{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9997305930800908,
  "eval_steps": 500,
  "global_step": 3247,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003078936227533387,
      "grad_norm": 1.7886259464726153,
      "learning_rate": 6.153846153846154e-07,
      "loss": 2.4065,
      "step": 1
    },
    {
      "epoch": 0.0015394681137666935,
      "grad_norm": 0.7856091829075883,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 2.2623,
      "step": 5
    },
    {
      "epoch": 0.003078936227533387,
      "grad_norm": 1.7492528772148481,
      "learning_rate": 6.153846153846155e-06,
      "loss": 2.3366,
      "step": 10
    },
    {
      "epoch": 0.004618404341300081,
      "grad_norm": 1.8810015033279819,
      "learning_rate": 9.230769230769232e-06,
      "loss": 2.2464,
      "step": 15
    },
    {
      "epoch": 0.006157872455066774,
      "grad_norm": 1.5384917280101869,
      "learning_rate": 1.230769230769231e-05,
      "loss": 2.4102,
      "step": 20
    },
    {
      "epoch": 0.007697340568833468,
      "grad_norm": 1.531064264819813,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 2.2919,
      "step": 25
    },
    {
      "epoch": 0.009236808682600161,
      "grad_norm": 1.5245321390728928,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 2.0073,
      "step": 30
    },
    {
      "epoch": 0.010776276796366856,
      "grad_norm": 2.3551208511966792,
      "learning_rate": 2.1538461538461542e-05,
      "loss": 2.0615,
      "step": 35
    },
    {
      "epoch": 0.012315744910133548,
      "grad_norm": 0.870620603646086,
      "learning_rate": 2.461538461538462e-05,
      "loss": 1.933,
      "step": 40
    },
    {
      "epoch": 0.013855213023900243,
      "grad_norm": 2.0784218387109545,
      "learning_rate": 2.7692307692307694e-05,
      "loss": 1.7955,
      "step": 45
    },
    {
      "epoch": 0.015394681137666935,
      "grad_norm": 1.7032213696426233,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 1.623,
      "step": 50
    },
    {
      "epoch": 0.01693414925143363,
      "grad_norm": 1.199804375605637,
      "learning_rate": 3.384615384615385e-05,
      "loss": 1.5078,
      "step": 55
    },
    {
      "epoch": 0.018473617365200323,
      "grad_norm": 0.6513653498128114,
      "learning_rate": 3.692307692307693e-05,
      "loss": 1.3395,
      "step": 60
    },
    {
      "epoch": 0.020013085478967015,
      "grad_norm": 0.35836956933859887,
      "learning_rate": 4e-05,
      "loss": 1.2971,
      "step": 65
    },
    {
      "epoch": 0.02155255359273371,
      "grad_norm": 0.27349495518377487,
      "learning_rate": 4.3076923076923084e-05,
      "loss": 1.197,
      "step": 70
    },
    {
      "epoch": 0.023092021706500404,
      "grad_norm": 0.3594788242269333,
      "learning_rate": 4.615384615384616e-05,
      "loss": 1.1353,
      "step": 75
    },
    {
      "epoch": 0.024631489820267097,
      "grad_norm": 0.20878396712118252,
      "learning_rate": 4.923076923076924e-05,
      "loss": 1.1295,
      "step": 80
    },
    {
      "epoch": 0.026170957934033793,
      "grad_norm": 0.15589572115960823,
      "learning_rate": 5.230769230769231e-05,
      "loss": 1.0822,
      "step": 85
    },
    {
      "epoch": 0.027710426047800486,
      "grad_norm": 0.10395722149316333,
      "learning_rate": 5.538461538461539e-05,
      "loss": 1.0227,
      "step": 90
    },
    {
      "epoch": 0.02924989416156718,
      "grad_norm": 0.11175530250493737,
      "learning_rate": 5.846153846153847e-05,
      "loss": 1.039,
      "step": 95
    },
    {
      "epoch": 0.03078936227533387,
      "grad_norm": 0.100290581587309,
      "learning_rate": 6.153846153846155e-05,
      "loss": 1.0021,
      "step": 100
    },
    {
      "epoch": 0.03232883038910057,
      "grad_norm": 0.1096086423198071,
      "learning_rate": 6.461538461538462e-05,
      "loss": 1.0098,
      "step": 105
    },
    {
      "epoch": 0.03386829850286726,
      "grad_norm": 0.11413161368432702,
      "learning_rate": 6.76923076923077e-05,
      "loss": 1.002,
      "step": 110
    },
    {
      "epoch": 0.03540776661663395,
      "grad_norm": 0.0883433361451938,
      "learning_rate": 7.076923076923078e-05,
      "loss": 1.0046,
      "step": 115
    },
    {
      "epoch": 0.036947234730400645,
      "grad_norm": 0.07140741468698422,
      "learning_rate": 7.384615384615386e-05,
      "loss": 0.9814,
      "step": 120
    },
    {
      "epoch": 0.03848670284416734,
      "grad_norm": 0.07872746773100796,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.9886,
      "step": 125
    },
    {
      "epoch": 0.04002617095793403,
      "grad_norm": 0.054900803691268776,
      "learning_rate": 8e-05,
      "loss": 1.0179,
      "step": 130
    },
    {
      "epoch": 0.04156563907170073,
      "grad_norm": 0.08778632715972133,
      "learning_rate": 8.307692307692309e-05,
      "loss": 0.9783,
      "step": 135
    },
    {
      "epoch": 0.04310510718546742,
      "grad_norm": 0.09659582528614778,
      "learning_rate": 8.615384615384617e-05,
      "loss": 0.9835,
      "step": 140
    },
    {
      "epoch": 0.044644575299234115,
      "grad_norm": 0.0703291247525324,
      "learning_rate": 8.923076923076924e-05,
      "loss": 0.963,
      "step": 145
    },
    {
      "epoch": 0.04618404341300081,
      "grad_norm": 0.06160228816589926,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.9635,
      "step": 150
    },
    {
      "epoch": 0.0477235115267675,
      "grad_norm": 0.067929217872403,
      "learning_rate": 9.53846153846154e-05,
      "loss": 0.9564,
      "step": 155
    },
    {
      "epoch": 0.049262979640534194,
      "grad_norm": 0.07059304417071058,
      "learning_rate": 9.846153846153848e-05,
      "loss": 0.9705,
      "step": 160
    },
    {
      "epoch": 0.050802447754300886,
      "grad_norm": 0.058717760542211006,
      "learning_rate": 0.00010153846153846153,
      "loss": 0.9493,
      "step": 165
    },
    {
      "epoch": 0.052341915868067586,
      "grad_norm": 0.07991988340066557,
      "learning_rate": 0.00010461538461538463,
      "loss": 0.9701,
      "step": 170
    },
    {
      "epoch": 0.05388138398183428,
      "grad_norm": 0.06759612940962959,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.9825,
      "step": 175
    },
    {
      "epoch": 0.05542085209560097,
      "grad_norm": 0.07707113030085129,
      "learning_rate": 0.00011076923076923077,
      "loss": 0.927,
      "step": 180
    },
    {
      "epoch": 0.056960320209367664,
      "grad_norm": 0.07777181849810574,
      "learning_rate": 0.00011384615384615384,
      "loss": 0.9597,
      "step": 185
    },
    {
      "epoch": 0.05849978832313436,
      "grad_norm": 0.06976692683939785,
      "learning_rate": 0.00011692307692307694,
      "loss": 0.9776,
      "step": 190
    },
    {
      "epoch": 0.06003925643690105,
      "grad_norm": 0.06744155796823793,
      "learning_rate": 0.00012,
      "loss": 0.9436,
      "step": 195
    },
    {
      "epoch": 0.06157872455066774,
      "grad_norm": 0.0906791338365497,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.9508,
      "step": 200
    },
    {
      "epoch": 0.06311819266443444,
      "grad_norm": 0.06833140260133326,
      "learning_rate": 0.00012615384615384615,
      "loss": 0.9305,
      "step": 205
    },
    {
      "epoch": 0.06465766077820113,
      "grad_norm": 0.07448730899375566,
      "learning_rate": 0.00012923076923076923,
      "loss": 0.959,
      "step": 210
    },
    {
      "epoch": 0.06619712889196783,
      "grad_norm": 0.06295871227457246,
      "learning_rate": 0.0001323076923076923,
      "loss": 0.9471,
      "step": 215
    },
    {
      "epoch": 0.06773659700573452,
      "grad_norm": 0.06671768002683286,
      "learning_rate": 0.0001353846153846154,
      "loss": 0.9475,
      "step": 220
    },
    {
      "epoch": 0.06927606511950121,
      "grad_norm": 0.06807246484129376,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.9176,
      "step": 225
    },
    {
      "epoch": 0.0708155332332679,
      "grad_norm": 0.06294758872039433,
      "learning_rate": 0.00014153846153846156,
      "loss": 0.9353,
      "step": 230
    },
    {
      "epoch": 0.0723550013470346,
      "grad_norm": 0.061076404800252414,
      "learning_rate": 0.0001446153846153846,
      "loss": 0.9029,
      "step": 235
    },
    {
      "epoch": 0.07389446946080129,
      "grad_norm": 0.07165204896011747,
      "learning_rate": 0.00014769230769230772,
      "loss": 0.9522,
      "step": 240
    },
    {
      "epoch": 0.07543393757456798,
      "grad_norm": 0.06142770831333145,
      "learning_rate": 0.00015076923076923077,
      "loss": 0.9604,
      "step": 245
    },
    {
      "epoch": 0.07697340568833468,
      "grad_norm": 0.055747916316771985,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.9469,
      "step": 250
    },
    {
      "epoch": 0.07851287380210137,
      "grad_norm": 0.06495847309063495,
      "learning_rate": 0.00015692307692307693,
      "loss": 0.8872,
      "step": 255
    },
    {
      "epoch": 0.08005234191586806,
      "grad_norm": 0.06953979479889788,
      "learning_rate": 0.00016,
      "loss": 0.9591,
      "step": 260
    },
    {
      "epoch": 0.08159181002963477,
      "grad_norm": 0.05997292205848339,
      "learning_rate": 0.0001630769230769231,
      "loss": 0.9317,
      "step": 265
    },
    {
      "epoch": 0.08313127814340146,
      "grad_norm": 0.06683367864789459,
      "learning_rate": 0.00016615384615384617,
      "loss": 0.9717,
      "step": 270
    },
    {
      "epoch": 0.08467074625716815,
      "grad_norm": 0.07414531970479905,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.9253,
      "step": 275
    },
    {
      "epoch": 0.08621021437093485,
      "grad_norm": 0.058289217740252974,
      "learning_rate": 0.00017230769230769234,
      "loss": 0.9423,
      "step": 280
    },
    {
      "epoch": 0.08774968248470154,
      "grad_norm": 0.06156915214820435,
      "learning_rate": 0.0001753846153846154,
      "loss": 0.9479,
      "step": 285
    },
    {
      "epoch": 0.08928915059846823,
      "grad_norm": 0.06839948189559994,
      "learning_rate": 0.00017846153846153847,
      "loss": 0.9547,
      "step": 290
    },
    {
      "epoch": 0.09082861871223492,
      "grad_norm": 0.07044677125831877,
      "learning_rate": 0.00018153846153846155,
      "loss": 0.9407,
      "step": 295
    },
    {
      "epoch": 0.09236808682600162,
      "grad_norm": 0.06454272103387144,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.9404,
      "step": 300
    },
    {
      "epoch": 0.09390755493976831,
      "grad_norm": 0.06540138303042092,
      "learning_rate": 0.0001876923076923077,
      "loss": 0.9283,
      "step": 305
    },
    {
      "epoch": 0.095447023053535,
      "grad_norm": 0.06606202830142367,
      "learning_rate": 0.0001907692307692308,
      "loss": 0.947,
      "step": 310
