch": 1.3719512195121952,
      "grad_norm": 2.6379387378692627,
      "learning_rate": 3.429878048780488e-05,
      "loss": 0.5505,
      "step": 225
    },
    {
      "epoch": 1.524390243902439,
      "grad_norm": 0.9585696458816528,
      "learning_rate": 3.8109756097560976e-05,
      "loss": 0.5509,
      "step": 250
    },
    {
      "epoch": 1.6768292682926829,
      "grad_norm": 1.4419804811477661,
      "learning_rate": 4.1920731707317077e-05,
      "loss": 0.5504,
      "step": 275
    },
    {
      "epoch": 1.8292682926829267,
      "grad_norm": 3.571232557296753,
      "learning_rate": 4.573170731707318e-05,
      "loss": 0.5359,
      "step": 300
    },
    {
      "epoch": 1.9817073170731707,
      "grad_norm": 0.8830344676971436,
      "learning_rate": 4.954268292682927e-05,
      "loss": 0.5286,
      "step": 325
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.30151843817787416,
      "eval_loss": 1.345648169517517,
      "eval_runtime": 40.9309,
      "eval_samples_per_second": 22.526,
      "eval_steps_per_second": 0.709,
      "step": 328
    },
    {
      "epoch": 2.1341463414634148,
      "grad_norm": 1.9142130613327026,
      "learning_rate": 4.962737127371274e-05,
      "loss": 0.5357,
      "step": 350
    },
    {
      "epoch": 2.2865853658536586,
      "grad_norm": 1.0479462146759033,
      "learning_rate": 4.92039295392954e-05,
      "loss": 0.5221,
      "step": 375
    },
    {
      "epoch": 2.4390243902439024,
      "grad_norm": 3.054068088531494,
      "learning_rate": 4.878048780487805e-05,
      "loss": 0.5064,
      "step": 400
    },
    {
      "epoch": 2.591463414634146,
      "grad_norm": 1.7171183824539185,
      "learning_rate": 4.8357046070460704e-05,
      "loss": 0.5064,
      "step": 425
    },
    {
      "epoch": 2.7439024390243905,
      "grad_norm": 1.1312766075134277,
      "learning_rate": 4.793360433604336e-05,
      "loss": 0.5036,
      "step": 450
    },
    {
      "epoch": 2.8963414634146343,
      "grad_norm": 1.6097897291183472,
      "learning_rate": 4.7510162601626016e-05,
      "loss": 0.4764,
      "step": 475
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4880694143167028,
      "eval_loss": 1.215592622756958,
      "eval_runtime": 40.8767,
      "eval_samples_per_second": 22.556,
      "eval_steps_per_second": 0.709,
      "step": 492
    },
    {
      "epoch": 3.048780487804878,
      "grad_norm": 2.613168954849243,
      "learning_rate": 4.7086720867208675e-05,
      "loss": 0.4773,
      "step": 500
    },
    {
      "epoch": 3.201219512195122,
      "grad_norm": 3.723191022872925,
      "learning_rate": 4.6663279132791334e-05,
      "loss": 0.4706,
      "step": 525
    },
    {
      "epoch": 3.3536585365853657,
      "grad_norm": 1.5962661504745483,
      "learning_rate": 4.6239837398373986e-05,
      "loss": 0.4653,
      "step": 550
    },
    {
      "epoch": 3.5060975609756095,
      "grad_norm": 1.0754482746124268,
      "learning_rate": 4.581639566395664e-05,
      "loss": 0.4571,
      "step": 575
    },
    {
      "epoch": 3.658536585365854,
      "grad_norm": 2.5520670413970947,
      "learning_rate": 4.53929539295393e-05,
      "loss": 0.4526,
      "step": 600
    },
    {
      "epoch": 3.8109756097560976,
      "grad_norm": 2.01554536819458,
      "learning_rate": 4.496951219512196e-05,
      "loss": 0.4418,
      "step": 625
    },
    {
      "epoch": 3.9634146341463414,
      "grad_norm": 4.0134148597717285,
      "learning_rate": 4.454607046070461e-05,
      "loss": 0.4397,
      "step": 650
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5498915401301518,
      "eval_loss": 1.120837688446045,
      "eval_runtime": 41.1962,
      "eval_samples_per_second": 22.381,
