{
  "best_metric": 0.9566261172294617,
  "best_model_checkpoint": "./kd_results/apple/mobilevit-small_alpha0.7_temp3.0_t3/checkpoint-2788",
  "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": 8.068937301635742,
      "learning_rate": 3.8109756097560976e-06,
      "loss": 0.7665,
      "step": 25
    },
    {
      "epoch": 0.3048780487804878,
      "grad_norm": 1.7976951599121094,
      "learning_rate": 7.621951219512195e-06,
      "loss": 0.5871,
      "step": 50
    },
    {
      "epoch": 0.4573170731707317,
      "grad_norm": 2.7205824851989746,
      "learning_rate": 1.1432926829268294e-05,
      "loss": 0.5684,
      "step": 75
    },
    {
      "epoch": 0.6097560975609756,
      "grad_norm": 2.772486686706543,
      "learning_rate": 1.524390243902439e-05,
      "loss": 0.5651,
      "step": 100
    },
    {
      "epoch": 0.7621951219512195,
      "grad_norm": 3.6146066188812256,
      "learning_rate": 1.9054878048780488e-05,
      "loss": 0.5625,
      "step": 125
    },
    {
      "epoch": 0.9146341463414634,
      "grad_norm": 2.03482723236084,
      "learning_rate": 2.286585365853659e-05,
      "loss": 0.5602,
      "step": 150
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2635574837310195,
      "eval_loss": 1.4016146659851074,
      "eval_runtime": 40.367,
      "eval_samples_per_second": 22.84,
      "eval_steps_per_second": 0.718,
      "step": 164
    },
    {
      "epoch": 1.0670731707317074,
      "grad_norm": 1.2862821817398071,
      "learning_rate": 2.6676829268292686e-05,
      "loss": 0.5641,
      "step": 175
    },
    {
      "epoch": 1.2195121951219512,
      "grad_norm": 1.9371393918991089,
      "learning_rate": 3.048780487804878e-05,
      "loss": 0.5582,
      "step": 200
    },
    {
      "epoch": 1.3719512195121952,
      "grad_norm": 3.7208340167999268,
      "learning_rate": 3.429878048780488e-05,
      "loss": 0.5542,
      "step": 225
    },
    {
      "epoch": 1.524390243902439,
      "grad_norm": 3.09954833984375,
      "learning_rate": 3.8109756097560976e-05,
      "loss": 0.5628,
      "step": 250
    },
    {
      "epoch": 1.6768292682926829,
      "grad_norm": 3.218473196029663,
      "learning_rate": 4.1920731707317077e-05,
      "loss": 0.558,
      "step": 275
    },
    {
      "epoch": 1.8292682926829267,
      "grad_norm": 3.0445876121520996,
      "learning_rate": 4.573170731707318e-05,
      "loss": 0.5355,
      "step": 300
    },
    {
      "epoch": 1.9817073170731707,
      "grad_norm": 1.5993388891220093,
      "learning_rate": 4.954268292682927e-05,
      "loss": 0.5293,
      "step": 325
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.31778741865509763,
      "eval_loss": 1.339619517326355,
      "eval_runtime": 40.45,
      "eval_samples_per_second": 22.794,
      "eval_steps_per_second": 0.717,
      "step": 328
    },
    {
      "epoch": 2.1341463414634148,
      "grad_norm": 1.1602020263671875,
      "learning_rate": 4.962737127371274e-05,
      "loss": 0.526,
      "step": 350
    },
    {
      "epoch": 2.2865853658536586,
      "grad_norm": 3.2051305770874023,
      "learning_rate": 4.92039295392954e-05,
      "loss": 0.5216,
      "step": 375
    },
    {
      "epoch": 2.4390243902439024,
      "grad_norm": 4.029102325439453,
      "learning_rate": 4.878048780487805e-05,
      "loss": 0.495,
      "step": 400
    },
    {
      "epoch": 2.591463414634146,
      "grad_norm": 4.257214546203613,
