{
  "best_metric": 1.8259689807891846,
  "best_model_checkpoint": "model_training/byt5_small/checkpoints-by_file-09-07-11-51/checkpoint-350",
  "epoch": 5.685279187817259,
  "eval_steps": 25,
  "global_step": 350,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08121827411167512,
      "grad_norm": 1332.5194091796875,
      "learning_rate": 5.102040816326531e-05,
      "loss": 57.087,
      "step": 5
    },
    {
      "epoch": 0.16243654822335024,
      "grad_norm": 73.31609344482422,
      "learning_rate": 0.00010204081632653062,
      "loss": 16.158,
      "step": 10
    },
    {
      "epoch": 0.2436548223350254,
      "grad_norm": 7.558203220367432,
      "learning_rate": 0.00015306122448979594,
      "loss": 4.2458,
      "step": 15
    },
    {
      "epoch": 0.3248730964467005,
      "grad_norm": 3.45859432220459,
      "learning_rate": 0.00020408163265306123,
      "loss": 3.5694,
      "step": 20
    },
    {
      "epoch": 0.40609137055837563,
      "grad_norm": 1.5817078351974487,
      "learning_rate": 0.00025510204081632655,
      "loss": 2.5571,
      "step": 25
    },
    {
      "epoch": 0.40609137055837563,
      "eval_loss": 1.86122727394104,
      "eval_runtime": 89.2369,
      "eval_samples_per_second": 11.206,
      "eval_steps_per_second": 1.401,
      "step": 25
    },
    {
      "epoch": 0.4873096446700508,
      "grad_norm": 0.7394866943359375,
      "learning_rate": 0.0003061224489795919,
      "loss": 2.1392,
      "step": 30
    },
    {
      "epoch": 0.5685279187817259,
      "grad_norm": 8.203688621520996,
      "learning_rate": 0.00035714285714285714,
      "loss": 2.0015,
      "step": 35
    },
    {
      "epoch": 0.649746192893401,
      "grad_norm": 8.245688438415527,
      "learning_rate": 0.00040816326530612246,
      "loss": 1.9816,
      "step": 40
    },
    {
      "epoch": 0.7309644670050761,
      "grad_norm": 1.8892487287521362,
      "learning_rate": 0.0004591836734693878,
      "loss": 2.3033,
      "step": 45
    },
    {
      "epoch": 0.8121827411167513,
      "grad_norm": 6.5841898918151855,
      "learning_rate": 0.0004999935985425297,
      "loss": 3.3388,
      "step": 50
    },
    {
      "epoch": 0.8121827411167513,
      "eval_loss": 2.1760518550872803,
      "eval_runtime": 88.9886,
      "eval_samples_per_second": 11.237,
      "eval_steps_per_second": 1.405,
      "step": 50
    },
    {
      "epoch": 0.8934010152284264,
      "grad_norm": 7.12192440032959,
      "learning_rate": 0.0004997695819512612,
      "loss": 3.6302,
      "step": 55
    },
    {
      "epoch": 0.9746192893401016,
      "grad_norm": 40.77882385253906,
      "learning_rate": 0.0004992258202402822,
      "loss": 2.7657,
      "step": 60
    },
    {
      "epoch": 1.0558375634517767,
      "grad_norm": 2.32853102684021,
      "learning_rate": 0.0004983630095117843,
      "loss": 2.8533,
      "step": 65
    },
    {
      "epoch": 1.1370558375634519,
      "grad_norm": 20.28560447692871,
      "learning_rate": 0.0004971822543018662,
      "loss": 2.6894,
      "step": 70
    },
    {
      "epoch": 1.218274111675127,
      "grad_norm": 13.825627326965332,
      "learning_rate": 0.0004956850661665511,
      "loss": 2.6103,
      "step": 75
    },
    {
      "epoch": 1.218274111675127,
      "eval_loss": 2.4702670574188232,
      "eval_runtime": 89.4229,
      "eval_samples_per_second": 11.183,
      "eval_steps_per_second": 1.398,
      "step": 75
    },
    {
      "epoch": 1.299492385786802,
      "grad_norm": 10.716843605041504,
      "learning_rate": 0.0004938733617467517,
      "loss": 2.565,
      "step": 80
    },
    {
      "epoch": 1.380710659898477,
      "grad_norm": 1.5047727823257446,
      "learning_rate": 0.0004917494603146632,
      "loss": 2.4888,
      "step": 85
    },
    {
      "epoch": 1.4619289340101522,
      "grad_norm": 5.189760684967041,