      "eval_steps_per_second": 0.704,
      "step": 656
    },
    {
      "epoch": 4.115853658536586,
      "grad_norm": 1.3298094272613525,
      "learning_rate": 4.412262872628726e-05,
      "loss": 0.4397,
      "step": 675
    },
    {
      "epoch": 4.2682926829268295,
      "grad_norm": 3.123973846435547,
      "learning_rate": 4.369918699186992e-05,
      "loss": 0.4399,
      "step": 700
    },
    {
      "epoch": 4.420731707317073,
      "grad_norm": 2.281238079071045,
      "learning_rate": 4.3275745257452574e-05,
      "loss": 0.4222,
      "step": 725
    },
    {
      "epoch": 4.573170731707317,
      "grad_norm": 4.78800630569458,
      "learning_rate": 4.285230352303523e-05,
      "loss": 0.4222,
      "step": 750
    },
    {
      "epoch": 4.725609756097561,
      "grad_norm": 2.5941879749298096,
      "learning_rate": 4.242886178861789e-05,
      "loss": 0.4172,
      "step": 775
    },
    {
      "epoch": 4.878048780487805,
      "grad_norm": 2.55926251411438,
      "learning_rate": 4.2005420054200545e-05,
      "loss": 0.4225,
      "step": 800
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5900216919739696,
      "eval_loss": 1.059511423110962,
      "eval_runtime": 40.9157,
      "eval_samples_per_second": 22.534,
      "eval_steps_per_second": 0.709,
      "step": 820
    },
    {
      "epoch": 5.030487804878049,
      "grad_norm": 6.088729381561279,
      "learning_rate": 4.15819783197832e-05,
      "loss": 0.4184,
      "step": 825
    },
    {
      "epoch": 5.182926829268292,
      "grad_norm": 2.796999931335449,
      "learning_rate": 4.1158536585365856e-05,
      "loss": 0.41,
      "step": 850
    },
    {
      "epoch": 5.335365853658536,
      "grad_norm": 1.5400503873825073,
      "learning_rate": 4.0735094850948516e-05,
      "loss": 0.3998,
      "step": 875
    },
    {
      "epoch": 5.487804878048781,
      "grad_norm": 1.9177120923995972,
      "learning_rate": 4.031165311653117e-05,
      "loss": 0.4178,
      "step": 900
    },
    {
      "epoch": 5.640243902439025,
      "grad_norm": 2.9486024379730225,
      "learning_rate": 3.988821138211382e-05,
      "loss": 0.3972,
      "step": 925
    },
    {
      "epoch": 5.7926829268292686,
      "grad_norm": 4.523825645446777,
      "learning_rate": 3.946476964769648e-05,
      "loss": 0.4069,
      "step": 950
    },
    {
      "epoch": 5.945121951219512,
      "grad_norm": 1.8625081777572632,
      "learning_rate": 3.904132791327913e-05,
      "loss": 0.3976,
      "step": 975
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6084598698481561,
      "eval_loss": 1.0301053524017334,
      "eval_runtime": 40.9069,
      "eval_samples_per_second": 22.539,
      "eval_steps_per_second": 0.709,
      "step": 984
    },
    {
      "epoch": 6.097560975609756,
      "grad_norm": 4.603743553161621,
      "learning_rate": 3.861788617886179e-05,
      "loss": 0.4078,
      "step": 1000
    },
    {
      "epoch": 6.25,
      "grad_norm": 1.958526611328125,
      "learning_rate": 3.8194444444444444e-05,
      "loss": 0.3866,
      "step": 1025
    },
    {
      "epoch": 6.402439024390244,
      "grad_norm": 1.1663492918014526,
      "learning_rate": 3.77710027100271e-05,
      "loss": 0.3953,
      "step": 1050
    },
    {
      "epoch": 6.554878048780488,
      "grad_norm": 2.688338041305542,
      "learning_rate": 3.7347560975609755e-05,
      "loss": 0.3879,
      "step": 1075
    },
    {
      "epoch": 6.7073170731707314,
      "grad_norm": 2.4757819175720215,
      "learning_rate": 3.6924119241192415e-05,
      "loss": 0.3906,
      "step": 1100
    },
    {
      "epoch": 6.859756097560975,
      "grad_norm": 1.4951151609420776,