      "learning_rate": 4.8357046070460704e-05,
      "loss": 0.5013,
      "step": 425
    },
    {
      "epoch": 2.7439024390243905,
      "grad_norm": 3.542896032333374,
      "learning_rate": 4.793360433604336e-05,
      "loss": 0.498,
      "step": 450
    },
    {
      "epoch": 2.8963414634146343,
      "grad_norm": 3.9297356605529785,
      "learning_rate": 4.7510162601626016e-05,
      "loss": 0.4722,
      "step": 475
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.47396963123644253,
      "eval_loss": 1.1894437074661255,
      "eval_runtime": 40.4246,
      "eval_samples_per_second": 22.808,
      "eval_steps_per_second": 0.717,
      "step": 492
    },
    {
      "epoch": 3.048780487804878,
      "grad_norm": 3.230293035507202,
      "learning_rate": 4.7086720867208675e-05,
      "loss": 0.4738,
      "step": 500
    },
    {
      "epoch": 3.201219512195122,
      "grad_norm": 3.6131033897399902,
      "learning_rate": 4.6663279132791334e-05,
      "loss": 0.4726,
      "step": 525
    },
    {
      "epoch": 3.3536585365853657,
      "grad_norm": 3.685091018676758,
      "learning_rate": 4.6239837398373986e-05,
      "loss": 0.4472,
      "step": 550
    },
    {
      "epoch": 3.5060975609756095,
      "grad_norm": 2.7749123573303223,
      "learning_rate": 4.581639566395664e-05,
      "loss": 0.4436,
      "step": 575
    },
    {
      "epoch": 3.658536585365854,
      "grad_norm": 3.262700080871582,
      "learning_rate": 4.53929539295393e-05,
      "loss": 0.4446,
      "step": 600
    },
    {
      "epoch": 3.8109756097560976,
      "grad_norm": 3.2966811656951904,
      "learning_rate": 4.496951219512196e-05,
      "loss": 0.4332,
      "step": 625
    },
    {
      "epoch": 3.9634146341463414,
      "grad_norm": 6.128195285797119,
      "learning_rate": 4.454607046070461e-05,
      "loss": 0.4285,
      "step": 650
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5336225596529284,
      "eval_loss": 1.1300371885299683,
      "eval_runtime": 41.4736,
      "eval_samples_per_second": 22.231,
      "eval_steps_per_second": 0.699,
      "step": 656
    },
    {
      "epoch": 4.115853658536586,
      "grad_norm": 1.3390623331069946,
      "learning_rate": 4.412262872628726e-05,
      "loss": 0.429,
      "step": 675
    },
    {
      "epoch": 4.2682926829268295,
      "grad_norm": 2.997009038925171,
      "learning_rate": 4.369918699186992e-05,
      "loss": 0.426,
      "step": 700
    },
    {
      "epoch": 4.420731707317073,
      "grad_norm": 3.32063364982605,
      "learning_rate": 4.3275745257452574e-05,
      "loss": 0.4108,
      "step": 725
    },
    {
      "epoch": 4.573170731707317,
      "grad_norm": 6.243458271026611,
      "learning_rate": 4.285230352303523e-05,
      "loss": 0.4249,
      "step": 750
    },
    {
      "epoch": 4.725609756097561,
      "grad_norm": 3.801074981689453,
      "learning_rate": 4.242886178861789e-05,
      "loss": 0.4069,
      "step": 775
    },
    {
      "epoch": 4.878048780487805,
      "grad_norm": 2.7360901832580566,
      "learning_rate": 4.2005420054200545e-05,
      "loss": 0.4026,
      "step": 800
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5824295010845987,
      "eval_loss": 1.0758073329925537,
      "eval_runtime": 41.8594,
      "eval_samples_per_second": 22.026,
      "eval_steps_per_second": 0.693,
      "step": 820
    },
    {
      "epoch": 5.030487804878049,
      "grad_norm": 4.799595832824707,
      "learning_rate": 4.15819783197832e-05,