    },
    {
      "epoch": 0.0969864911673017,
      "grad_norm": 0.06565881788931771,
      "learning_rate": 0.00019384615384615385,
      "loss": 0.9608,
      "step": 315
    },
    {
      "epoch": 0.09852595928106839,
      "grad_norm": 0.05723384408289347,
      "learning_rate": 0.00019692307692307696,
      "loss": 0.922,
      "step": 320
    },
    {
      "epoch": 0.10006542739483508,
      "grad_norm": 0.08595339007629281,
      "learning_rate": 0.0002,
      "loss": 0.9297,
      "step": 325
    },
    {
      "epoch": 0.10160489550860177,
      "grad_norm": 0.06146101896381126,
      "learning_rate": 0.00019999855506507185,
      "loss": 0.9568,
      "step": 330
    },
    {
      "epoch": 0.10314436362236847,
      "grad_norm": 0.06043681352930332,
      "learning_rate": 0.00019999422030204418,
      "loss": 0.9392,
      "step": 335
    },
    {
      "epoch": 0.10468383173613517,
      "grad_norm": 0.058784129825152236,
      "learning_rate": 0.00019998699583618593,
      "loss": 0.9335,
      "step": 340
    },
    {
      "epoch": 0.10622329984990186,
      "grad_norm": 0.07339711767455756,
      "learning_rate": 0.00019997688187627482,
      "loss": 0.9388,
      "step": 345
    },
    {
      "epoch": 0.10776276796366856,
      "grad_norm": 0.06863424893787205,
      "learning_rate": 0.0001999638787145911,
      "loss": 0.9552,
      "step": 350
    },
    {
      "epoch": 0.10930223607743525,
      "grad_norm": 0.06416167746235103,
      "learning_rate": 0.0001999479867269092,
      "loss": 0.9398,
      "step": 355
    },
    {
      "epoch": 0.11084170419120194,
      "grad_norm": 0.07293566060961007,
      "learning_rate": 0.00019992920637248697,
      "loss": 0.9283,
      "step": 360
    },
    {
      "epoch": 0.11238117230496864,
      "grad_norm": 0.05795397130511251,
      "learning_rate": 0.00019990753819405213,
      "loss": 0.9156,
      "step": 365
    },
    {
      "epoch": 0.11392064041873533,
      "grad_norm": 0.05925802755851074,
      "learning_rate": 0.00019988298281778684,
      "loss": 0.9436,
      "step": 370
    },
    {
      "epoch": 0.11546010853250202,
      "grad_norm": 0.07382813289359652,
      "learning_rate": 0.00019985554095330955,
      "loss": 0.9079,
      "step": 375
    },
    {
      "epoch": 0.11699957664626871,
      "grad_norm": 0.07244233003395625,
      "learning_rate": 0.0001998252133936544,
      "loss": 0.9372,
      "step": 380
    },
    {
      "epoch": 0.1185390447600354,
      "grad_norm": 0.059856414735217266,
      "learning_rate": 0.00019979200101524845,
      "loss": 0.9462,
      "step": 385
    },
    {
      "epoch": 0.1200785128738021,
      "grad_norm": 0.06311186958064242,
      "learning_rate": 0.00019975590477788613,
      "loss": 0.9337,
      "step": 390
    },
    {
      "epoch": 0.12161798098756879,
      "grad_norm": 0.054257277491550376,
      "learning_rate": 0.0001997169257247018,
      "loss": 0.8951,
      "step": 395
    },
    {
      "epoch": 0.12315744910133548,
      "grad_norm": 0.05597241826184769,
      "learning_rate": 0.00019967506498213931,
      "loss": 0.9489,
      "step": 400
    },
    {
      "epoch": 0.12469691721510218,
      "grad_norm": 0.0575366172657177,
      "learning_rate": 0.00019963032375991966,
      "loss": 0.9612,
      "step": 405
    },
    {
      "epoch": 0.12623638532886888,
      "grad_norm": 0.05498473779799205,
      "learning_rate": 0.00019958270335100595,
      "loss": 0.9167,
      "step": 410
    },
    {
      "epoch": 0.12777585344263556,
      "grad_norm": 0.05451363935096892,
      "learning_rate": 0.00019953220513156602,
      "loss": 0.9657,
      "step": 415
    },
    {
      "epoch": 0.12931532155640227,
      "grad_norm": 0.05273299621055326,
      "learning_rate": 0.0001994788305609327,
      "loss": 0.9268,
      "step": 420
    },
    {
      "epoch": 0.13085478967016895,
      "grad_norm": 0.05850087024230285,
      "learning_rate": 0.00019942258118156163,
      "loss": 0.9237,
      "step": 425
    },
    {
      "epoch": 0.13239425778393565,
      "grad_norm": 0.05234156950662489,
      "learning_rate": 0.00019936345861898663,
      "loss": 0.9429,
      "step": 430
    },
    {
      "epoch": 0.13393372589770233,
      "grad_norm": 0.1134035523279181,
      "learning_rate": 0.0001993014645817728,
      "loss": 0.9358,
      "step": 435
    },
    {
      "epoch": 0.13547319401146904,
      "grad_norm": 0.048751485298395646,
      "learning_rate": 0.00019923660086146723,
      "loss": 0.9318,
      "step": 440
    },
    {
      "epoch": 0.13701266212523572,
      "grad_norm": 0.055361809469683866,
      "learning_rate": 0.0001991688693325469,
      "loss": 0.9289,
      "step": 445
    },
    {
      "epoch": 0.13855213023900242,
      "grad_norm": 0.058284785462008666,
      "learning_rate": 0.00019909827195236493,
      "loss": 0.9291,
      "step": 450
    },
    {
      "epoch": 0.14009159835276913,
      "grad_norm": 0.05368624032603138,
      "learning_rate": 0.00019902481076109372,
      "loss": 0.9497,
      "step": 455
    },
    {
      "epoch": 0.1416310664665358,
      "grad_norm": 0.05055173721774897,
      "learning_rate": 0.00019894848788166604,
      "loss": 0.9404,
      "step": 460
    },
    {
      "epoch": 0.14317053458030252,
      "grad_norm": 0.05824332224771405,
      "learning_rate": 0.00019886930551971387,
      "loss": 0.955,
      "step": 465
    },
    {
      "epoch": 0.1447100026940692,
      "grad_norm": 0.07214196511041003,
      "learning_rate": 0.0001987872659635043,
      "loss": 0.9202,
      "step": 470
    },
    {
      "epoch": 0.1462494708078359,
      "grad_norm": 0.058974012454576195,
      "learning_rate": 0.00019870237158387384,
      "loss": 0.9317,
      "step": 475
    },
    {
      "epoch": 0.14778893892160258,
      "grad_norm": 0.0496321161862721,
      "learning_rate": 0.00019861462483415952,
      "loss": 0.9236,
      "step": 480
    },
    {
      "epoch": 0.1493284070353693,
      "grad_norm": 0.05467253796263981,
      "learning_rate": 0.0001985240282501282,
      "loss": 0.9253,
      "step": 485
    },
    {
      "epoch": 0.15086787514913597,
      "grad_norm": 0.04861402346330219,
      "learning_rate": 0.0001984305844499033,
      "loss": 0.9106,
      "step": 490
    },
    {
      "epoch": 0.15240734326290267,
      "grad_norm": 0.05397477192351095,
      "learning_rate": 0.00019833429613388902,
      "loss": 0.9287,
      "step": 495
    },
    {
      "epoch": 0.15394681137666935,
      "grad_norm": 0.05808861261090533,
      "learning_rate": 0.0001982351660846924,
      "loss": 0.9266,
      "step": 500
    },
    {
      "epoch": 0.15548627949043606,
      "grad_norm": 0.049151963300644703,
      "learning_rate": 0.00019813319716704278,
      "loss": 0.9255,
      "step": 505
    },
    {
      "epoch": 0.15702574760420274,
      "grad_norm": 0.060327540548373824,
      "learning_rate": 0.00019802839232770921,
      "loss": 0.9636,
      "step": 510
    },
    {
      "epoch": 0.15856521571796944,
      "grad_norm": 0.050076605968131446,
      "learning_rate": 0.00019792075459541518,
      "loss": 0.9083,
      "step": 515
    },
    {
      "epoch": 0.16010468383173612,
      "grad_norm": 0.05710597237534321,
      "learning_rate": 0.00019781028708075102,
      "loss": 0.9154,
      "step": 520
    },
    {
      "epoch": 0.16164415194550283,
      "grad_norm": 0.05119017067438771,
      "learning_rate": 0.00019769699297608417,
      "loss": 0.9144,
      "step": 525
    },
    {
      "epoch": 0.16318362005926954,
      "grad_norm": 0.06546986936590168,
      "learning_rate": 0.00019758087555546682,
      "loss": 0.951,
      "step": 530
    },
    {
      "epoch": 0.16472308817303621,
      "grad_norm": 0.05293989495985302,
      "learning_rate": 0.0001974619381745413,
      "loss": 0.9443,
      "step": 535
    },
    {
      "epoch": 0.16626255628680292,
      "grad_norm": 0.05268282028511608,
      "learning_rate": 0.00019734018427044307,
      "loss": 0.9213,
      "step": 540
    },
    {
      "epoch": 0.1678020244005696,
      "grad_norm": 0.05073432746268571,
      "learning_rate": 0.0001972156173617016,
      "loss": 0.9235,
      "step": 545
    },
    {
      "epoch": 0.1693414925143363,
      "grad_norm": 0.052364528405400936,
      "learning_rate": 0.00019708824104813837,
      "loss": 0.9234,
      "step": 550
    },
    {
      "epoch": 0.17088096062810298,
      "grad_norm": 0.05163092419435362,
      "learning_rate": 0.00019695805901076308,
      "loss": 0.943,
      "step": 555
    },
    {
      "epoch": 0.1724204287418697,
      "grad_norm": 0.05013282990770516,
      "learning_rate": 0.00019682507501166718,
      "loss": 0.9674,
      "step": 560
    },
    {
      "epoch": 0.17395989685563637,
      "grad_norm": 0.05496807027320449,
      "learning_rate": 0.00019668929289391523,
      "loss": 0.9121,
      "step": 565
    },
    {
      "epoch": 0.17549936496940308,
      "grad_norm": 0.05295664915113558,
      "learning_rate": 0.00019655071658143366,
      "loss": 0.9309,
      "step": 570
    },
    {
      "epoch": 0.17703883308316976,
      "grad_norm": 0.05770430579600094,
      "learning_rate": 0.00019640935007889755,
      "loss": 0.9525,
      "step": 575
    },
    {
      "epoch": 0.17857830119693646,
      "grad_norm": 0.04967368156780794,
      "learning_rate": 0.0001962651974716149,
      "loss": 0.9421,
      "step": 580
    },
    {
      "epoch": 0.18011776931070314,
      "grad_norm": 0.05542180958681771,
      "learning_rate": 0.0001961182629254084,
      "loss": 0.9032,
      "step": 585
    },
    {
      "epoch": 0.18165723742446985,
      "grad_norm": 0.05157460362945081,
      "learning_rate": 0.00019596855068649522,
      "loss": 0.9281,
      "step": 590
    },
    {
      "epoch": 0.18319670553823653,
      "grad_norm": 0.05474461334176702,
      "learning_rate": 0.00019581606508136426,
      "loss": 0.9405,
      "step": 595
    },
    {
      "epoch": 0.18473617365200323,
      "grad_norm": 0.04999123312677221,
      "learning_rate": 0.00019566081051665098,
      "loss": 0.894,
      "step": 600
    },
    {
      "epoch": 0.18627564176576994,
      "grad_norm": 0.04663933495586406,
      "learning_rate": 0.00019550279147901036,
      "loss": 0.9582,
      "step": 605
    },
    {
      "epoch": 0.18781510987953662,
      "grad_norm": 0.05748400574383358,
      "learning_rate": 0.00019534201253498682,
      "loss": 0.9202,
      "step": 610
    },
    {
      "epoch": 0.18935457799330332,
      "grad_norm": 0.061058791061366406,
      "learning_rate": 0.0001951784783308827,
      "loss": 0.8908,
      "step": 615
    },
    {
      "epoch": 0.19089404610707,
      "grad_norm": 0.05602302672114044,
      "learning_rate": 0.0001950121935926236,
      "loss": 0.9195,
      "step": 620
    },
    {
      "epoch": 0.1924335142208367,
      "grad_norm": 0.05101288671521673,