      "learning_rate": 3.6500677506775074e-05,
      "loss": 0.3816,
      "step": 1125
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5986984815618221,
      "eval_loss": 1.035401463508606,
      "eval_runtime": 40.7855,
      "eval_samples_per_second": 22.606,
      "eval_steps_per_second": 0.711,
      "step": 1148
    },
    {
      "epoch": 7.012195121951219,
      "grad_norm": 3.0987966060638428,
      "learning_rate": 3.6077235772357726e-05,
      "loss": 0.3912,
      "step": 1150
    },
    {
      "epoch": 7.164634146341464,
      "grad_norm": 1.7649365663528442,
      "learning_rate": 3.565379403794038e-05,
      "loss": 0.3816,
      "step": 1175
    },
    {
      "epoch": 7.317073170731708,
      "grad_norm": 2.6136391162872314,
      "learning_rate": 3.523035230352303e-05,
      "loss": 0.3743,
      "step": 1200
    },
    {
      "epoch": 7.469512195121951,
      "grad_norm": 2.337141990661621,
      "learning_rate": 3.48069105691057e-05,
      "loss": 0.372,
      "step": 1225
    },
    {
      "epoch": 7.621951219512195,
      "grad_norm": 4.140254497528076,
      "learning_rate": 3.438346883468835e-05,
      "loss": 0.383,
      "step": 1250
    },
    {
      "epoch": 7.774390243902439,
      "grad_norm": 2.4736337661743164,
      "learning_rate": 3.3960027100271e-05,
      "loss": 0.388,
      "step": 1275
    },
    {
      "epoch": 7.926829268292683,
      "grad_norm": 1.4317537546157837,
      "learning_rate": 3.353658536585366e-05,
      "loss": 0.3825,
      "step": 1300
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6182212581344902,
      "eval_loss": 1.0171048641204834,
      "eval_runtime": 40.7984,
      "eval_samples_per_second": 22.599,
      "eval_steps_per_second": 0.711,
      "step": 1312
    },
    {
      "epoch": 8.079268292682928,
      "grad_norm": 2.8469619750976562,
      "learning_rate": 3.3113143631436314e-05,
      "loss": 0.3668,
      "step": 1325
    },
    {
      "epoch": 8.231707317073171,
      "grad_norm": 2.6326467990875244,
      "learning_rate": 3.268970189701897e-05,
      "loss": 0.3612,
      "step": 1350
    },
    {
      "epoch": 8.384146341463415,
      "grad_norm": 4.997561454772949,
      "learning_rate": 3.226626016260163e-05,
      "loss": 0.3771,
      "step": 1375
    },
    {
      "epoch": 8.536585365853659,
      "grad_norm": 1.6547101736068726,
      "learning_rate": 3.1842818428184285e-05,
      "loss": 0.3637,
      "step": 1400
    },
    {
      "epoch": 8.689024390243903,
      "grad_norm": 2.6040332317352295,
      "learning_rate": 3.141937669376694e-05,
      "loss": 0.3719,
      "step": 1425
    },
    {
      "epoch": 8.841463414634147,
      "grad_norm": 2.0173676013946533,
      "learning_rate": 3.0995934959349596e-05,
      "loss": 0.3598,
      "step": 1450
    },
    {
      "epoch": 8.99390243902439,
      "grad_norm": 2.6045796871185303,
      "learning_rate": 3.0572493224932255e-05,
      "loss": 0.3696,
      "step": 1475
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6268980477223427,
      "eval_loss": 0.9927615523338318,
      "eval_runtime": 40.444,
      "eval_samples_per_second": 22.797,
      "eval_steps_per_second": 0.717,
      "step": 1476
    },
    {
      "epoch": 9.146341463414634,
      "grad_norm": 2.540935516357422,
      "learning_rate": 3.0149051490514908e-05,
      "loss": 0.3505,
      "step": 1500
    },
    {
      "epoch": 9.298780487804878,
      "grad_norm": 1.7361254692077637,
      "learning_rate": 2.972560975609756e-05,
      "loss": 0.3669,
      "step": 1525
    },
    {
      "epoch": 9.451219512195122,
      "grad_norm": 2.2732503414154053,
      "learning_rate": 2.930216802168022e-05,
      "loss": 0.3598,
      "step": 1550