      "loss": 0.4173,
      "step": 825
    },
    {
      "epoch": 5.182926829268292,
      "grad_norm": 2.9078187942504883,
      "learning_rate": 4.1158536585365856e-05,
      "loss": 0.3998,
      "step": 850
    },
    {
      "epoch": 5.335365853658536,
      "grad_norm": 3.774778366088867,
      "learning_rate": 4.0735094850948516e-05,
      "loss": 0.3861,
      "step": 875
    },
    {
      "epoch": 5.487804878048781,
      "grad_norm": 4.949164390563965,
      "learning_rate": 4.031165311653117e-05,
      "loss": 0.3931,
      "step": 900
    },
    {
      "epoch": 5.640243902439025,
      "grad_norm": 3.3501133918762207,
      "learning_rate": 3.988821138211382e-05,
      "loss": 0.392,
      "step": 925
    },
    {
      "epoch": 5.7926829268292686,
      "grad_norm": 1.3912969827651978,
      "learning_rate": 3.946476964769648e-05,
      "loss": 0.3871,
      "step": 950
    },
    {
      "epoch": 5.945121951219512,
      "grad_norm": 1.8055741786956787,
      "learning_rate": 3.904132791327913e-05,
      "loss": 0.3918,
      "step": 975
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5900216919739696,
      "eval_loss": 1.0469610691070557,
      "eval_runtime": 40.9778,
      "eval_samples_per_second": 22.5,
      "eval_steps_per_second": 0.708,
      "step": 984
    },
    {
      "epoch": 6.097560975609756,
      "grad_norm": 5.702024459838867,
      "learning_rate": 3.861788617886179e-05,
      "loss": 0.3849,
      "step": 1000
    },
    {
      "epoch": 6.25,
      "grad_norm": 2.4023022651672363,
      "learning_rate": 3.8194444444444444e-05,
      "loss": 0.3821,
      "step": 1025
    },
    {
      "epoch": 6.402439024390244,
      "grad_norm": 1.873747706413269,
      "learning_rate": 3.77710027100271e-05,
      "loss": 0.384,
      "step": 1050
    },
    {
      "epoch": 6.554878048780488,
      "grad_norm": 3.79156231880188,
      "learning_rate": 3.7347560975609755e-05,
      "loss": 0.3828,
      "step": 1075
    },
    {
      "epoch": 6.7073170731707314,
      "grad_norm": 4.098422527313232,
      "learning_rate": 3.6924119241192415e-05,
      "loss": 0.3773,
      "step": 1100
    },
    {
      "epoch": 6.859756097560975,
      "grad_norm": 1.7072761058807373,
      "learning_rate": 3.6500677506775074e-05,
      "loss": 0.3733,
      "step": 1125
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6095444685466378,
      "eval_loss": 1.0197956562042236,
      "eval_runtime": 40.4455,
      "eval_samples_per_second": 22.796,
      "eval_steps_per_second": 0.717,
      "step": 1148
    },
    {
      "epoch": 7.012195121951219,
      "grad_norm": 1.8962244987487793,
      "learning_rate": 3.6077235772357726e-05,
      "loss": 0.3679,
      "step": 1150
    },
    {
      "epoch": 7.164634146341464,
      "grad_norm": 2.354288101196289,
      "learning_rate": 3.565379403794038e-05,
      "loss": 0.3651,
      "step": 1175
    },
    {
      "epoch": 7.317073170731708,
      "grad_norm": 2.6500818729400635,
      "learning_rate": 3.523035230352303e-05,
      "loss": 0.3563,
      "step": 1200
    },
    {
      "epoch": 7.469512195121951,
      "grad_norm": 4.6288933753967285,
      "learning_rate": 3.48069105691057e-05,
      "loss": 0.3521,
      "step": 1225
    },
    {
      "epoch": 7.621951219512195,
      "grad_norm": 2.5447182655334473,
      "learning_rate": 3.438346883468835e-05,
      "loss": 0.3664,
      "step": 1250
    },
    {