      "learning_rate": 0.00019484316312562205,
      "loss": 0.9041,
      "step": 625
    },
    {
      "epoch": 0.1939729823346034,
      "grad_norm": 0.04764728716875625,
      "learning_rate": 0.00019467139181463862,
      "loss": 0.9414,
      "step": 630
    },
    {
      "epoch": 0.1955124504483701,
      "grad_norm": 0.05001047990653959,
      "learning_rate": 0.00019449688462364056,
      "loss": 0.9191,
      "step": 635
    },
    {
      "epoch": 0.19705191856213677,
      "grad_norm": 0.0551438348203942,
      "learning_rate": 0.00019431964659565867,
      "loss": 0.9265,
      "step": 640
    },
    {
      "epoch": 0.19859138667590348,
      "grad_norm": 0.05127776312579667,
      "learning_rate": 0.0001941396828526412,
      "loss": 0.927,
      "step": 645
    },
    {
      "epoch": 0.20013085478967016,
      "grad_norm": 0.05648580243210199,
      "learning_rate": 0.00019395699859530623,
      "loss": 0.9023,
      "step": 650
    },
    {
      "epoch": 0.20167032290343687,
      "grad_norm": 0.0548444576545457,
      "learning_rate": 0.00019377159910299093,
      "loss": 0.9127,
      "step": 655
    },
    {
      "epoch": 0.20320979101720354,
      "grad_norm": 0.04982308571194646,
      "learning_rate": 0.00019358348973349943,
      "loss": 0.896,
      "step": 660
    },
    {
      "epoch": 0.20474925913097025,
      "grad_norm": 0.05789843448294904,
      "learning_rate": 0.00019339267592294763,
      "loss": 0.9235,
      "step": 665
    },
    {
      "epoch": 0.20628872724473693,
      "grad_norm": 0.050550388560361946,
      "learning_rate": 0.00019319916318560635,
      "loss": 0.9062,
      "step": 670
    },
    {
      "epoch": 0.20782819535850364,
      "grad_norm": 0.054057089770620986,
      "learning_rate": 0.00019300295711374187,
      "loss": 0.9225,
      "step": 675
    },
    {
      "epoch": 0.20936766347227034,
      "grad_norm": 0.05060226186203681,
      "learning_rate": 0.00019280406337745428,
      "loss": 0.9006,
      "step": 680
    },
    {
      "epoch": 0.21090713158603702,
      "grad_norm": 0.060487703264749865,
      "learning_rate": 0.00019260248772451377,
      "loss": 0.9271,
      "step": 685
    },
    {
      "epoch": 0.21244659969980373,
      "grad_norm": 0.05087519096486981,
      "learning_rate": 0.0001923982359801943,
      "loss": 0.924,
      "step": 690
    },
    {
      "epoch": 0.2139860678135704,
      "grad_norm": 0.050600273155108975,
      "learning_rate": 0.00019219131404710552,
      "loss": 0.9467,
      "step": 695
    },
    {
      "epoch": 0.21552553592733711,
      "grad_norm": 0.05205827603309181,
      "learning_rate": 0.00019198172790502196,
      "loss": 0.906,
      "step": 700
    },
    {
      "epoch": 0.2170650040411038,
      "grad_norm": 0.047259129341584924,
      "learning_rate": 0.0001917694836107104,
      "loss": 0.9164,
      "step": 705
    },
    {
      "epoch": 0.2186044721548705,
      "grad_norm": 0.045967289482132204,
      "learning_rate": 0.00019155458729775467,
      "loss": 0.9324,
      "step": 710
    },
    {
      "epoch": 0.22014394026863718,
      "grad_norm": 0.047454682336569326,
      "learning_rate": 0.0001913370451763786,
      "loss": 0.9025,
      "step": 715
    },
    {
      "epoch": 0.22168340838240388,
      "grad_norm": 0.04616693147301789,
      "learning_rate": 0.00019111686353326631,
      "loss": 0.9055,
      "step": 720
    },
    {
      "epoch": 0.22322287649617056,
      "grad_norm": 0.049953829560007476,
      "learning_rate": 0.00019089404873138082,
      "loss": 0.9355,
      "step": 725
    },
    {
      "epoch": 0.22476234460993727,
      "grad_norm": 0.050646893052510175,
      "learning_rate": 0.00019066860720977986,
      "loss": 0.9284,
      "step": 730
    },
    {
      "epoch": 0.22630181272370395,
      "grad_norm": 0.05034487250112421,
      "learning_rate": 0.00019044054548343002,
      "loss": 0.9101,
      "step": 735
    },
    {
      "epoch": 0.22784128083747066,
      "grad_norm": 0.04887230407763029,
      "learning_rate": 0.0001902098701430184,
      "loss": 0.9405,
      "step": 740
    },
    {
      "epoch": 0.22938074895123733,
      "grad_norm": 0.05156805123913926,
      "learning_rate": 0.00018997658785476214,
      "loss": 0.9408,
      "step": 745
    },
    {
      "epoch": 0.23092021706500404,
      "grad_norm": 0.045216425675149156,
      "learning_rate": 0.00018974070536021572,
      "loss": 0.9478,
      "step": 750
    },
    {
      "epoch": 0.23245968517877075,
      "grad_norm": 0.053986948133609866,
      "learning_rate": 0.00018950222947607625,
      "loss": 0.9329,
      "step": 755
    },
    {
      "epoch": 0.23399915329253743,
      "grad_norm": 0.04883518041049424,
      "learning_rate": 0.0001892611670939865,
      "loss": 0.8833,
      "step": 760
    },
    {
      "epoch": 0.23553862140630413,
      "grad_norm": 0.05491819215808765,
      "learning_rate": 0.00018901752518033548,
      "loss": 0.9292,
      "step": 765
    },
    {
      "epoch": 0.2370780895200708,
      "grad_norm": 0.048033151778087875,
      "learning_rate": 0.0001887713107760575,
      "loss": 0.9279,
      "step": 770
    },
    {
      "epoch": 0.23861755763383752,
      "grad_norm": 0.05085333839214927,
      "learning_rate": 0.00018852253099642833,
      "loss": 0.9014,
      "step": 775
    },
    {
      "epoch": 0.2401570257476042,
      "grad_norm": 0.047876992365331196,
      "learning_rate": 0.0001882711930308599,
      "loss": 0.9312,
      "step": 780
    },
    {
      "epoch": 0.2416964938613709,
      "grad_norm": 0.048396575556683336,
      "learning_rate": 0.00018801730414269225,
      "loss": 0.9176,
      "step": 785
    },
    {
      "epoch": 0.24323596197513758,
      "grad_norm": 0.05047818799105187,
      "learning_rate": 0.0001877608716689839,
      "loss": 0.9412,
      "step": 790
    },
    {
      "epoch": 0.2447754300889043,
      "grad_norm": 0.050841254906904526,
      "learning_rate": 0.00018750190302029956,
      "loss": 0.9013,
      "step": 795
    },
    {
      "epoch": 0.24631489820267097,
      "grad_norm": 0.05248393498729352,
      "learning_rate": 0.00018724040568049612,
      "loss": 0.922,
      "step": 800
    },
    {
      "epoch": 0.24785436631643767,
      "grad_norm": 0.05277634783788006,
      "learning_rate": 0.00018697638720650646,
      "loss": 0.9136,
      "step": 805
    },
    {
      "epoch": 0.24939383443020435,
      "grad_norm": 0.04577816551575828,
      "learning_rate": 0.00018670985522812084,
      "loss": 0.8914,
      "step": 810
    },
    {
      "epoch": 0.25093330254397106,
      "grad_norm": 0.047133784056639066,
      "learning_rate": 0.0001864408174477665,
      "loss": 0.9204,
      "step": 815
    },
    {
      "epoch": 0.25247277065773777,
      "grad_norm": 0.047470050981737545,
      "learning_rate": 0.00018616928164028523,
      "loss": 0.8883,
      "step": 820
    },
    {
      "epoch": 0.2540122387715045,
      "grad_norm": 0.0451286945270303,
      "learning_rate": 0.00018589525565270844,
      "loss": 0.9137,
      "step": 825
    },
    {
      "epoch": 0.2555517068852711,
      "grad_norm": 0.05014389409036849,
      "learning_rate": 0.0001856187474040306,
      "loss": 0.9281,
      "step": 830
    },
    {
      "epoch": 0.25709117499903783,
      "grad_norm": 0.050236591187208905,
      "learning_rate": 0.00018533976488498016,
      "loss": 0.9101,
      "step": 835
    },
    {
      "epoch": 0.25863064311280454,
      "grad_norm": 0.05001661471805326,
      "learning_rate": 0.0001850583161577889,
      "loss": 0.9349,
      "step": 840
    },
    {
      "epoch": 0.26017011122657124,
      "grad_norm": 0.04645985082158435,
      "learning_rate": 0.00018477440935595873,
      "loss": 0.9524,
      "step": 845
    },
    {
      "epoch": 0.2617095793403379,
      "grad_norm": 0.048589216000343084,
      "learning_rate": 0.00018448805268402672,
      "loss": 0.9331,
      "step": 850
    },
    {
      "epoch": 0.2632490474541046,
      "grad_norm": 0.04873829009307298,
      "learning_rate": 0.00018419925441732804,
      "loss": 0.914,
      "step": 855
    },
    {
      "epoch": 0.2647885155678713,
      "grad_norm": 0.052282665220643836,
      "learning_rate": 0.00018390802290175673,
      "loss": 0.9244,
      "step": 860
    },
    {
      "epoch": 0.266327983681638,
      "grad_norm": 0.04761057179968087,
      "learning_rate": 0.00018361436655352456,
      "loss": 0.9032,
      "step": 865
    },
    {
      "epoch": 0.26786745179540467,
      "grad_norm": 0.04691078478567084,
      "learning_rate": 0.00018331829385891783,
      "loss": 0.9025,
      "step": 870
    },
    {
      "epoch": 0.26940691990917137,
      "grad_norm": 0.04689928092597921,
      "learning_rate": 0.00018301981337405212,
      "loss": 0.923,
      "step": 875
    },
    {
      "epoch": 0.2709463880229381,
      "grad_norm": 0.050329806919839966,
      "learning_rate": 0.00018271893372462497,
      "loss": 0.9251,
      "step": 880
    },
    {
      "epoch": 0.2724858561367048,
      "grad_norm": 0.04966113875059789,
      "learning_rate": 0.00018241566360566665,
      "loss": 0.9053,
      "step": 885
    },
    {
      "epoch": 0.27402532425047144,
      "grad_norm": 0.05275353859230967,
      "learning_rate": 0.00018211001178128892,
      "loss": 0.9212,
      "step": 890
    },
    {
      "epoch": 0.27556479236423814,
      "grad_norm": 0.04759465019587229,
      "learning_rate": 0.00018180198708443173,
      "loss": 0.9324,
      "step": 895
    },
    {
      "epoch": 0.27710426047800485,
      "grad_norm": 0.05028165140499341,
      "learning_rate": 0.00018149159841660795,
      "loss": 0.9149,
      "step": 900
    },
    {
      "epoch": 0.27864372859177156,
      "grad_norm": 0.05511573197678425,
      "learning_rate": 0.00018117885474764613,
      "loss": 0.956,
      "step": 905
    },
    {
      "epoch": 0.28018319670553826,
      "grad_norm": 0.05105095212128217,
      "learning_rate": 0.00018086376511543126,
      "loss": 0.9385,
      "step": 910
    },
    {
      "epoch": 0.2817226648193049,
      "grad_norm": 0.050601170231212314,
      "learning_rate": 0.00018054633862564368,
      "loss": 0.919,
      "step": 915
    },
    {
      "epoch": 0.2832621329330716,
      "grad_norm": 0.05104843063118903,
      "learning_rate": 0.0001802265844514958,
      "loss": 0.8866,
      "step": 920
    },
    {
      "epoch": 0.2848016010468383,
      "grad_norm": 0.04820628993315859,
      "learning_rate": 0.0001799045118334671,
      "loss": 0.9294,
      "step": 925
    },
    {
      "epoch": 0.28634106916060503,
      "grad_norm": 0.05584344656415011,
      "learning_rate": 0.00017958013007903713,
      "loss": 0.9308,
      "step": 930
    },
    {
      "epoch": 0.2878805372743717,
      "grad_norm": 0.05127226028923933,
      "learning_rate": 0.0001792534485624164,
      "loss": 0.9179,
      "step": 935
    },
    {
      "epoch": 0.2894200053881384,