    },
    {
      "epoch": 9.603658536585366,
      "grad_norm": 1.3575024604797363,
      "learning_rate": 2.8878726287262875e-05,
      "loss": 0.3561,
      "step": 1575
    },
    {
      "epoch": 9.75609756097561,
      "grad_norm": 1.395545482635498,
      "learning_rate": 2.8455284552845528e-05,
      "loss": 0.371,
      "step": 1600
    },
    {
      "epoch": 9.908536585365853,
      "grad_norm": 2.107233762741089,
      "learning_rate": 2.8031842818428184e-05,
      "loss": 0.3513,
      "step": 1625
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6225596529284165,
      "eval_loss": 1.0023045539855957,
      "eval_runtime": 40.1352,
      "eval_samples_per_second": 22.972,
      "eval_steps_per_second": 0.723,
      "step": 1640
    },
    {
      "epoch": 10.060975609756097,
      "grad_norm": 1.7307239770889282,
      "learning_rate": 2.7608401084010843e-05,
      "loss": 0.3592,
      "step": 1650
    },
    {
      "epoch": 10.213414634146341,
      "grad_norm": 2.54693341255188,
      "learning_rate": 2.71849593495935e-05,
      "loss": 0.3523,
      "step": 1675
    },
    {
      "epoch": 10.365853658536585,
      "grad_norm": 2.0898828506469727,
      "learning_rate": 2.676151761517615e-05,
      "loss": 0.3498,
      "step": 1700
    },
    {
      "epoch": 10.518292682926829,
      "grad_norm": 1.3730179071426392,
      "learning_rate": 2.633807588075881e-05,
      "loss": 0.3481,
      "step": 1725
    },
    {
      "epoch": 10.670731707317072,
      "grad_norm": 2.239461660385132,
      "learning_rate": 2.5914634146341466e-05,
      "loss": 0.3463,
      "step": 1750
    },
    {
      "epoch": 10.823170731707316,
      "grad_norm": 2.5770819187164307,
      "learning_rate": 2.549119241192412e-05,
      "loss": 0.3552,
      "step": 1775
    },
    {
      "epoch": 10.975609756097562,
      "grad_norm": 2.877971649169922,
      "learning_rate": 2.5067750677506774e-05,
      "loss": 0.3489,
      "step": 1800
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6095444685466378,
      "eval_loss": 1.0006794929504395,
      "eval_runtime": 40.8889,
      "eval_samples_per_second": 22.549,
      "eval_steps_per_second": 0.709,
      "step": 1804
    },
    {
      "epoch": 11.128048780487806,
      "grad_norm": 3.003967761993408,
      "learning_rate": 2.464430894308943e-05,
      "loss": 0.3499,
      "step": 1825
    },
    {
      "epoch": 11.28048780487805,
      "grad_norm": 1.321793794631958,
      "learning_rate": 2.422086720867209e-05,
      "loss": 0.3447,
      "step": 1850
    },
    {
      "epoch": 11.432926829268293,
      "grad_norm": 3.9306182861328125,
      "learning_rate": 2.3797425474254745e-05,
      "loss": 0.3463,
      "step": 1875
    },
    {
      "epoch": 11.585365853658537,
      "grad_norm": 1.7785139083862305,
      "learning_rate": 2.3373983739837398e-05,
      "loss": 0.3399,
      "step": 1900
    },
    {
      "epoch": 11.737804878048781,
      "grad_norm": 2.1510891914367676,
      "learning_rate": 2.2950542005420057e-05,
      "loss": 0.3518,
      "step": 1925
    },
    {
      "epoch": 11.890243902439025,
      "grad_norm": 3.6793370246887207,
      "learning_rate": 2.252710027100271e-05,
      "loss": 0.3529,
      "step": 1950
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6377440347071583,
      "eval_loss": 0.9781045317649841,
      "eval_runtime": 41.6978,
      "eval_samples_per_second": 22.111,
      "eval_steps_per_second": 0.695,
      "step": 1968
    },
    {
      "epoch": 12.042682926829269,
      "grad_norm": 1.4457000494003296,
      "learning_rate": 2.210365853658537e-05,
      "loss": 0.3445,
      "step": 1975
    },
    {
      "epoch": 12.195121951219512,