      "epoch": 7.774390243902439,
      "grad_norm": 3.541546583175659,
      "learning_rate": 3.3960027100271e-05,
      "loss": 0.3613,
      "step": 1275
    },
    {
      "epoch": 7.926829268292683,
      "grad_norm": 3.149768590927124,
      "learning_rate": 3.353658536585366e-05,
      "loss": 0.3622,
      "step": 1300
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6355748373101953,
      "eval_loss": 0.9974873661994934,
      "eval_runtime": 41.2546,
      "eval_samples_per_second": 22.349,
      "eval_steps_per_second": 0.703,
      "step": 1312
    },
    {
      "epoch": 8.079268292682928,
      "grad_norm": 3.627375602722168,
      "learning_rate": 3.3113143631436314e-05,
      "loss": 0.3529,
      "step": 1325
    },
    {
      "epoch": 8.231707317073171,
      "grad_norm": 4.361692428588867,
      "learning_rate": 3.268970189701897e-05,
      "loss": 0.344,
      "step": 1350
    },
    {
      "epoch": 8.384146341463415,
      "grad_norm": 3.2309417724609375,
      "learning_rate": 3.226626016260163e-05,
      "loss": 0.3594,
      "step": 1375
    },
    {
      "epoch": 8.536585365853659,
      "grad_norm": 1.337557315826416,
      "learning_rate": 3.1842818428184285e-05,
      "loss": 0.3381,
      "step": 1400
    },
    {
      "epoch": 8.689024390243903,
      "grad_norm": 2.6758828163146973,
      "learning_rate": 3.141937669376694e-05,
      "loss": 0.3564,
      "step": 1425
    },
    {
      "epoch": 8.841463414634147,
      "grad_norm": 3.8586676120758057,
      "learning_rate": 3.0995934959349596e-05,
      "loss": 0.3403,
      "step": 1450
    },
    {
      "epoch": 8.99390243902439,
      "grad_norm": 3.7615859508514404,
      "learning_rate": 3.0572493224932255e-05,
      "loss": 0.3504,
      "step": 1475
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6052060737527115,
      "eval_loss": 0.997992753982544,
      "eval_runtime": 41.0691,
      "eval_samples_per_second": 22.45,
      "eval_steps_per_second": 0.706,
      "step": 1476
    },
    {
      "epoch": 9.146341463414634,
      "grad_norm": 1.4145981073379517,
      "learning_rate": 3.0149051490514908e-05,
      "loss": 0.3369,
      "step": 1500
    },
    {
      "epoch": 9.298780487804878,
      "grad_norm": 3.1288559436798096,
      "learning_rate": 2.972560975609756e-05,
      "loss": 0.3488,
      "step": 1525
    },
    {
      "epoch": 9.451219512195122,
      "grad_norm": 3.6095070838928223,
      "learning_rate": 2.930216802168022e-05,
      "loss": 0.3373,
      "step": 1550
    },
    {
      "epoch": 9.603658536585366,
      "grad_norm": 3.40952205657959,
      "learning_rate": 2.8878726287262875e-05,
      "loss": 0.3333,
      "step": 1575
    },
    {
      "epoch": 9.75609756097561,
      "grad_norm": 3.9528234004974365,
      "learning_rate": 2.8455284552845528e-05,
      "loss": 0.3442,
      "step": 1600
    },
    {
      "epoch": 9.908536585365853,
      "grad_norm": 2.3063442707061768,
      "learning_rate": 2.8031842818428184e-05,
      "loss": 0.3345,
      "step": 1625
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6171366594360087,
      "eval_loss": 1.0102119445800781,
      "eval_runtime": 41.0313,
      "eval_samples_per_second": 22.471,
      "eval_steps_per_second": 0.707,
      "step": 1640
    },
    {
      "epoch": 10.060975609756097,
      "grad_norm": 2.8857736587524414,
      "learning_rate": 2.7608401084010843e-05,
      "loss": 0.3346,
      "step": 1650