      "grad_norm": 0.04901315749056508,
      "learning_rate": 0.00017892447672427563,
      "loss": 0.9166,
      "step": 940
    },
    {
      "epoch": 0.2909594735019051,
      "grad_norm": 0.044541590041367256,
      "learning_rate": 0.00017859322407147272,
      "loss": 0.9296,
      "step": 945
    },
    {
      "epoch": 0.2924989416156718,
      "grad_norm": 0.04514811142602194,
      "learning_rate": 0.00017825970017677832,
      "loss": 0.9403,
      "step": 950
    },
    {
      "epoch": 0.29403840972943845,
      "grad_norm": 0.0497766203514951,
      "learning_rate": 0.00017792391467859886,
      "loss": 0.8969,
      "step": 955
    },
    {
      "epoch": 0.29557787784320516,
      "grad_norm": 0.046772051107154496,
      "learning_rate": 0.0001775858772806983,
      "loss": 0.9436,
      "step": 960
    },
    {
      "epoch": 0.29711734595697187,
      "grad_norm": 0.045795964863886866,
      "learning_rate": 0.00017724559775191744,
      "loss": 0.9097,
      "step": 965
    },
    {
      "epoch": 0.2986568140707386,
      "grad_norm": 0.05044904668621051,
      "learning_rate": 0.00017690308592589182,
      "loss": 0.9214,
      "step": 970
    },
    {
      "epoch": 0.3001962821845053,
      "grad_norm": 0.052229656751926,
      "learning_rate": 0.0001765583517007675,
      "loss": 0.8822,
      "step": 975
    },
    {
      "epoch": 0.30173575029827193,
      "grad_norm": 0.049422521625903534,
      "learning_rate": 0.00017621140503891488,
      "loss": 0.8945,
      "step": 980
    },
    {
      "epoch": 0.30327521841203864,
      "grad_norm": 0.054031889194258664,
      "learning_rate": 0.00017586225596664102,
      "loss": 0.9078,
      "step": 985
    },
    {
      "epoch": 0.30481468652580535,
      "grad_norm": 0.04906883819137053,
      "learning_rate": 0.00017551091457389966,
      "loss": 0.9123,
      "step": 990
    },
    {
      "epoch": 0.30635415463957205,
      "grad_norm": 0.046767739088330644,
      "learning_rate": 0.00017515739101399983,
      "loss": 0.9042,
      "step": 995
    },
    {
      "epoch": 0.3078936227533387,
      "grad_norm": 0.04562148158039666,
      "learning_rate": 0.00017480169550331231,
      "loss": 0.8927,
      "step": 1000
    },
    {
      "epoch": 0.3094330908671054,
      "grad_norm": 0.047045878433571706,
      "learning_rate": 0.00017444383832097442,
      "loss": 0.9057,
      "step": 1005
    },
    {
      "epoch": 0.3109725589808721,
      "grad_norm": 0.0515310037789988,
      "learning_rate": 0.00017408382980859305,
      "loss": 0.9173,
      "step": 1010
    },
    {
      "epoch": 0.3125120270946388,
      "grad_norm": 0.04945975879857112,
      "learning_rate": 0.00017372168036994566,
      "loss": 0.8967,
      "step": 1015
    },
    {
      "epoch": 0.3140514952084055,
      "grad_norm": 0.045915450946519304,
      "learning_rate": 0.00017335740047067972,
      "loss": 0.9063,
      "step": 1020
    },
    {
      "epoch": 0.3155909633221722,
      "grad_norm": 0.04696718756630669,
      "learning_rate": 0.0001729910006380102,
      "loss": 0.9204,
      "step": 1025
    },
    {
      "epoch": 0.3171304314359389,
      "grad_norm": 0.054017714335650606,
      "learning_rate": 0.00017262249146041546,
      "loss": 0.9488,
      "step": 1030
    },
    {
      "epoch": 0.3186698995497056,
      "grad_norm": 0.046208799015592085,
      "learning_rate": 0.00017225188358733107,
      "loss": 0.8934,
      "step": 1035
    },
    {
      "epoch": 0.32020936766347224,
      "grad_norm": 0.05439874345483133,
      "learning_rate": 0.00017187918772884232,
      "loss": 0.9364,
      "step": 1040
    },
    {
      "epoch": 0.32174883577723895,
      "grad_norm": 0.045811341739051416,
      "learning_rate": 0.00017150441465537447,
      "loss": 0.935,
      "step": 1045
    },
    {
      "epoch": 0.32328830389100566,
      "grad_norm": 0.04551232779575255,
      "learning_rate": 0.00017112757519738154,
      "loss": 0.9106,
      "step": 1050
    },
    {
      "epoch": 0.32482777200477236,
      "grad_norm": 0.050527589367316904,
      "learning_rate": 0.0001707486802450335,
      "loss": 0.9217,
      "step": 1055
    },
    {
      "epoch": 0.32636724011853907,
      "grad_norm": 0.04871891012305943,
      "learning_rate": 0.00017036774074790132,
      "loss": 0.8848,
      "step": 1060
    },
    {
      "epoch": 0.3279067082323057,
      "grad_norm": 0.048363348820198084,
      "learning_rate": 0.00016998476771464072,
      "loss": 0.916,
      "step": 1065
    },
    {
      "epoch": 0.32944617634607243,
      "grad_norm": 0.04799882669750604,
      "learning_rate": 0.00016959977221267392,
      "loss": 0.8934,
      "step": 1070
    },
    {
      "epoch": 0.33098564445983913,
      "grad_norm": 0.049532832158656184,
      "learning_rate": 0.0001692127653678699,
      "loss": 0.9207,
      "step": 1075
    },
    {
      "epoch": 0.33252511257360584,
      "grad_norm": 0.049670369145985355,
      "learning_rate": 0.00016882375836422284,
      "loss": 0.9112,
      "step": 1080
    },
    {
      "epoch": 0.3340645806873725,
      "grad_norm": 0.05066949742909503,
      "learning_rate": 0.00016843276244352885,
      "loss": 0.9369,
      "step": 1085
    },
    {
      "epoch": 0.3356040488011392,
      "grad_norm": 0.04945281177695019,
      "learning_rate": 0.00016803978890506113,
      "loss": 0.9444,
      "step": 1090
    },
    {
      "epoch": 0.3371435169149059,
      "grad_norm": 0.05353405785121247,
      "learning_rate": 0.00016764484910524358,
      "loss": 0.9059,
      "step": 1095
    },
    {
      "epoch": 0.3386829850286726,
      "grad_norm": 0.04906241678051854,
      "learning_rate": 0.00016724795445732243,
      "loss": 0.8797,
      "step": 1100
    },
    {
      "epoch": 0.34022245314243926,
      "grad_norm": 0.057154558228479126,
      "learning_rate": 0.00016684911643103642,
      "loss": 0.9355,
      "step": 1105
    },
    {
      "epoch": 0.34176192125620597,
      "grad_norm": 0.049827926481343945,
      "learning_rate": 0.0001664483465522855,
      "loss": 0.9132,
      "step": 1110
    },
    {
      "epoch": 0.3433013893699727,
      "grad_norm": 0.04854134175156812,
      "learning_rate": 0.00016604565640279754,
      "loss": 0.9221,
      "step": 1115
    },
    {
      "epoch": 0.3448408574837394,
      "grad_norm": 0.05196607300498118,
      "learning_rate": 0.0001656410576197938,
      "loss": 0.9293,
      "step": 1120
    },
    {
      "epoch": 0.3463803255975061,
      "grad_norm": 0.04834785242459327,
      "learning_rate": 0.0001652345618956526,
      "loss": 0.9454,
      "step": 1125
    },
    {
      "epoch": 0.34791979371127274,
      "grad_norm": 0.04617356598468755,
      "learning_rate": 0.00016482618097757122,
      "loss": 0.9275,
      "step": 1130
    },
    {
      "epoch": 0.34945926182503945,
      "grad_norm": 0.04868318584802262,
      "learning_rate": 0.00016441592666722684,
      "loss": 0.9141,
      "step": 1135
    },
    {
      "epoch": 0.35099872993880615,
      "grad_norm": 0.04447091312889386,
      "learning_rate": 0.00016400381082043507,
      "loss": 0.956,
      "step": 1140
    },
    {
      "epoch": 0.35253819805257286,
      "grad_norm": 0.04567651469536377,
      "learning_rate": 0.00016358984534680748,
      "loss": 0.9272,
      "step": 1145
    },
    {
      "epoch": 0.3540776661663395,
      "grad_norm": 0.04650640057414895,
      "learning_rate": 0.00016317404220940758,
      "loss": 0.9025,
      "step": 1150
    },
    {
      "epoch": 0.3556171342801062,
      "grad_norm": 0.04372629674005958,
      "learning_rate": 0.00016275641342440483,
      "loss": 0.9294,
      "step": 1155
    },
    {
      "epoch": 0.3571566023938729,
      "grad_norm": 0.050895686251958096,
      "learning_rate": 0.0001623369710607277,
      "loss": 0.8862,
      "step": 1160
    },
    {
      "epoch": 0.35869607050763963,
      "grad_norm": 0.04714825832143354,
      "learning_rate": 0.00016191572723971455,
      "loss": 0.9477,
      "step": 1165
    },
    {
      "epoch": 0.3602355386214063,
      "grad_norm": 0.051310184729132374,
      "learning_rate": 0.00016149269413476353,
      "loss": 0.883,
      "step": 1170
    },
    {
      "epoch": 0.361775006735173,
      "grad_norm": 0.05062348820990283,
      "learning_rate": 0.00016106788397098095,
      "loss": 0.8848,
      "step": 1175
    },
    {
      "epoch": 0.3633144748489397,
      "grad_norm": 0.04566500242723035,
      "learning_rate": 0.0001606413090248276,
      "loss": 0.8814,
      "step": 1180
    },
    {
      "epoch": 0.3648539429627064,
      "grad_norm": 0.051602640362734715,
      "learning_rate": 0.00016021298162376428,
      "loss": 0.9051,
      "step": 1185
    },
    {
      "epoch": 0.36639341107647305,
      "grad_norm": 0.048994793028383776,
      "learning_rate": 0.00015978291414589542,
      "loss": 0.9181,
      "step": 1190
    },
    {
      "epoch": 0.36793287919023976,
      "grad_norm": 0.04717320718802025,
      "learning_rate": 0.0001593511190196115,
      "loss": 0.9299,
      "step": 1195
    },
    {
      "epoch": 0.36947234730400647,
      "grad_norm": 0.050428580066856195,
      "learning_rate": 0.00015891760872322963,
      "loss": 0.8928,
      "step": 1200
    },
    {
      "epoch": 0.37101181541777317,
      "grad_norm": 0.043926223439215366,
      "learning_rate": 0.00015848239578463325,
      "loss": 0.9203,
      "step": 1205
    },
    {
      "epoch": 0.3725512835315399,
      "grad_norm": 0.051179038558784154,
      "learning_rate": 0.00015804549278090982,
      "loss": 0.8929,
      "step": 1210
    },
    {
      "epoch": 0.37409075164530653,
      "grad_norm": 0.05028409083764298,
      "learning_rate": 0.00015760691233798757,
      "loss": 0.8955,
      "step": 1215
    },
    {
      "epoch": 0.37563021975907324,
      "grad_norm": 0.0478697697271406,
      "learning_rate": 0.00015716666713027055,
      "loss": 0.9083,
      "step": 1220
    },
    {
      "epoch": 0.37716968787283994,
      "grad_norm": 0.05223173395313064,
      "learning_rate": 0.00015672476988027228,
      "loss": 0.9158,
      "step": 1225
    },
    {
      "epoch": 0.37870915598660665,
      "grad_norm": 0.05595336863735006,
      "learning_rate": 0.0001562812333582482,
      "loss": 0.8856,
      "step": 1230
    },
    {
      "epoch": 0.3802486241003733,
      "grad_norm": 0.047063201161414325,
      "learning_rate": 0.00015583607038182655,
      "loss": 0.9077,
      "step": 1235
    },
    {
      "epoch": 0.38178809221414,
      "grad_norm": 0.05077602726776445,
      "learning_rate": 0.000155389293815638,
      "loss": 0.9109,
      "step": 1240
    },
    {
      "epoch": 0.3833275603279067,
      "grad_norm": 0.04803050608192193,
      "learning_rate": 0.00015494091657094385,
      "loss": 0.9117,
      "step": 1245
    },
    {
      "epoch": 0.3848670284416734,
      "grad_norm": 0.052600153553541314,