      "learning_rate": 0.0004893160808047222,
      "loss": 2.4684,
      "step": 90
    },
    {
      "epoch": 1.5431472081218274,
      "grad_norm": 1.0872949361801147,
      "learning_rate": 0.00048657633833293557,
      "loss": 2.4044,
      "step": 95
    },
    {
      "epoch": 1.6243654822335025,
      "grad_norm": 0.9995412230491638,
      "learning_rate": 0.0004835337402090316,
      "loss": 2.3281,
      "step": 100
    },
    {
      "epoch": 1.6243654822335025,
      "eval_loss": 2.164686918258667,
      "eval_runtime": 89.4619,
      "eval_samples_per_second": 11.178,
      "eval_steps_per_second": 1.397,
      "step": 100
    },
    {
      "epoch": 1.7055837563451777,
      "grad_norm": 1.2759788036346436,
      "learning_rate": 0.0004801921814465414,
      "loss": 2.2785,
      "step": 105
    },
    {
      "epoch": 1.7868020304568528,
      "grad_norm": 0.9360132217407227,
      "learning_rate": 0.00047655593977655674,
      "loss": 2.214,
      "step": 110
    },
    {
      "epoch": 1.868020304568528,
      "grad_norm": 0.33227458596229553,
      "learning_rate": 0.0004726296701715489,
      "loss": 2.1815,
      "step": 115
    },
    {
      "epoch": 1.9492385786802031,
      "grad_norm": 0.32235994935035706,
      "learning_rate": 0.00046841839888625623,
      "loss": 2.1927,
      "step": 120
    },
    {
      "epoch": 2.030456852791878,
      "grad_norm": 0.5263159275054932,
      "learning_rate": 0.0004639275170232734,
      "loss": 2.1143,
      "step": 125
    },
    {
      "epoch": 2.030456852791878,
      "eval_loss": 2.0160372257232666,
      "eval_runtime": 88.8689,
      "eval_samples_per_second": 11.253,
      "eval_steps_per_second": 1.407,
      "step": 125
    },
    {
      "epoch": 2.1116751269035534,
      "grad_norm": 0.4549558162689209,
      "learning_rate": 0.0004591627736315743,
      "loss": 2.1318,
      "step": 130
    },
    {
      "epoch": 2.1928934010152283,
      "grad_norm": 0.670222818851471,
      "learning_rate": 0.0004541302683468084,
      "loss": 2.1073,
      "step": 135
    },
    {
      "epoch": 2.2741116751269037,
      "grad_norm": 0.536625862121582,
      "learning_rate": 0.0004488364435827881,
      "loss": 2.1021,
      "step": 140
    },
    {
      "epoch": 2.3553299492385786,
      "grad_norm": 0.3629292845726013,
      "learning_rate": 0.00044328807628416644,
      "loss": 2.0923,
      "step": 145
    },
    {
      "epoch": 2.436548223350254,
      "grad_norm": 0.2985295057296753,
      "learning_rate": 0.0004374922692508611,
      "loss": 2.0753,
      "step": 150
    },
    {
      "epoch": 2.436548223350254,
      "eval_loss": 1.948063611984253,
      "eval_runtime": 88.8793,
      "eval_samples_per_second": 11.251,
      "eval_steps_per_second": 1.406,
      "step": 150
    },
    {
      "epoch": 2.517766497461929,
      "grad_norm": 0.26300689578056335,
      "learning_rate": 0.0004314564420453311,
      "loss": 2.0447,
      "step": 155
    },
    {
      "epoch": 2.598984771573604,
      "grad_norm": 0.1663161963224411,
      "learning_rate": 0.0004251883214943475,
      "loss": 2.0429,
      "step": 160
    },
    {
      "epoch": 2.6802030456852792,
      "grad_norm": 0.6070680022239685,
      "learning_rate": 0.0004186959317974155,
      "loss": 2.0699,
      "step": 165
    },
    {
      "epoch": 2.761421319796954,
      "grad_norm": 1.7647202014923096,
      "learning_rate": 0.00041198758425451266,
      "loss": 2.0304,
      "step": 170
    },
    {
      "epoch": 2.8426395939086295,
      "grad_norm": 0.5461482405662537,
      "learning_rate": 0.00040507186662629185,
      "loss": 2.0449,
      "step": 175
    },
    {
      "epoch": 2.8426395939086295,
      "eval_loss": 1.8957690000534058,
      "eval_runtime": 88.8966,
      "eval_samples_per_second": 11.249,
      "eval_steps_per_second": 1.406,