    },
    {
      "epoch": 10.213414634146341,
      "grad_norm": 1.7870771884918213,
      "learning_rate": 2.71849593495935e-05,
      "loss": 0.3275,
      "step": 1675
    },
    {
      "epoch": 10.365853658536585,
      "grad_norm": 4.46469783782959,
      "learning_rate": 2.676151761517615e-05,
      "loss": 0.3398,
      "step": 1700
    },
    {
      "epoch": 10.518292682926829,
      "grad_norm": 2.4383885860443115,
      "learning_rate": 2.633807588075881e-05,
      "loss": 0.329,
      "step": 1725
    },
    {
      "epoch": 10.670731707317072,
      "grad_norm": 2.1877756118774414,
      "learning_rate": 2.5914634146341466e-05,
      "loss": 0.3239,
      "step": 1750
    },
    {
      "epoch": 10.823170731707316,
      "grad_norm": 3.2625539302825928,
      "learning_rate": 2.549119241192412e-05,
      "loss": 0.3329,
      "step": 1775
    },
    {
      "epoch": 10.975609756097562,
      "grad_norm": 4.354867458343506,
      "learning_rate": 2.5067750677506774e-05,
      "loss": 0.3362,
      "step": 1800
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6008676789587852,
      "eval_loss": 1.012461543083191,
      "eval_runtime": 41.0714,
      "eval_samples_per_second": 22.449,
      "eval_steps_per_second": 0.706,
      "step": 1804
    },
    {
      "epoch": 11.128048780487806,
      "grad_norm": 5.102630138397217,
      "learning_rate": 2.464430894308943e-05,
      "loss": 0.3373,
      "step": 1825
    },
    {
      "epoch": 11.28048780487805,
      "grad_norm": 3.905867338180542,
      "learning_rate": 2.422086720867209e-05,
      "loss": 0.3228,
      "step": 1850
    },
    {
      "epoch": 11.432926829268293,
      "grad_norm": 3.5207550525665283,
      "learning_rate": 2.3797425474254745e-05,
      "loss": 0.3215,
      "step": 1875
    },
    {
      "epoch": 11.585365853658537,
      "grad_norm": 4.22362756729126,
      "learning_rate": 2.3373983739837398e-05,
      "loss": 0.321,
      "step": 1900
    },
    {
      "epoch": 11.737804878048781,
      "grad_norm": 2.190922498703003,
      "learning_rate": 2.2950542005420057e-05,
      "loss": 0.3286,
      "step": 1925
    },
    {
      "epoch": 11.890243902439025,
      "grad_norm": 4.008134365081787,
      "learning_rate": 2.252710027100271e-05,
      "loss": 0.3281,
      "step": 1950
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6301518438177874,
      "eval_loss": 0.9796279668807983,
      "eval_runtime": 41.1914,
      "eval_samples_per_second": 22.383,
      "eval_steps_per_second": 0.704,
      "step": 1968
    },
    {
      "epoch": 12.042682926829269,
      "grad_norm": 1.8417494297027588,
      "learning_rate": 2.210365853658537e-05,
      "loss": 0.3275,
      "step": 1975
    },
    {
      "epoch": 12.195121951219512,
      "grad_norm": 2.1919643878936768,
      "learning_rate": 2.1680216802168024e-05,
      "loss": 0.312,
      "step": 2000
    },
    {
      "epoch": 12.347560975609756,
      "grad_norm": 4.401681900024414,
      "learning_rate": 2.1256775067750677e-05,
      "loss": 0.3213,
      "step": 2025
    },
    {
      "epoch": 12.5,
      "grad_norm": 2.55501389503479,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.3194,
      "step": 2050
    },
    {
      "epoch": 12.652439024390244,
      "grad_norm": 1.9148985147476196,
      "learning_rate": 2.040989159891599e-05,
      "loss": 0.3181,
      "step": 2075
    },
    {
      "epoch": 12.804878048780488,
      "grad_norm": 2.983046054840088,