      "learning_rate": 0.00015449095160526292,
      "loss": 0.9351,
      "step": 1250
    },
    {
      "epoch": 0.38640649655544007,
      "grad_norm": 0.05078425004669163,
      "learning_rate": 0.00015403941192199718,
      "loss": 0.9158,
      "step": 1255
    },
    {
      "epoch": 0.3879459646692068,
      "grad_norm": 0.047300542169020104,
      "learning_rate": 0.0001535863105700558,
      "loss": 0.9222,
      "step": 1260
    },
    {
      "epoch": 0.3894854327829735,
      "grad_norm": 0.0499337783219946,
      "learning_rate": 0.00015313166064347814,
      "loss": 0.8993,
      "step": 1265
    },
    {
      "epoch": 0.3910249008967402,
      "grad_norm": 0.04421937665282515,
      "learning_rate": 0.00015267547528105538,
      "loss": 0.9147,
      "step": 1270
    },
    {
      "epoch": 0.3925643690105069,
      "grad_norm": 0.044571852373292344,
      "learning_rate": 0.0001522177676659508,
      "loss": 0.9415,
      "step": 1275
    },
    {
      "epoch": 0.39410383712427355,
      "grad_norm": 0.046596705359225846,
      "learning_rate": 0.00015175855102531887,
      "loss": 0.8979,
      "step": 1280
    },
    {
      "epoch": 0.39564330523804025,
      "grad_norm": 0.045724837857095094,
      "learning_rate": 0.00015129783862992283,
      "loss": 0.899,
      "step": 1285
    },
    {
      "epoch": 0.39718277335180696,
      "grad_norm": 0.053727794778771945,
      "learning_rate": 0.0001508356437937512,
      "loss": 0.9224,
      "step": 1290
    },
    {
      "epoch": 0.39872224146557367,
      "grad_norm": 0.0510139813187336,
      "learning_rate": 0.00015037197987363338,
      "loss": 0.9109,
      "step": 1295
    },
    {
      "epoch": 0.4002617095793403,
      "grad_norm": 0.04914990325645437,
      "learning_rate": 0.0001499068602688532,
      "loss": 0.9385,
      "step": 1300
    },
    {
      "epoch": 0.401801177693107,
      "grad_norm": 0.04759626671434196,
      "learning_rate": 0.00014944029842076185,
      "loss": 0.91,
      "step": 1305
    },
    {
      "epoch": 0.40334064580687373,
      "grad_norm": 0.048748643361189885,
      "learning_rate": 0.0001489723078123896,
      "loss": 0.9186,
      "step": 1310
    },
    {
      "epoch": 0.40488011392064044,
      "grad_norm": 0.047860946511890864,
      "learning_rate": 0.00014850290196805594,
      "loss": 0.9218,
      "step": 1315
    },
    {
      "epoch": 0.4064195820344071,
      "grad_norm": 0.04550478467051693,
      "learning_rate": 0.00014803209445297887,
      "loss": 0.8886,
      "step": 1320
    },
    {
      "epoch": 0.4079590501481738,
      "grad_norm": 0.04833318781042355,
      "learning_rate": 0.00014755989887288285,
      "loss": 0.9221,
      "step": 1325
    },
    {
      "epoch": 0.4094985182619405,
      "grad_norm": 0.05020532883685666,
      "learning_rate": 0.00014708632887360564,
      "loss": 0.9162,
      "step": 1330
    },
    {
      "epoch": 0.4110379863757072,
      "grad_norm": 0.04679679704616009,
      "learning_rate": 0.0001466113981407039,
      "loss": 0.9232,
      "step": 1335
    },
    {
      "epoch": 0.41257745448947386,
      "grad_norm": 0.050508164787588954,
      "learning_rate": 0.00014613512039905765,
      "loss": 0.911,
      "step": 1340
    },
    {
      "epoch": 0.41411692260324057,
      "grad_norm": 0.04830099079596549,
      "learning_rate": 0.00014565750941247386,
      "loss": 0.8917,
      "step": 1345
    },
    {
      "epoch": 0.4156563907170073,
      "grad_norm": 0.048848878196708544,
      "learning_rate": 0.0001451785789832884,
      "loss": 0.898,
      "step": 1350
    },
    {
      "epoch": 0.417195858830774,
      "grad_norm": 0.05215004480641985,
      "learning_rate": 0.00014469834295196743,
      "loss": 0.9239,
      "step": 1355
    },
    {
      "epoch": 0.4187353269445407,
      "grad_norm": 0.04994491938163197,
      "learning_rate": 0.00014421681519670722,
      "loss": 0.9189,
      "step": 1360
    },
    {
      "epoch": 0.42027479505830734,
      "grad_norm": 0.048232822341065185,
      "learning_rate": 0.0001437340096330332,
      "loss": 0.9255,
      "step": 1365
    },
    {
      "epoch": 0.42181426317207404,
      "grad_norm": 0.04919041827285363,
      "learning_rate": 0.0001432499402133979,
      "loss": 0.9164,
      "step": 1370
    },
    {
      "epoch": 0.42335373128584075,
      "grad_norm": 0.04623444150419545,
      "learning_rate": 0.0001427646209267775,
      "loss": 0.9087,
      "step": 1375
    },
    {
      "epoch": 0.42489319939960746,
      "grad_norm": 0.048472709471847446,
      "learning_rate": 0.00014227806579826774,
      "loss": 0.8969,
      "step": 1380
    },
    {
      "epoch": 0.4264326675133741,
      "grad_norm": 0.049327204957984185,
      "learning_rate": 0.00014179028888867867,
      "loss": 0.949,
      "step": 1385
    },
    {
      "epoch": 0.4279721356271408,
      "grad_norm": 0.05555195843727436,
      "learning_rate": 0.00014130130429412815,
      "loss": 0.9044,
      "step": 1390
    },
    {
      "epoch": 0.4295116037409075,
      "grad_norm": 0.051799419936980215,
      "learning_rate": 0.0001408111261456346,
      "loss": 0.8921,
      "step": 1395
    },
    {
      "epoch": 0.43105107185467423,
      "grad_norm": 0.046270078670221725,
      "learning_rate": 0.00014031976860870855,
      "loss": 0.8874,
      "step": 1400
    },
    {
      "epoch": 0.4325905399684409,
      "grad_norm": 0.04941817581350544,
      "learning_rate": 0.00013982724588294335,
      "loss": 0.9001,
      "step": 1405
    },
    {
      "epoch": 0.4341300080822076,
      "grad_norm": 0.04887226355400606,
      "learning_rate": 0.00013933357220160476,
      "loss": 0.9353,
      "step": 1410
    },
    {
      "epoch": 0.4356694761959743,
      "grad_norm": 0.04821337321743884,
      "learning_rate": 0.00013883876183121973,
      "loss": 0.9272,
      "step": 1415
    },
    {
      "epoch": 0.437208944309741,
      "grad_norm": 0.049898799489967206,
      "learning_rate": 0.000138342829071164,
      "loss": 0.9174,
      "step": 1420
    },
    {
      "epoch": 0.4387484124235077,
      "grad_norm": 0.048656203845601646,
      "learning_rate": 0.00013784578825324885,
      "loss": 0.8906,
      "step": 1425
    },
    {
      "epoch": 0.44028788053727436,
      "grad_norm": 0.04620511855527758,
      "learning_rate": 0.00013734765374130717,
      "loss": 0.9087,
      "step": 1430
    },
    {
      "epoch": 0.44182734865104106,
      "grad_norm": 0.04999574904136337,
      "learning_rate": 0.00013684843993077788,
      "loss": 0.8928,
      "step": 1435
    },
    {
      "epoch": 0.44336681676480777,
      "grad_norm": 0.04817308835362055,
      "learning_rate": 0.00013634816124829063,
      "loss": 0.8994,
      "step": 1440
    },
    {
      "epoch": 0.4449062848785745,
      "grad_norm": 0.044807384662721876,
      "learning_rate": 0.0001358468321512481,
      "loss": 0.9323,
      "step": 1445
    },
    {
      "epoch": 0.4464457529923411,
      "grad_norm": 0.04745745776343273,
      "learning_rate": 0.00013534446712740877,
      "loss": 0.9087,
      "step": 1450
    },
    {
      "epoch": 0.44798522110610783,
      "grad_norm": 0.049663621405145894,
      "learning_rate": 0.0001348410806944681,
      "loss": 0.8972,
      "step": 1455
    },
    {
      "epoch": 0.44952468921987454,
      "grad_norm": 0.05332229094490998,
      "learning_rate": 0.00013433668739963882,
      "loss": 0.9014,
      "step": 1460
    },
    {
      "epoch": 0.45106415733364125,
      "grad_norm": 0.04572502563110564,
      "learning_rate": 0.00013383130181923071,
      "loss": 0.9036,
      "step": 1465
    },
    {
      "epoch": 0.4526036254474079,
      "grad_norm": 0.04805188541183551,
      "learning_rate": 0.00013332493855822936,
      "loss": 0.8927,
      "step": 1470
    },
    {
      "epoch": 0.4541430935611746,
      "grad_norm": 0.04946728536218321,
      "learning_rate": 0.00013281761224987398,
      "loss": 0.884,
      "step": 1475
    },
    {
      "epoch": 0.4556825616749413,
      "grad_norm": 0.04761071510226589,
      "learning_rate": 0.00013230933755523466,
      "loss": 0.9059,
      "step": 1480
    },
    {
      "epoch": 0.457222029788708,
      "grad_norm": 0.04991160344244152,
      "learning_rate": 0.00013180012916278854,
      "loss": 0.9204,
      "step": 1485
    },
    {
      "epoch": 0.45876149790247467,
      "grad_norm": 0.04997242360939681,
      "learning_rate": 0.00013129000178799548,
      "loss": 0.9116,
      "step": 1490
    },
    {
      "epoch": 0.4603009660162414,
      "grad_norm": 0.04894858600532897,
      "learning_rate": 0.00013077897017287272,
      "loss": 0.8855,
      "step": 1495
    },
    {
      "epoch": 0.4618404341300081,
      "grad_norm": 0.05029992919887125,
      "learning_rate": 0.00013026704908556888,
      "loss": 0.9303,
      "step": 1500
    },
    {
      "epoch": 0.4633799022437748,
      "grad_norm": 0.050898543846493435,
      "learning_rate": 0.0001297542533199371,
      "loss": 0.9209,
      "step": 1505
    },
    {
      "epoch": 0.4649193703575415,
      "grad_norm": 0.04963049003889223,
      "learning_rate": 0.00012924059769510768,
      "loss": 0.9156,
      "step": 1510
    },
    {
      "epoch": 0.46645883847130815,
      "grad_norm": 0.049519531874776083,
      "learning_rate": 0.00012872609705505964,
      "loss": 0.9273,
      "step": 1515
    },
    {
      "epoch": 0.46799830658507485,
      "grad_norm": 0.05098025085500099,
      "learning_rate": 0.00012821076626819196,
      "loss": 0.9145,
      "step": 1520
    },
    {
      "epoch": 0.46953777469884156,
      "grad_norm": 0.052520698863764266,
      "learning_rate": 0.00012769462022689363,
      "loss": 0.9059,
      "step": 1525
    },
    {
      "epoch": 0.47107724281260827,
      "grad_norm": 0.046885402287568066,
      "learning_rate": 0.0001271776738471136,
      "loss": 0.9084,
      "step": 1530
    },
    {
      "epoch": 0.4726167109263749,
      "grad_norm": 0.05134637502690939,
      "learning_rate": 0.00012665994206792938,
      "loss": 0.9333,
      "step": 1535
    },
    {
      "epoch": 0.4741561790401416,
      "grad_norm": 0.044736051310860794,
      "learning_rate": 0.00012614143985111565,
      "loss": 0.966,
      "step": 1540
    },
    {
      "epoch": 0.47569564715390833,
      "grad_norm": 0.052866581288498866,
      "learning_rate": 0.00012562218218071164,
      "loss": 0.8973,
      "step": 1545
    },
    {
      "epoch": 0.47723511526767504,
      "grad_norm": 0.04912631183968688,
      "learning_rate": 0.0001251021840625883,
      "loss": 0.901,
      "step": 1550
    },
    {
      "epoch": 0.4787745833814417,
      "grad_norm": 0.05796784976691198,
      "learning_rate": 0.00012458146052401442,
      "loss": 0.9028,
      "step": 1555
    },
    {
      "epoch": 0.4803140514952084,
      "grad_norm": 0.048382356277699094,
      "learning_rate": 0.00012406002661322264,