      "step": 175
    },
    {
      "epoch": 2.9238578680203045,
      "grad_norm": 1.156761646270752,
      "learning_rate": 0.0003979576321403705,
      "loss": 1.9947,
      "step": 180
    },
    {
      "epoch": 3.00507614213198,
      "grad_norm": 0.465610533952713,
      "learning_rate": 0.0003906539881577793,
      "loss": 2.0121,
      "step": 185
    },
    {
      "epoch": 3.0862944162436547,
      "grad_norm": 0.1722594052553177,
      "learning_rate": 0.0003831702845140801,
      "loss": 2.0141,
      "step": 190
    },
    {
      "epoch": 3.16751269035533,
      "grad_norm": 0.24845992028713226,
      "learning_rate": 0.00037551610155007613,
      "loss": 1.9536,
      "step": 195
    },
    {
      "epoch": 3.248730964467005,
      "grad_norm": 0.4477495551109314,
      "learning_rate": 0.00036770123784744027,
      "loss": 2.0201,
      "step": 200
    },
    {
      "epoch": 3.248730964467005,
      "eval_loss": 1.8864437341690063,
      "eval_runtime": 88.9059,
      "eval_samples_per_second": 11.248,
      "eval_steps_per_second": 1.406,
      "step": 200
    },
    {
      "epoch": 3.3299492385786804,
      "grad_norm": 0.5112472176551819,
      "learning_rate": 0.00035973569768495855,
      "loss": 1.9846,
      "step": 205
    },
    {
      "epoch": 3.4111675126903553,
      "grad_norm": 0.6808192133903503,
      "learning_rate": 0.0003516296782314491,
      "loss": 1.9839,
      "step": 210
    },
    {
      "epoch": 3.4923857868020303,
      "grad_norm": 0.24290819466114044,
      "learning_rate": 0.00034339355649175095,
      "loss": 2.0036,
      "step": 215
    },
    {
      "epoch": 3.5736040609137056,
      "grad_norm": 0.34100961685180664,
      "learning_rate": 0.00033503787602249364,
      "loss": 2.0178,
      "step": 220
    },
    {
      "epoch": 3.6548223350253806,
      "grad_norm": 0.7999327182769775,
      "learning_rate": 0.00032657333343465356,
      "loss": 1.9947,
      "step": 225
    },
    {
      "epoch": 3.6548223350253806,
      "eval_loss": 1.8772450685501099,
      "eval_runtime": 88.947,
      "eval_samples_per_second": 11.243,
      "eval_steps_per_second": 1.405,
      "step": 225
    },
    {
      "epoch": 3.736040609137056,
      "grad_norm": 0.3293197453022003,
      "learning_rate": 0.0003180107647001769,
      "loss": 2.0078,
      "step": 230
    },
    {
      "epoch": 3.817258883248731,
      "grad_norm": 0.19176721572875977,
      "learning_rate": 0.0003093611312801979,
      "loss": 1.9594,
      "step": 235
    },
    {
      "epoch": 3.8984771573604062,
      "grad_norm": 0.24580970406532288,
      "learning_rate": 0.00030063550609261025,
      "loss": 1.9514,
      "step": 240
    },
    {
      "epoch": 3.979695431472081,
      "grad_norm": 0.28957948088645935,
      "learning_rate": 0.000291845059336957,
      "loss": 1.9809,
      "step": 245
    },
    {
      "epoch": 4.060913705583756,
      "grad_norm": 0.19108451902866364,
      "learning_rate": 0.0002830010441947834,
      "loss": 1.9295,
      "step": 250
    },
    {
      "epoch": 4.060913705583756,
      "eval_loss": 1.8687773942947388,
      "eval_runtime": 88.8506,
      "eval_samples_per_second": 11.255,
      "eval_steps_per_second": 1.407,
      "step": 250
    },
    {
      "epoch": 4.1421319796954315,
      "grad_norm": 0.18111436069011688,
      "learning_rate": 0.00027411478242376017,
      "loss": 1.9789,
      "step": 255
    },
    {
      "epoch": 4.223350253807107,
      "grad_norm": 0.4630391299724579,
      "learning_rate": 0.00026519764986401774,
      "loss": 1.9657,
      "step": 260
    },
    {
      "epoch": 4.304568527918782,
      "grad_norm": 0.2104698121547699,
      "learning_rate": 0.000256261061875247,
      "loss": 1.9403,
      "step": 265
    },
    {
      "epoch": 4.385786802030457,
      "grad_norm": 0.4305352568626404,