      "learning_rate": 1.9986449864498648e-05,
      "loss": 0.3096,
      "step": 2100
    },
    {
      "epoch": 12.957317073170731,
      "grad_norm": 2.8490724563598633,
      "learning_rate": 1.95630081300813e-05,
      "loss": 0.3146,
      "step": 2125
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6616052060737527,
      "eval_loss": 0.9591848850250244,
      "eval_runtime": 40.9905,
      "eval_samples_per_second": 22.493,
      "eval_steps_per_second": 0.707,
      "step": 2132
    },
    {
      "epoch": 13.109756097560975,
      "grad_norm": 2.6641592979431152,
      "learning_rate": 1.9139566395663956e-05,
      "loss": 0.3186,
      "step": 2150
    },
    {
      "epoch": 13.262195121951219,
      "grad_norm": 2.278855800628662,
      "learning_rate": 1.8716124661246615e-05,
      "loss": 0.3172,
      "step": 2175
    },
    {
      "epoch": 13.414634146341463,
      "grad_norm": 2.147651195526123,
      "learning_rate": 1.8292682926829268e-05,
      "loss": 0.3151,
      "step": 2200
    },
    {
      "epoch": 13.567073170731707,
      "grad_norm": 1.7433544397354126,
      "learning_rate": 1.7869241192411927e-05,
      "loss": 0.3065,
      "step": 2225
    },
    {
      "epoch": 13.71951219512195,
      "grad_norm": 3.9199421405792236,
      "learning_rate": 1.744579945799458e-05,
      "loss": 0.3051,
      "step": 2250
    },
    {
      "epoch": 13.871951219512194,
      "grad_norm": 1.637223243713379,
      "learning_rate": 1.702235772357724e-05,
      "loss": 0.3172,
      "step": 2275
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6475054229934925,
      "eval_loss": 0.9733015298843384,
      "eval_runtime": 40.3794,
      "eval_samples_per_second": 22.833,
      "eval_steps_per_second": 0.718,
      "step": 2296
    },
    {
      "epoch": 14.024390243902438,
      "grad_norm": 3.8937485218048096,
      "learning_rate": 1.659891598915989e-05,
      "loss": 0.3116,
      "step": 2300
    },
    {
      "epoch": 14.176829268292684,
      "grad_norm": 3.694873809814453,
      "learning_rate": 1.6175474254742547e-05,
      "loss": 0.3108,
      "step": 2325
    },
    {
      "epoch": 14.329268292682928,
      "grad_norm": 3.432950735092163,
      "learning_rate": 1.5752032520325206e-05,
      "loss": 0.3052,
      "step": 2350
    },
    {
      "epoch": 14.481707317073171,
      "grad_norm": 1.0661758184432983,
      "learning_rate": 1.532859078590786e-05,
      "loss": 0.3057,
      "step": 2375
    },
    {
      "epoch": 14.634146341463415,
      "grad_norm": 3.632337808609009,
      "learning_rate": 1.4905149051490516e-05,
      "loss": 0.312,
      "step": 2400
    },
    {
      "epoch": 14.786585365853659,
      "grad_norm": 1.7915321588516235,
      "learning_rate": 1.448170731707317e-05,
      "loss": 0.3089,
      "step": 2425
    },
    {
      "epoch": 14.939024390243903,
      "grad_norm": 2.4583942890167236,
      "learning_rate": 1.4058265582655828e-05,
      "loss": 0.3091,
      "step": 2450
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6540130151843818,
      "eval_loss": 0.9678862690925598,
      "eval_runtime": 40.1458,
      "eval_samples_per_second": 22.966,
      "eval_steps_per_second": 0.722,
      "step": 2460
    },
    {
      "epoch": 15.091463414634147,
      "grad_norm": 3.025009870529175,
      "learning_rate": 1.3634823848238485e-05,
      "loss": 0.298,
      "step": 2475
    },
    {
      "epoch": 15.24390243902439,