      "loss": 0.8968,
      "step": 1560
    },
    {
      "epoch": 0.4818535196089751,
      "grad_norm": 0.04824535557964773,
      "learning_rate": 0.00012353789739897437,
      "loss": 0.9223,
      "step": 1565
    },
    {
      "epoch": 0.4833929877227418,
      "grad_norm": 0.04791248494600298,
      "learning_rate": 0.00012301508797012432,
      "loss": 0.905,
      "step": 1570
    },
    {
      "epoch": 0.4849324558365085,
      "grad_norm": 0.04920168920976926,
      "learning_rate": 0.00012249161343518466,
      "loss": 0.8898,
      "step": 1575
    },
    {
      "epoch": 0.48647192395027516,
      "grad_norm": 0.04821219071584707,
      "learning_rate": 0.00012196748892188816,
      "loss": 0.9172,
      "step": 1580
    },
    {
      "epoch": 0.48801139206404187,
      "grad_norm": 0.04801131628609144,
      "learning_rate": 0.00012144272957675108,
      "loss": 0.9079,
      "step": 1585
    },
    {
      "epoch": 0.4895508601778086,
      "grad_norm": 0.0494886933485408,
      "learning_rate": 0.00012091735056463562,
      "loss": 0.8918,
      "step": 1590
    },
    {
      "epoch": 0.4910903282915753,
      "grad_norm": 0.05157629416069951,
      "learning_rate": 0.00012039136706831145,
      "loss": 0.9354,
      "step": 1595
    },
    {
      "epoch": 0.49262979640534194,
      "grad_norm": 0.04980659124460385,
      "learning_rate": 0.00011986479428801709,
      "loss": 0.9131,
      "step": 1600
    },
    {
      "epoch": 0.49416926451910864,
      "grad_norm": 0.04888303131555791,
      "learning_rate": 0.00011933764744102058,
      "loss": 0.8903,
      "step": 1605
    },
    {
      "epoch": 0.49570873263287535,
      "grad_norm": 0.05003112563947073,
      "learning_rate": 0.00011880994176117976,
      "loss": 0.9179,
      "step": 1610
    },
    {
      "epoch": 0.49724820074664206,
      "grad_norm": 0.04777538154853683,
      "learning_rate": 0.00011828169249850201,
      "loss": 0.9163,
      "step": 1615
    },
    {
      "epoch": 0.4987876688604087,
      "grad_norm": 0.05074786015766115,
      "learning_rate": 0.00011775291491870351,
      "loss": 0.8999,
      "step": 1620
    },
    {
      "epoch": 0.5003271369741754,
      "grad_norm": 0.049063505070354355,
      "learning_rate": 0.00011722362430276816,
      "loss": 0.907,
      "step": 1625
    },
    {
      "epoch": 0.5018666050879421,
      "grad_norm": 0.049098848873215875,
      "learning_rate": 0.00011669383594650593,
      "loss": 0.9324,
      "step": 1630
    },
    {
      "epoch": 0.5034060732017088,
      "grad_norm": 0.046020639564825015,
      "learning_rate": 0.00011616356516011083,
      "loss": 0.8911,
      "step": 1635
    },
    {
      "epoch": 0.5049455413154755,
      "grad_norm": 0.05089226172884991,
      "learning_rate": 0.00011563282726771847,
      "loss": 0.9177,
      "step": 1640
    },
    {
      "epoch": 0.5064850094292422,
      "grad_norm": 0.04938039025170907,
      "learning_rate": 0.0001151016376069632,
      "loss": 0.9405,
      "step": 1645
    },
    {
      "epoch": 0.508024477543009,
      "grad_norm": 0.048691616868991525,
      "learning_rate": 0.00011457001152853493,
      "loss": 0.8958,
      "step": 1650
    },
    {
      "epoch": 0.5095639456567755,
      "grad_norm": 0.047488356666995184,
      "learning_rate": 0.00011403796439573544,
      "loss": 0.9082,
      "step": 1655
    },
    {
      "epoch": 0.5111034137705422,
      "grad_norm": 0.04706768719557111,
      "learning_rate": 0.00011350551158403442,
      "loss": 0.9331,
      "step": 1660
    },
    {
      "epoch": 0.512642881884309,
      "grad_norm": 0.04385967327877815,
      "learning_rate": 0.0001129726684806252,
      "loss": 0.8857,
      "step": 1665
    },
    {
      "epoch": 0.5141823499980757,
      "grad_norm": 0.051532938748309255,
      "learning_rate": 0.00011243945048398003,
      "loss": 0.8908,
      "step": 1670
    },
    {
      "epoch": 0.5157218181118424,
      "grad_norm": 0.04623181890622743,
      "learning_rate": 0.000111905873003405,
      "loss": 0.9026,
      "step": 1675
    },
    {
      "epoch": 0.5172612862256091,
      "grad_norm": 0.05177657958980996,
      "learning_rate": 0.00011137195145859494,
      "loss": 0.8843,
      "step": 1680
    },
    {
      "epoch": 0.5188007543393758,
      "grad_norm": 0.046296108567482355,
      "learning_rate": 0.00011083770127918762,
      "loss": 0.8838,
      "step": 1685
    },
    {
      "epoch": 0.5203402224531425,
      "grad_norm": 0.051247343947526836,
      "learning_rate": 0.00011030313790431788,
      "loss": 0.9165,
      "step": 1690
    },
    {
      "epoch": 0.5218796905669091,
      "grad_norm": 0.051691250574067094,
      "learning_rate": 0.00010976827678217161,
      "loss": 0.8813,
      "step": 1695
    },
    {
      "epoch": 0.5234191586806758,
      "grad_norm": 0.04567288707386652,
      "learning_rate": 0.00010923313336953913,
      "loss": 0.8939,
      "step": 1700
    },
    {
      "epoch": 0.5249586267944425,
      "grad_norm": 0.04586936870346886,
      "learning_rate": 0.00010869772313136861,
      "loss": 0.9012,
      "step": 1705
    },
    {
      "epoch": 0.5264980949082092,
      "grad_norm": 0.04460595879520053,
      "learning_rate": 0.00010816206154031916,
      "loss": 0.8875,
      "step": 1710
    },
    {
      "epoch": 0.5280375630219759,
      "grad_norm": 0.050944410064677675,
      "learning_rate": 0.00010762616407631356,
      "loss": 0.9415,
      "step": 1715
    },
    {
      "epoch": 0.5295770311357426,
      "grad_norm": 0.059744943371876844,
      "learning_rate": 0.00010709004622609116,
      "loss": 0.9385,
      "step": 1720
    },
    {
      "epoch": 0.5311164992495093,
      "grad_norm": 0.05553649610942733,
      "learning_rate": 0.00010655372348276006,
      "loss": 0.8979,
      "step": 1725
    },
    {
      "epoch": 0.532655967363276,
      "grad_norm": 0.051926822886669166,
      "learning_rate": 0.00010601721134534959,
      "loss": 0.9113,
      "step": 1730
    },
    {
      "epoch": 0.5341954354770427,
      "grad_norm": 0.04941826428941035,
      "learning_rate": 0.00010548052531836223,
      "loss": 0.905,
      "step": 1735
    },
    {
      "epoch": 0.5357349035908093,
      "grad_norm": 0.0527044621920537,
      "learning_rate": 0.00010494368091132576,
      "loss": 0.9028,
      "step": 1740
    },
    {
      "epoch": 0.537274371704576,
      "grad_norm": 0.048514900283774504,
      "learning_rate": 0.00010440669363834483,
      "loss": 0.8898,
      "step": 1745
    },
    {
      "epoch": 0.5388138398183427,
      "grad_norm": 0.046357229010524525,
      "learning_rate": 0.00010386957901765277,
      "loss": 0.9095,
      "step": 1750
    },
    {
      "epoch": 0.5403533079321095,
      "grad_norm": 0.0464242270463998,
      "learning_rate": 0.00010333235257116313,
      "loss": 0.8576,
      "step": 1755
    },
    {
      "epoch": 0.5418927760458762,
      "grad_norm": 0.04854933121211971,
      "learning_rate": 0.00010279502982402103,
      "loss": 0.918,
      "step": 1760
    },
    {
      "epoch": 0.5434322441596429,
      "grad_norm": 0.04881303108310671,
      "learning_rate": 0.00010225762630415457,
      "loss": 0.8918,
      "step": 1765
    },
    {
      "epoch": 0.5449717122734096,
      "grad_norm": 0.04968155994030721,
      "learning_rate": 0.00010172015754182607,
      "loss": 0.9163,
      "step": 1770
    },
    {
      "epoch": 0.5465111803871763,
      "grad_norm": 0.050904812829817965,
      "learning_rate": 0.00010118263906918331,
      "loss": 0.9132,
      "step": 1775
    },
    {
      "epoch": 0.5480506485009429,
      "grad_norm": 0.04909198498496398,
      "learning_rate": 0.00010064508641981054,
      "loss": 0.8792,
      "step": 1780
    },
    {
      "epoch": 0.5495901166147096,
      "grad_norm": 0.05019294870663726,
      "learning_rate": 0.0001001075151282798,
      "loss": 0.9235,
      "step": 1785
    },
    {
      "epoch": 0.5511295847284763,
      "grad_norm": 0.04910118909724831,
      "learning_rate": 9.956994072970179e-05,
      "loss": 0.9058,
      "step": 1790
    },
    {
      "epoch": 0.552669052842243,
      "grad_norm": 0.04468508395156117,
      "learning_rate": 9.903237875927698e-05,
      "loss": 0.9365,
      "step": 1795
    },
    {
      "epoch": 0.5542085209560097,
      "grad_norm": 0.04546291013701692,
      "learning_rate": 9.849484475184672e-05,
      "loss": 0.8978,
      "step": 1800
    },
    {
      "epoch": 0.5557479890697764,
      "grad_norm": 0.05064130811494575,
      "learning_rate": 9.795735424144428e-05,
      "loss": 0.8882,
      "step": 1805
    },
    {
      "epoch": 0.5572874571835431,
      "grad_norm": 0.04926396297485791,
      "learning_rate": 9.74199227608459e-05,
      "loss": 0.9096,
      "step": 1810
    },
    {
      "epoch": 0.5588269252973098,
      "grad_norm": 0.04864342218523345,
      "learning_rate": 9.688256584112192e-05,
      "loss": 0.8931,
      "step": 1815
    },
    {
      "epoch": 0.5603663934110765,
      "grad_norm": 0.05149354719899412,
      "learning_rate": 9.634529901118799e-05,
      "loss": 0.8993,
      "step": 1820
    },
    {
      "epoch": 0.5619058615248431,
      "grad_norm": 0.048833615891704936,
      "learning_rate": 9.580813779735624e-05,
      "loss": 0.9103,
      "step": 1825
    },
    {
      "epoch": 0.5634453296386098,
      "grad_norm": 0.04911921426500149,
      "learning_rate": 9.52710977228867e-05,
      "loss": 0.9014,
      "step": 1830
    },
    {
      "epoch": 0.5649847977523765,
      "grad_norm": 0.05022485311358825,
      "learning_rate": 9.473419430753864e-05,
      "loss": 0.8591,
      "step": 1835
    },
    {
      "epoch": 0.5665242658661432,
      "grad_norm": 0.04955602494518542,
      "learning_rate": 9.419744306712197e-05,
      "loss": 0.8835,
      "step": 1840
    },
    {
      "epoch": 0.56806373397991,
      "grad_norm": 0.05007271924525583,
      "learning_rate": 9.3660859513049e-05,
      "loss": 0.9356,
      "step": 1845
    },
    {
      "epoch": 0.5696032020936767,
      "grad_norm": 0.05122824824006998,
      "learning_rate": 9.312445915188609e-05,
      "loss": 0.9023,
      "step": 1850
    },
    {
      "epoch": 0.5711426702074434,
      "grad_norm": 0.050844115063622415,
      "learning_rate": 9.258825748490558e-05,
      "loss": 0.9229,
      "step": 1855
    },
    {
      "epoch": 0.5726821383212101,
      "grad_norm": 0.05035025571223379,
      "learning_rate": 9.205227000763788e-05,
      "loss": 0.9195,
      "step": 1860
    },
    {
      "epoch": 0.5742216064349767,
      "grad_norm": 0.049262791068879354,
      "learning_rate": 9.151651220942349e-05,
      "loss": 0.9216,
      "step": 1865
    },
    {
      "epoch": 0.5757610745487434,
      "grad_norm": 0.04738660153724957,
      "learning_rate": 9.098099957296552e-05,
      "loss": 0.9125,
      "step": 1870
    },
    {
      "epoch": 0.5773005426625101,