      "learning_rate": 0.0002473164587232079,
      "loss": 1.9762,
      "step": 270
    },
    {
      "epoch": 4.467005076142132,
      "grad_norm": 0.25098443031311035,
      "learning_rate": 0.0002383752909343547,
      "loss": 1.9564,
      "step": 275
    },
    {
      "epoch": 4.467005076142132,
      "eval_loss": 1.8538275957107544,
      "eval_runtime": 88.8543,
      "eval_samples_per_second": 11.254,
      "eval_steps_per_second": 1.407,
      "step": 275
    },
    {
      "epoch": 4.548223350253807,
      "grad_norm": 0.3380163908004761,
      "learning_rate": 0.0002294490046373259,
      "loss": 1.9096,
      "step": 280
    },
    {
      "epoch": 4.629441624365482,
      "grad_norm": 0.18323899805545807,
      "learning_rate": 0.00022054902691006405,
      "loss": 1.9775,
      "step": 285
    },
    {
      "epoch": 4.710659898477157,
      "grad_norm": 0.13836567103862762,
      "learning_rate": 0.00021168675115132315,
      "loss": 1.9888,
      "step": 290
    },
    {
      "epoch": 4.791878172588833,
      "grad_norm": 0.20029225945472717,
      "learning_rate": 0.00020287352249529153,
      "loss": 1.9547,
      "step": 295
    },
    {
      "epoch": 4.873096446700508,
      "grad_norm": 0.16367541253566742,
      "learning_rate": 0.00019412062328800044,
      "loss": 1.9243,
      "step": 300
    },
    {
      "epoch": 4.873096446700508,
      "eval_loss": 1.8513847589492798,
      "eval_runtime": 88.8985,
      "eval_samples_per_second": 11.249,
      "eval_steps_per_second": 1.406,
      "step": 300
    },
    {
      "epoch": 4.9543147208121825,
      "grad_norm": 0.14919321238994598,
      "learning_rate": 0.000185439258644112,
      "loss": 1.9703,
      "step": 305
    },
    {
      "epoch": 5.035532994923858,
      "grad_norm": 0.20038799941539764,
      "learning_rate": 0.00017684054210257517,
      "loss": 1.9606,
      "step": 310
    },
    {
      "epoch": 5.116751269035533,
      "grad_norm": 0.3811420798301697,
      "learning_rate": 0.00016833548139951395,
      "loss": 1.971,
      "step": 315
    },
    {
      "epoch": 5.197969543147208,
      "grad_norm": 0.34453725814819336,
      "learning_rate": 0.0001599349643765599,
      "loss": 1.9342,
      "step": 320
    },
    {
      "epoch": 5.279187817258883,
      "grad_norm": 0.22745971381664276,
      "learning_rate": 0.0001516497450426686,
      "loss": 1.958,
      "step": 325
    },
    {
      "epoch": 5.279187817258883,
      "eval_loss": 1.8567323684692383,
      "eval_runtime": 88.9414,
      "eval_samples_per_second": 11.243,
      "eval_steps_per_second": 1.405,
      "step": 325
    },
    {
      "epoch": 5.3604060913705585,
      "grad_norm": 0.3236718475818634,
      "learning_rate": 0.00014349042980726362,
      "loss": 1.927,
      "step": 330
    },
    {
      "epoch": 5.441624365482234,
      "grad_norm": 0.5001394748687744,
      "learning_rate": 0.0001354674639023318,
      "loss": 1.9415,
      "step": 335
    },
    {
      "epoch": 5.522842639593908,
      "grad_norm": 0.5372613668441772,
      "learning_rate": 0.00012759111801085066,
      "loss": 1.9346,
      "step": 340
    },
    {
      "epoch": 5.604060913705584,
      "grad_norm": 0.39717650413513184,
      "learning_rate": 0.00011987147511866788,
      "loss": 1.9491,
      "step": 345
    },
    {
      "epoch": 5.685279187817259,
      "grad_norm": 0.34586790204048157,
      "learning_rate": 0.00011231841760666186,
      "loss": 1.872,
      "step": 350
    },
    {
      "epoch": 5.685279187817259,
      "eval_loss": 1.8259689807891846,
      "eval_runtime": 88.9255,
      "eval_samples_per_second": 11.245,
      "eval_steps_per_second": 1.406,
      "step": 350
    }
  ],
  "logging_steps": 5,
  "max_steps": 488,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.435956091813888e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}