      "grad_norm": 2.660372495651245,
      "learning_rate": 2.1680216802168024e-05,
      "loss": 0.3313,
      "step": 2000
    },
    {
      "epoch": 12.347560975609756,
      "grad_norm": 1.3028616905212402,
      "learning_rate": 2.1256775067750677e-05,
      "loss": 0.348,
      "step": 2025
    },
    {
      "epoch": 12.5,
      "grad_norm": 2.393099308013916,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.3442,
      "step": 2050
    },
    {
      "epoch": 12.652439024390244,
      "grad_norm": 2.0544044971466064,
      "learning_rate": 2.040989159891599e-05,
      "loss": 0.3444,
      "step": 2075
    },
    {
      "epoch": 12.804878048780488,
      "grad_norm": 1.6202657222747803,
      "learning_rate": 1.9986449864498648e-05,
      "loss": 0.3352,
      "step": 2100
    },
    {
      "epoch": 12.957317073170731,
      "grad_norm": 2.0553081035614014,
      "learning_rate": 1.95630081300813e-05,
      "loss": 0.3323,
      "step": 2125
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6453362255965293,
      "eval_loss": 0.9597899913787842,
      "eval_runtime": 40.6478,
      "eval_samples_per_second": 22.683,
      "eval_steps_per_second": 0.713,
      "step": 2132
    },
    {
      "epoch": 13.109756097560975,
      "grad_norm": 2.281384229660034,
      "learning_rate": 1.9139566395663956e-05,
      "loss": 0.3375,
      "step": 2150
    },
    {
      "epoch": 13.262195121951219,
      "grad_norm": 1.460560917854309,
      "learning_rate": 1.8716124661246615e-05,
      "loss": 0.3421,
      "step": 2175
    },
    {
      "epoch": 13.414634146341463,
      "grad_norm": 2.1000704765319824,
      "learning_rate": 1.8292682926829268e-05,
      "loss": 0.3382,
      "step": 2200
    },
    {
      "epoch": 13.567073170731707,
      "grad_norm": 2.0283122062683105,
      "learning_rate": 1.7869241192411927e-05,
      "loss": 0.3337,
      "step": 2225
    },
    {
      "epoch": 13.71951219512195,
      "grad_norm": 4.288078784942627,
      "learning_rate": 1.744579945799458e-05,
      "loss": 0.3241,
      "step": 2250
    },
    {
      "epoch": 13.871951219512194,
      "grad_norm": 1.8883591890335083,
      "learning_rate": 1.702235772357724e-05,
      "loss": 0.3355,
      "step": 2275
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6388286334056399,
      "eval_loss": 0.9914635419845581,
      "eval_runtime": 40.4312,
      "eval_samples_per_second": 22.804,
      "eval_steps_per_second": 0.717,
      "step": 2296
    },
    {
      "epoch": 14.024390243902438,
      "grad_norm": 2.0896174907684326,
      "learning_rate": 1.659891598915989e-05,
      "loss": 0.3388,
      "step": 2300
    },
    {
      "epoch": 14.176829268292684,
      "grad_norm": 1.939850926399231,
      "learning_rate": 1.6175474254742547e-05,
      "loss": 0.3339,
      "step": 2325
    },
    {
      "epoch": 14.329268292682928,
      "grad_norm": 1.495898723602295,
      "learning_rate": 1.5752032520325206e-05,
      "loss": 0.3284,
      "step": 2350
    },
    {
      "epoch": 14.481707317073171,
      "grad_norm": 1.344751238822937,
      "learning_rate": 1.532859078590786e-05,
      "loss": 0.3287,
      "step": 2375
    },
    {
      "epoch": 14.634146341463415,
      "grad_norm": 2.6487138271331787,
      "learning_rate": 1.4905149051490516e-05,
      "loss": 0.329,
      "step": 2400
    },
    {
      "epoch": 14.786585365853659,
      "grad_norm": 1.6167285442352295,
      "learning_rate": 1.448170731707317e-05,
      "loss": 0.3271,
      "step": 2425
    },
    {
      "epoch": 14.939024390243903,
      "grad_norm": 3.4504077434539795,
      "learning_rate": 1.4058265582655828e-05,
      "loss": 0.3328,
      "step": 2450