      "grad_norm": 0.048935820953397643,
      "learning_rate": 9.044574757388224e-05,
      "loss": 0.9094,
      "step": 1875
    },
    {
      "epoch": 0.5788400107762768,
      "grad_norm": 0.04691306454158609,
      "learning_rate": 8.991077168025976e-05,
      "loss": 0.905,
      "step": 1880
    },
    {
      "epoch": 0.5803794788900435,
      "grad_norm": 0.05133111412345183,
      "learning_rate": 8.937608735220527e-05,
      "loss": 0.9186,
      "step": 1885
    },
    {
      "epoch": 0.5819189470038102,
      "grad_norm": 0.051329822743825367,
      "learning_rate": 8.884171004139996e-05,
      "loss": 0.9025,
      "step": 1890
    },
    {
      "epoch": 0.5834584151175769,
      "grad_norm": 0.04706601914619586,
      "learning_rate": 8.830765519065262e-05,
      "loss": 0.8676,
      "step": 1895
    },
    {
      "epoch": 0.5849978832313436,
      "grad_norm": 0.05352891934650077,
      "learning_rate": 8.777393823345343e-05,
      "loss": 0.9192,
      "step": 1900
    },
    {
      "epoch": 0.5865373513451103,
      "grad_norm": 0.04834968781184054,
      "learning_rate": 8.724057459352784e-05,
      "loss": 0.8724,
      "step": 1905
    },
    {
      "epoch": 0.5880768194588769,
      "grad_norm": 0.044222947012426655,
      "learning_rate": 8.670757968439086e-05,
      "loss": 0.9333,
      "step": 1910
    },
    {
      "epoch": 0.5896162875726436,
      "grad_norm": 0.05079157358067232,
      "learning_rate": 8.617496890890179e-05,
      "loss": 0.9094,
      "step": 1915
    },
    {
      "epoch": 0.5911557556864103,
      "grad_norm": 0.050375433903915766,
      "learning_rate": 8.564275765881887e-05,
      "loss": 0.9156,
      "step": 1920
    },
    {
      "epoch": 0.592695223800177,
      "grad_norm": 0.04868715846487233,
      "learning_rate": 8.511096131435454e-05,
      "loss": 0.8904,
      "step": 1925
    },
    {
      "epoch": 0.5942346919139437,
      "grad_norm": 0.04810372148191546,
      "learning_rate": 8.457959524373109e-05,
      "loss": 0.9052,
      "step": 1930
    },
    {
      "epoch": 0.5957741600277104,
      "grad_norm": 0.04755974196810061,
      "learning_rate": 8.404867480273636e-05,
      "loss": 0.9271,
      "step": 1935
    },
    {
      "epoch": 0.5973136281414771,
      "grad_norm": 0.0478039967873912,
      "learning_rate": 8.351821533428023e-05,
      "loss": 0.9033,
      "step": 1940
    },
    {
      "epoch": 0.5988530962552439,
      "grad_norm": 0.04906160648695682,
      "learning_rate": 8.298823216795093e-05,
      "loss": 0.9237,
      "step": 1945
    },
    {
      "epoch": 0.6003925643690106,
      "grad_norm": 0.046540573735668354,
      "learning_rate": 8.245874061957224e-05,
      "loss": 0.9118,
      "step": 1950
    },
    {
      "epoch": 0.6019320324827772,
      "grad_norm": 0.046166297970054594,
      "learning_rate": 8.192975599076078e-05,
      "loss": 0.8924,
      "step": 1955
    },
    {
      "epoch": 0.6034715005965439,
      "grad_norm": 0.044047187769735666,
      "learning_rate": 8.140129356848387e-05,
      "loss": 0.8964,
      "step": 1960
    },
    {
      "epoch": 0.6050109687103106,
      "grad_norm": 0.04678512152897659,
      "learning_rate": 8.087336862461783e-05,
      "loss": 0.8889,
      "step": 1965
    },
    {
      "epoch": 0.6065504368240773,
      "grad_norm": 0.05171025733167019,
      "learning_rate": 8.034599641550642e-05,
      "loss": 0.9185,
      "step": 1970
    },
    {
      "epoch": 0.608089904937844,
      "grad_norm": 0.05197154775293005,
      "learning_rate": 7.981919218152016e-05,
      "loss": 0.8856,
      "step": 1975
    },
    {
      "epoch": 0.6096293730516107,
      "grad_norm": 0.04806811361587056,
      "learning_rate": 7.929297114661581e-05,
      "loss": 0.8966,
      "step": 1980
    },
    {
      "epoch": 0.6111688411653774,
      "grad_norm": 0.0483073053451646,
      "learning_rate": 7.876734851789643e-05,
      "loss": 0.921,
      "step": 1985
    },
    {
      "epoch": 0.6127083092791441,
      "grad_norm": 0.052489394135255815,
      "learning_rate": 7.824233948517185e-05,
      "loss": 0.9218,
      "step": 1990
    },
    {
      "epoch": 0.6142477773929107,
      "grad_norm": 0.04794477112095112,
      "learning_rate": 7.771795922051999e-05,
      "loss": 0.9177,
      "step": 1995
    },
    {
      "epoch": 0.6157872455066774,
      "grad_norm": 0.04956784535077608,
      "learning_rate": 7.719422287784798e-05,
      "loss": 0.8914,
      "step": 2000
    },
    {
      "epoch": 0.6173267136204441,
      "grad_norm": 0.04659852412787985,
      "learning_rate": 7.667114559245451e-05,
      "loss": 0.8891,
      "step": 2005
    },
    {
      "epoch": 0.6188661817342108,
      "grad_norm": 0.05397377178528671,
      "learning_rate": 7.614874248059238e-05,
      "loss": 0.8999,
      "step": 2010
    },
    {
      "epoch": 0.6204056498479775,
      "grad_norm": 0.04880480656639066,
      "learning_rate": 7.56270286390316e-05,
      "loss": 0.8914,
      "step": 2015
    },
    {
      "epoch": 0.6219451179617442,
      "grad_norm": 0.0595785459438196,
      "learning_rate": 7.510601914462331e-05,
      "loss": 0.8818,
      "step": 2020
    },
    {
      "epoch": 0.6234845860755109,
      "grad_norm": 0.048788834960320694,
      "learning_rate": 7.458572905386381e-05,
      "loss": 0.8838,
      "step": 2025
    },
    {
      "epoch": 0.6250240541892776,
      "grad_norm": 0.045705258650394785,
      "learning_rate": 7.406617340245957e-05,
      "loss": 0.9223,
      "step": 2030
    },
    {
      "epoch": 0.6265635223030444,
      "grad_norm": 0.04669796206524976,
      "learning_rate": 7.354736720489273e-05,
      "loss": 0.8891,
      "step": 2035
    },
    {
      "epoch": 0.628102990416811,
      "grad_norm": 0.04916213749618641,
      "learning_rate": 7.302932545398721e-05,
      "loss": 0.9074,
      "step": 2040
    },
    {
      "epoch": 0.6296424585305777,
      "grad_norm": 0.04569107557257168,
      "learning_rate": 7.251206312047547e-05,
      "loss": 0.8764,
      "step": 2045
    },
    {
      "epoch": 0.6311819266443444,
      "grad_norm": 0.04964260088103693,
      "learning_rate": 7.199559515256573e-05,
      "loss": 0.924,
      "step": 2050
    },
    {
      "epoch": 0.6327213947581111,
      "grad_norm": 0.048960093238618815,
      "learning_rate": 7.14799364755101e-05,
      "loss": 0.8949,
      "step": 2055
    },
    {
      "epoch": 0.6342608628718778,
      "grad_norm": 0.054026565865804656,
      "learning_rate": 7.096510199117327e-05,
      "loss": 0.9077,
      "step": 2060
    },
    {
      "epoch": 0.6358003309856445,
      "grad_norm": 0.05251895743669816,
      "learning_rate": 7.045110657760179e-05,
      "loss": 0.9142,
      "step": 2065
    },
    {
      "epoch": 0.6373397990994112,
      "grad_norm": 0.05001620423587143,
      "learning_rate": 6.993796508859418e-05,
      "loss": 0.8995,
      "step": 2070
    },
    {
      "epoch": 0.6388792672131779,
      "grad_norm": 0.050303627652658243,
      "learning_rate": 6.942569235327167e-05,
      "loss": 0.9019,
      "step": 2075
    },
    {
      "epoch": 0.6404187353269445,
      "grad_norm": 0.04941507039840511,
      "learning_rate": 6.891430317564964e-05,
      "loss": 0.9084,
      "step": 2080
    },
    {
      "epoch": 0.6419582034407112,
      "grad_norm": 0.04935121720371743,
      "learning_rate": 6.840381233420973e-05,
      "loss": 0.9054,
      "step": 2085
    },
    {
      "epoch": 0.6434976715544779,
      "grad_norm": 0.04517252434388429,
      "learning_rate": 6.789423458147292e-05,
      "loss": 0.8993,
      "step": 2090
    },
    {
      "epoch": 0.6450371396682446,
      "grad_norm": 0.05115514741866341,
      "learning_rate": 6.738558464357305e-05,
      "loss": 0.9209,
      "step": 2095
    },
    {
      "epoch": 0.6465766077820113,
      "grad_norm": 0.05140726619238067,
      "learning_rate": 6.687787721983136e-05,
      "loss": 0.9355,
      "step": 2100
    },
    {
      "epoch": 0.648116075895778,
      "grad_norm": 0.05038813028960742,
      "learning_rate": 6.63711269823317e-05,
      "loss": 0.8943,
      "step": 2105
    },
    {
      "epoch": 0.6496555440095447,
      "grad_norm": 0.055624576968490075,
      "learning_rate": 6.586534857549638e-05,
      "loss": 0.9085,
      "step": 2110
    },
    {
      "epoch": 0.6511950121233114,
      "grad_norm": 0.05129186851296076,
      "learning_rate": 6.536055661566312e-05,
      "loss": 0.927,
      "step": 2115
    },
    {
      "epoch": 0.6527344802370781,
      "grad_norm": 0.04555360600540732,
      "learning_rate": 6.485676569066258e-05,
      "loss": 0.9395,
      "step": 2120
    },
    {
      "epoch": 0.6542739483508447,
      "grad_norm": 0.05296379155269382,
      "learning_rate": 6.43539903593969e-05,
      "loss": 0.9081,
      "step": 2125
    },
    {
      "epoch": 0.6558134164646114,
      "grad_norm": 0.04919069940254436,
      "learning_rate": 6.385224515141879e-05,
      "loss": 0.8944,
      "step": 2130
    },
    {
      "epoch": 0.6573528845783781,
      "grad_norm": 0.0522053813927617,
      "learning_rate": 6.335154456651178e-05,
      "loss": 0.9116,
      "step": 2135
    },
    {
      "epoch": 0.6588923526921449,
      "grad_norm": 0.051576528357524944,
      "learning_rate": 6.285190307427114e-05,
      "loss": 0.935,
      "step": 2140
    },
    {
      "epoch": 0.6604318208059116,
      "grad_norm": 0.049706711499682615,
      "learning_rate": 6.235333511368573e-05,
      "loss": 0.9228,
      "step": 2145
    },
    {
      "epoch": 0.6619712889196783,
      "grad_norm": 0.05032192517838721,
      "learning_rate": 6.185585509272078e-05,
      "loss": 0.9146,
      "step": 2150
    },
    {
      "epoch": 0.663510757033445,
      "grad_norm": 0.052718683382770616,
      "learning_rate": 6.135947738790145e-05,
      "loss": 0.9233,
      "step": 2155
    },
    {
      "epoch": 0.6650502251472117,
      "grad_norm": 0.0493293346157755,
      "learning_rate": 6.0864216343897365e-05,
      "loss": 0.9053,
      "step": 2160
    },
    {
      "epoch": 0.6665896932609783,
      "grad_norm": 0.05315667999825589,
      "learning_rate": 6.0370086273108205e-05,
      "loss": 0.9295,
      "step": 2165
    },
    {
      "epoch": 0.668129161374745,
      "grad_norm": 0.047212072599722384,
      "learning_rate": 5.987710145524992e-05,
      "loss": 0.9253,
      "step": 2170
    },
    {
      "epoch": 0.6696686294885117,
      "grad_norm": 0.050669226462072614,
      "learning_rate": 5.938527613694214e-05,
      "loss": 0.9083,
      "step": 2175
    },
    {
      "epoch": 0.6712080976022784,
      "grad_norm": 0.05100220067507159,
      "learning_rate": 5.8894624531296486e-05,
      "loss": 0.8645,
      "step": 2180
    },
    {
      "epoch": 0.6727475657160451,
      "grad_norm": 0.051987668437823965,
      "learning_rate": 5.840516081750583e-05,
      "loss": 0.8996,
      "step": 2185
    },