      "grad_norm": 3.4300453662872314,
      "learning_rate": 1.321138211382114e-05,
      "loss": 0.3074,
      "step": 2500
    },
    {
      "epoch": 15.396341463414634,
      "grad_norm": 1.541405200958252,
      "learning_rate": 1.2787940379403795e-05,
      "loss": 0.3062,
      "step": 2525
    },
    {
      "epoch": 15.548780487804878,
      "grad_norm": 2.387010097503662,
      "learning_rate": 1.2364498644986451e-05,
      "loss": 0.2945,
      "step": 2550
    },
    {
      "epoch": 15.701219512195122,
      "grad_norm": 2.736443519592285,
      "learning_rate": 1.1941056910569107e-05,
      "loss": 0.3052,
      "step": 2575
    },
    {
      "epoch": 15.853658536585366,
      "grad_norm": 1.431464672088623,
      "learning_rate": 1.1517615176151763e-05,
      "loss": 0.3071,
      "step": 2600
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6507592190889371,
      "eval_loss": 0.9691238403320312,
      "eval_runtime": 39.6989,
      "eval_samples_per_second": 23.225,
      "eval_steps_per_second": 0.73,
      "step": 2624
    },
    {
      "epoch": 16.00609756097561,
      "grad_norm": 2.9343862533569336,
      "learning_rate": 1.1094173441734418e-05,
      "loss": 0.306,
      "step": 2625
    },
    {
      "epoch": 16.158536585365855,
      "grad_norm": 2.295952081680298,
      "learning_rate": 1.0670731707317074e-05,
      "loss": 0.2991,
      "step": 2650
    },
    {
      "epoch": 16.3109756097561,
      "grad_norm": 4.23213529586792,
      "learning_rate": 1.0247289972899728e-05,
      "loss": 0.2971,
      "step": 2675
    },
    {
      "epoch": 16.463414634146343,
      "grad_norm": 2.456118583679199,
      "learning_rate": 9.823848238482384e-06,
      "loss": 0.2962,
      "step": 2700
    },
    {
      "epoch": 16.615853658536587,
      "grad_norm": 1.444655179977417,
      "learning_rate": 9.400406504065042e-06,
      "loss": 0.303,
      "step": 2725
    },
    {
      "epoch": 16.76829268292683,
      "grad_norm": 3.4977266788482666,
      "learning_rate": 8.976964769647697e-06,
      "loss": 0.2963,
      "step": 2750
    },
    {
      "epoch": 16.920731707317074,
      "grad_norm": 2.6125636100769043,
      "learning_rate": 8.553523035230353e-06,
      "loss": 0.3073,
      "step": 2775
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6616052060737527,
      "eval_loss": 0.9566261172294617,
      "eval_runtime": 40.7013,
      "eval_samples_per_second": 22.653,
      "eval_steps_per_second": 0.713,
      "step": 2788
    },
    {
      "epoch": 17.073170731707318,
      "grad_norm": 2.241692066192627,
      "learning_rate": 8.130081300813009e-06,
      "loss": 0.3013,
      "step": 2800
    },
    {
      "epoch": 17.225609756097562,
      "grad_norm": 1.112600326538086,
      "learning_rate": 7.706639566395663e-06,
      "loss": 0.2911,
      "step": 2825
    },
    {
      "epoch": 17.378048780487806,
      "grad_norm": 2.2839741706848145,
      "learning_rate": 7.283197831978319e-06,
      "loss": 0.3048,
      "step": 2850
    },
    {
      "epoch": 17.53048780487805,
      "grad_norm": 2.7641336917877197,
      "learning_rate": 6.859756097560977e-06,
      "loss": 0.3041,
      "step": 2875
    },
    {
      "epoch": 17.682926829268293,
      "grad_norm": 1.2625651359558105,
      "learning_rate": 6.4363143631436324e-06,
      "loss": 0.3002,
      "step": 2900
    },
    {
      "epoch": 17.835365853658537,
      "grad_norm": 2.472018241882324,