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6377440347071583,
      "eval_loss": 0.98011314868927,
      "eval_runtime": 41.1599,
      "eval_samples_per_second": 22.4,
      "eval_steps_per_second": 0.705,
      "step": 2460
    },
    {
      "epoch": 15.091463414634147,
      "grad_norm": 2.9397025108337402,
      "learning_rate": 1.3634823848238485e-05,
      "loss": 0.3158,
      "step": 2475
    },
    {
      "epoch": 15.24390243902439,
      "grad_norm": 2.264390707015991,
      "learning_rate": 1.321138211382114e-05,
      "loss": 0.3349,
      "step": 2500
    },
    {
      "epoch": 15.396341463414634,
      "grad_norm": 1.4004125595092773,
      "learning_rate": 1.2787940379403795e-05,
      "loss": 0.3313,
      "step": 2525
    },
    {
      "epoch": 15.548780487804878,
      "grad_norm": 1.2873611450195312,
      "learning_rate": 1.2364498644986451e-05,
      "loss": 0.3249,
      "step": 2550
    },
    {
      "epoch": 15.701219512195122,
      "grad_norm": 1.9618803262710571,
      "learning_rate": 1.1941056910569107e-05,
      "loss": 0.3268,
      "step": 2575
    },
    {
      "epoch": 15.853658536585366,
      "grad_norm": 1.3931300640106201,
      "learning_rate": 1.1517615176151763e-05,
      "loss": 0.3305,
      "step": 2600
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6388286334056399,
      "eval_loss": 0.9744916558265686,
      "eval_runtime": 40.1084,
      "eval_samples_per_second": 22.988,
      "eval_steps_per_second": 0.723,
      "step": 2624
    },
    {
      "epoch": 16.00609756097561,
      "grad_norm": 2.658966302871704,
      "learning_rate": 1.1094173441734418e-05,
      "loss": 0.3226,
      "step": 2625
    },
    {
      "epoch": 16.158536585365855,
      "grad_norm": 1.2235839366912842,
      "learning_rate": 1.0670731707317074e-05,
      "loss": 0.3204,
      "step": 2650
    },
    {
      "epoch": 16.3109756097561,
      "grad_norm": 2.9110801219940186,
      "learning_rate": 1.0247289972899728e-05,
      "loss": 0.3217,
      "step": 2675
    },
    {
      "epoch": 16.463414634146343,
      "grad_norm": 1.3170216083526611,
      "learning_rate": 9.823848238482384e-06,
      "loss": 0.3191,
      "step": 2700
    },
    {
      "epoch": 16.615853658536587,
      "grad_norm": 2.4817311763763428,
      "learning_rate": 9.400406504065042e-06,
      "loss": 0.3259,
      "step": 2725
    },
    {
      "epoch": 16.76829268292683,
      "grad_norm": 2.6858763694763184,
      "learning_rate": 8.976964769647697e-06,
      "loss": 0.3195,
      "step": 2750
    },
    {
      "epoch": 16.920731707317074,
      "grad_norm": 2.6210761070251465,
      "learning_rate": 8.553523035230353e-06,
      "loss": 0.3268,
      "step": 2775
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6496746203904555,
      "eval_loss": 0.9655945897102356,
      "eval_runtime": 39.7959,
      "eval_samples_per_second": 23.168,
      "eval_steps_per_second": 0.729,
      "step": 2788
    },
    {
      "epoch": 17.073170731707318,
      "grad_norm": 2.4724555015563965,
      "learning_rate": 8.130081300813009e-06,
      "loss": 0.3202,
      "step": 2800
    },
    {
      "epoch": 17.225609756097562,
      "grad_norm": 2.4132657051086426,
      "learning_rate": 7.706639566395663e-06,
      "loss": 0.3162,
      "step": 2825
    },
    {
      "epoch": 17.378048780487806,
      "grad_norm": 2.5661656856536865,
      "learning_rate": 7.283197831978319e-06,
      "loss": 0.325,
      "step": 2850
    },
    {
      "epoch": 17.53048780487805,
      "grad_norm": 1.742690086364746,
      "learning_rate": 6.859756097560977e-06,
      "loss": 0.323,
      "step": 2875
    },
    {
      "epoch": 17.682926829268293,
      "grad_norm": 2.458004951477051,