      "learning_rate": 6.0128726287262874e-06,
      "loss": 0.2953,
      "step": 2925
    },
    {
      "epoch": 17.98780487804878,
      "grad_norm": 1.4805189371109009,
      "learning_rate": 5.589430894308943e-06,
      "loss": 0.2953,
      "step": 2950
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.648590021691974,
      "eval_loss": 0.9578619003295898,
      "eval_runtime": 41.3628,
      "eval_samples_per_second": 22.291,
      "eval_steps_per_second": 0.701,
      "step": 2952
    },
    {
      "epoch": 18.140243902439025,
      "grad_norm": 1.534975290298462,
      "learning_rate": 5.165989159891599e-06,
      "loss": 0.2983,
      "step": 2975
    },
    {
      "epoch": 18.29268292682927,
      "grad_norm": 3.629396677017212,
      "learning_rate": 4.742547425474255e-06,
      "loss": 0.2982,
      "step": 3000
    },
    {
      "epoch": 18.445121951219512,
      "grad_norm": 1.30571448802948,
      "learning_rate": 4.319105691056911e-06,
      "loss": 0.2972,
      "step": 3025
    },
    {
      "epoch": 18.597560975609756,
      "grad_norm": 1.2913093566894531,
      "learning_rate": 3.8956639566395666e-06,
      "loss": 0.2961,
      "step": 3050
    },
    {
      "epoch": 18.75,
      "grad_norm": 1.5529708862304688,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.2923,
      "step": 3075
    },
    {
      "epoch": 18.902439024390244,
      "grad_norm": 2.1228458881378174,
      "learning_rate": 3.0487804878048782e-06,
      "loss": 0.3014,
      "step": 3100
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6605206073752712,
      "eval_loss": 0.9578254818916321,
      "eval_runtime": 40.314,
      "eval_samples_per_second": 22.87,
      "eval_steps_per_second": 0.719,
      "step": 3116
    },
    {
      "epoch": 19.054878048780488,
      "grad_norm": 2.608959436416626,
      "learning_rate": 2.625338753387534e-06,
      "loss": 0.2944,
      "step": 3125
    },
    {
      "epoch": 19.20731707317073,
      "grad_norm": 1.4581996202468872,
      "learning_rate": 2.2018970189701894e-06,
      "loss": 0.3009,
      "step": 3150
    },
    {
      "epoch": 19.359756097560975,
      "grad_norm": 1.045812964439392,
      "learning_rate": 1.7784552845528455e-06,
      "loss": 0.2979,
      "step": 3175
    },
    {
      "epoch": 19.51219512195122,
      "grad_norm": 1.0459766387939453,
      "learning_rate": 1.3550135501355015e-06,
      "loss": 0.2938,
      "step": 3200
    },
    {
      "epoch": 19.664634146341463,
      "grad_norm": 2.3936879634857178,
      "learning_rate": 9.315718157181572e-07,
      "loss": 0.2976,
      "step": 3225
    },
    {
      "epoch": 19.817073170731707,
      "grad_norm": 3.6012637615203857,
      "learning_rate": 5.081300813008131e-07,
      "loss": 0.2921,
      "step": 3250
    },
    {
      "epoch": 19.96951219512195,
      "grad_norm": 1.0907233953475952,
      "learning_rate": 8.468834688346885e-08,
      "loss": 0.2909,
      "step": 3275
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6550976138828634,
      "eval_loss": 0.959929883480072,
      "eval_runtime": 40.9305,
      "eval_samples_per_second": 22.526,
      "eval_steps_per_second": 0.709,
      "step": 3280
    },
    {
      "epoch": 20.0,
      "step": 3280,
      "total_flos": 0.0,
      "train_loss": 0.37087090215305,
      "train_runtime": 14479.2181,
      "train_samples_per_second": 7.216,
      "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
}