      "learning_rate": 6.4363143631436324e-06,
      "loss": 0.3219,
      "step": 2900
    },
    {
      "epoch": 17.835365853658537,
      "grad_norm": 2.420703649520874,
      "learning_rate": 6.0128726287262874e-06,
      "loss": 0.3204,
      "step": 2925
    },
    {
      "epoch": 17.98780487804878,
      "grad_norm": 2.132844924926758,
      "learning_rate": 5.589430894308943e-06,
      "loss": 0.3165,
      "step": 2950
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6529284164859002,
      "eval_loss": 0.9608322381973267,
      "eval_runtime": 40.2408,
      "eval_samples_per_second": 22.912,
      "eval_steps_per_second": 0.721,
      "step": 2952
    },
    {
      "epoch": 18.140243902439025,
      "grad_norm": 1.4758714437484741,
      "learning_rate": 5.165989159891599e-06,
      "loss": 0.325,
      "step": 2975
    },
    {
      "epoch": 18.29268292682927,
      "grad_norm": 1.8934224843978882,
      "learning_rate": 4.742547425474255e-06,
      "loss": 0.3196,
      "step": 3000
    },
    {
      "epoch": 18.445121951219512,
      "grad_norm": 1.881678581237793,
      "learning_rate": 4.319105691056911e-06,
      "loss": 0.3189,
      "step": 3025
    },
    {
      "epoch": 18.597560975609756,
      "grad_norm": 1.4729632139205933,
      "learning_rate": 3.8956639566395666e-06,
      "loss": 0.3165,
      "step": 3050
    },
    {
      "epoch": 18.75,
      "grad_norm": 1.7952040433883667,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.3202,
      "step": 3075
    },
    {
      "epoch": 18.902439024390244,
      "grad_norm": 1.5464168787002563,
      "learning_rate": 3.0487804878048782e-06,
      "loss": 0.3222,
      "step": 3100
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6420824295010846,
      "eval_loss": 0.9698135256767273,
      "eval_runtime": 40.337,
      "eval_samples_per_second": 22.857,
      "eval_steps_per_second": 0.719,
      "step": 3116
    },
    {
      "epoch": 19.054878048780488,
      "grad_norm": 2.1197171211242676,
      "learning_rate": 2.625338753387534e-06,
      "loss": 0.3119,
      "step": 3125
    },
    {
      "epoch": 19.20731707317073,
      "grad_norm": 2.011740207672119,
      "learning_rate": 2.2018970189701894e-06,
      "loss": 0.3228,
      "step": 3150
    },
    {
      "epoch": 19.359756097560975,
      "grad_norm": 1.4066367149353027,
      "learning_rate": 1.7784552845528455e-06,
      "loss": 0.3187,
      "step": 3175
    },
    {
      "epoch": 19.51219512195122,
      "grad_norm": 1.4676731824874878,
      "learning_rate": 1.3550135501355015e-06,
      "loss": 0.3217,
      "step": 3200
    },
    {
      "epoch": 19.664634146341463,
      "grad_norm": 1.0230920314788818,
      "learning_rate": 9.315718157181572e-07,
      "loss": 0.3182,
      "step": 3225
    },
    {
      "epoch": 19.817073170731707,
      "grad_norm": 1.3352421522140503,
      "learning_rate": 5.081300813008131e-07,
      "loss": 0.3134,
      "step": 3250
    },
    {
      "epoch": 19.96951219512195,
      "grad_norm": 1.4204283952713013,
      "learning_rate": 8.468834688346885e-08,
      "loss": 0.3161,
      "step": 3275
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6420824295010846,
      "eval_loss": 0.9688836336135864,
      "eval_runtime": 40.2191,
      "eval_samples_per_second": 22.924,
      "eval_steps_per_second": 0.721,
      "step": 3280
    },
    {
      "epoch": 20.0,
      "step": 3280,
      "total_flos": 0.0,
      "train_loss": 0.3857773442093919,
      "train_runtime": 14443.1896,
      "train_samples_per_second": 7.234,
      "train_steps_per_second": 0.227
    }
  ],
  "logging_steps": 25,
  "max_steps": 3280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}