    {
      "epoch": 0.6742870338298118,
      "grad_norm": 0.04753179189866846,
      "learning_rate": 5.791689914043447e-05,
      "loss": 0.9343,
      "step": 2190
    },
    {
      "epoch": 0.6758265019435785,
      "grad_norm": 0.0480193892017521,
      "learning_rate": 5.742985361020945e-05,
      "loss": 0.9033,
      "step": 2195
    },
    {
      "epoch": 0.6773659700573452,
      "grad_norm": 0.04740490974672915,
      "learning_rate": 5.69440383018127e-05,
      "loss": 0.9184,
      "step": 2200
    },
    {
      "epoch": 0.6789054381711119,
      "grad_norm": 0.05175060569501943,
      "learning_rate": 5.6459467254674435e-05,
      "loss": 0.8675,
      "step": 2205
    },
    {
      "epoch": 0.6804449062848785,
      "grad_norm": 0.05185903028595489,
      "learning_rate": 5.597615447226724e-05,
      "loss": 0.9068,
      "step": 2210
    },
    {
      "epoch": 0.6819843743986452,
      "grad_norm": 0.04719298233429231,
      "learning_rate": 5.549411392170154e-05,
      "loss": 0.8831,
      "step": 2215
    },
    {
      "epoch": 0.6835238425124119,
      "grad_norm": 0.05015798124154065,
      "learning_rate": 5.501335953332187e-05,
      "loss": 0.9059,
      "step": 2220
    },
    {
      "epoch": 0.6850633106261786,
      "grad_norm": 0.04871438606328049,
      "learning_rate": 5.453390520030439e-05,
      "loss": 0.9056,
      "step": 2225
    },
    {
      "epoch": 0.6866027787399454,
      "grad_norm": 0.050636785484572754,
      "learning_rate": 5.405576477825538e-05,
      "loss": 0.8878,
      "step": 2230
    },
    {
      "epoch": 0.6881422468537121,
      "grad_norm": 0.04750867999768763,
      "learning_rate": 5.3578952084810765e-05,
      "loss": 0.8914,
      "step": 2235
    },
    {
      "epoch": 0.6896817149674788,
      "grad_norm": 0.04779134258739018,
      "learning_rate": 5.310348089923681e-05,
      "loss": 0.9338,
      "step": 2240
    },
    {
      "epoch": 0.6912211830812455,
      "grad_norm": 0.04979302324195998,
      "learning_rate": 5.2629364962032004e-05,
      "loss": 0.9119,
      "step": 2245
    },
    {
      "epoch": 0.6927606511950122,
      "grad_norm": 0.05335301731834932,
      "learning_rate": 5.2156617974529886e-05,
      "loss": 0.914,
      "step": 2250
    },
    {
      "epoch": 0.6943001193087788,
      "grad_norm": 0.05265619261702106,
      "learning_rate": 5.1685253598503116e-05,
      "loss": 0.9205,
      "step": 2255
    },
    {
      "epoch": 0.6958395874225455,
      "grad_norm": 0.04725682617133063,
      "learning_rate": 5.1215285455768794e-05,
      "loss": 0.8847,
      "step": 2260
    },
    {
      "epoch": 0.6973790555363122,
      "grad_norm": 0.04942017634222197,
      "learning_rate": 5.074672712779456e-05,
      "loss": 0.8925,
      "step": 2265
    },
    {
      "epoch": 0.6989185236500789,
      "grad_norm": 0.04927308607421081,
      "learning_rate": 5.0279592155306286e-05,
      "loss": 0.8891,
      "step": 2270
    },
    {
      "epoch": 0.7004579917638456,
      "grad_norm": 0.04990850824904343,
      "learning_rate": 4.9813894037896747e-05,
      "loss": 0.893,
      "step": 2275
    },
    {
      "epoch": 0.7019974598776123,
      "grad_norm": 0.048947292463279043,
      "learning_rate": 4.93496462336354e-05,
      "loss": 0.9209,
      "step": 2280
    },
    {
      "epoch": 0.703536927991379,
      "grad_norm": 0.047946341032061454,
      "learning_rate": 4.8886862158679714e-05,
      "loss": 0.9199,
      "step": 2285
    },
    {
      "epoch": 0.7050763961051457,
      "grad_norm": 0.049190715621887184,
      "learning_rate": 4.8425555186887096e-05,
      "loss": 0.9036,
      "step": 2290
    },
    {
      "epoch": 0.7066158642189123,
      "grad_norm": 0.04856052193361857,
      "learning_rate": 4.796573864942868e-05,
      "loss": 0.9258,
      "step": 2295
    },
    {
      "epoch": 0.708155332332679,
      "grad_norm": 0.0530378049345599,
      "learning_rate": 4.750742583440397e-05,
      "loss": 0.8822,
      "step": 2300
    },
    {
      "epoch": 0.7096948004464457,
      "grad_norm": 0.048240884761695274,
      "learning_rate": 4.7050629986456873e-05,
      "loss": 0.8987,
      "step": 2305
    },
    {
      "epoch": 0.7112342685602124,
      "grad_norm": 0.050819290693906345,
      "learning_rate": 4.65953643063929e-05,
      "loss": 0.9141,
      "step": 2310
    },
    {
      "epoch": 0.7127737366739791,
      "grad_norm": 0.05000033360017204,
      "learning_rate": 4.6141641950797645e-05,
      "loss": 0.8723,
      "step": 2315
    },
    {
      "epoch": 0.7143132047877458,
      "grad_norm": 0.052399778768556976,
      "learning_rate": 4.5689476031656784e-05,
      "loss": 0.9292,
      "step": 2320
    },
    {
      "epoch": 0.7158526729015126,
      "grad_norm": 0.0455297122865129,
      "learning_rate": 4.523887961597688e-05,
      "loss": 0.9161,
      "step": 2325
    },
    {
      "epoch": 0.7173921410152793,
      "grad_norm": 0.049687782977331535,
      "learning_rate": 4.4789865725407934e-05,
      "loss": 0.9097,
      "step": 2330
    },
    {
      "epoch": 0.718931609129046,
      "grad_norm": 0.048946818451028745,
      "learning_rate": 4.434244733586699e-05,
      "loss": 0.8835,
      "step": 2335
    },
    {
      "epoch": 0.7204710772428126,
      "grad_norm": 0.04958169600360128,
      "learning_rate": 4.389663737716324e-05,
      "loss": 0.8982,
      "step": 2340
    },
    {
      "epoch": 0.7220105453565793,
      "grad_norm": 0.046259849431565525,
      "learning_rate": 4.3452448732624264e-05,
      "loss": 0.8846,
      "step": 2345
    },
    {
      "epoch": 0.723550013470346,
      "grad_norm": 0.052582557198390514,
      "learning_rate": 4.3009894238723856e-05,
      "loss": 0.885,
      "step": 2350
    },
    {
      "epoch": 0.7250894815841127,
      "grad_norm": 0.05260226314003436,
      "learning_rate": 4.256898668471092e-05,
      "loss": 0.9124,
      "step": 2355
    },
    {
      "epoch": 0.7266289496978794,
      "grad_norm": 0.049890083882035696,
      "learning_rate": 4.212973881223994e-05,
      "loss": 0.898,
      "step": 2360
    },
    {
      "epoch": 0.7281684178116461,
      "grad_norm": 0.05306403301581372,
      "learning_rate": 4.1692163315002784e-05,
      "loss": 0.9059,
      "step": 2365
    },
    {
      "epoch": 0.7297078859254128,
      "grad_norm": 0.048478281578994746,
      "learning_rate": 4.125627283836184e-05,
      "loss": 0.9035,
      "step": 2370
    },
    {
      "epoch": 0.7312473540391795,
      "grad_norm": 0.049739063696613614,
      "learning_rate": 4.082207997898457e-05,
      "loss": 0.8676,
      "step": 2375
    },
    {
      "epoch": 0.7327868221529461,
      "grad_norm": 0.049794405765263006,
      "learning_rate": 4.0389597284479595e-05,
      "loss": 0.8854,
      "step": 2380
    },
    {
      "epoch": 0.7343262902667128,
      "grad_norm": 0.052010748653662224,
      "learning_rate": 3.995883725303392e-05,
      "loss": 0.904,
      "step": 2385
    },
    {
      "epoch": 0.7358657583804795,
      "grad_norm": 0.05056533072375817,
      "learning_rate": 3.952981233305183e-05,
      "loss": 0.905,
      "step": 2390
    },
    {
      "epoch": 0.7374052264942462,
      "grad_norm": 0.04981878712032599,
      "learning_rate": 3.9102534922795166e-05,
      "loss": 0.8988,
      "step": 2395
    },
    {
      "epoch": 0.7389446946080129,
      "grad_norm": 0.049479950724968795,
      "learning_rate": 3.867701737002502e-05,
      "loss": 0.891,
      "step": 2400
    },
    {
      "epoch": 0.7404841627217796,
      "grad_norm": 0.049303364479271874,
      "learning_rate": 3.825327197164483e-05,
      "loss": 0.8792,
      "step": 2405
    },
    {
      "epoch": 0.7420236308355463,
      "grad_norm": 0.04976774531974654,
      "learning_rate": 3.7831310973345216e-05,
      "loss": 0.8929,
      "step": 2410
    },
    {
      "epoch": 0.743563098949313,
      "grad_norm": 0.04848591906961566,
      "learning_rate": 3.741114656924983e-05,
      "loss": 0.912,
      "step": 2415
    },
    {
      "epoch": 0.7451025670630798,
      "grad_norm": 0.05246275968531801,
      "learning_rate": 3.699279090156315e-05,
      "loss": 0.9162,
      "step": 2420
    },
    {
      "epoch": 0.7466420351768464,
      "grad_norm": 0.04784683370164075,
      "learning_rate": 3.6576256060219486e-05,
      "loss": 0.8821,
      "step": 2425
    },
    {
      "epoch": 0.7481815032906131,
      "grad_norm": 0.046538610019559695,
      "learning_rate": 3.616155408253367e-05,
      "loss": 0.8899,
      "step": 2430
    },
    {
      "epoch": 0.7497209714043798,
      "grad_norm": 0.04653797000431008,
      "learning_rate": 3.574869695285315e-05,
      "loss": 0.8906,
      "step": 2435
    },
    {
      "epoch": 0.7512604395181465,
      "grad_norm": 0.04854417696784218,
      "learning_rate": 3.5337696602211614e-05,
      "loss": 0.8736,
      "step": 2440
    },
    {
      "epoch": 0.7527999076319132,
      "grad_norm": 0.051959288038197514,
      "learning_rate": 3.492856490798439e-05,
      "loss": 0.9068,
      "step": 2445
    },
    {
      "epoch": 0.7543393757456799,
      "grad_norm": 0.05055945751383996,
      "learning_rate": 3.4521313693544966e-05,
      "loss": 0.9248,
      "step": 2450
    },
    {
      "epoch": 0.7558788438594466,
      "grad_norm": 0.04850985539708475,
      "learning_rate": 3.4115954727923395e-05,
      "loss": 0.9046,
      "step": 2455
    },
    {
      "epoch": 0.7574183119732133,
      "grad_norm": 0.048636140509240505,
      "learning_rate": 3.371249972546624e-05,
      "loss": 0.8935,
      "step": 2460
    },
    {
      "epoch": 0.7589577800869799,
      "grad_norm": 0.04693875087419482,
      "learning_rate": 3.3310960345497974e-05,
      "loss": 0.9293,
      "step": 2465
    },
    {
      "epoch": 0.7604972482007466,
      "grad_norm": 0.05068418571049861,
      "learning_rate": 3.291134819198417e-05,
      "loss": 0.9066,
      "step": 2470
    },
    {
      "epoch": 0.7620367163145133,
      "grad_norm": 0.049732111001641315,
      "learning_rate": 3.251367481319596e-05,
      "loss": 0.9084,
      "step": 2475
    },
    {
      "epoch": 0.76357618442828,
      "grad_norm": 0.04877480542908784,
      "learning_rate": 3.2117951701376436e-05,
      "loss": 0.885,
      "step": 2480
    },
    {
      "epoch": 0.7651156525420467,
      "grad_norm": 0.050155447150674604,
      "learning_rate": 3.172419029240853e-05,
      "loss": 0.9149,
      "step": 2485
    },
    {
      "epoch": 0.7666551206558134,
      "grad_norm": 0.05030249813110325,
      "learning_rate": 3.133240196548447e-05,
      "loss": 0.9008,
      "step": 2490
    },
    {
      "epoch": 0.7681945887695801,
      "grad_norm": 0.049906298979659446,
      "learning_rate": 3.0942598042777073e-05,
      "loss": 0.9373,
      "step": 2495
    },
    {
      "epoch": 0.7697340568833468,
      "grad_norm": 0.05100248290333388,
      "learning_rate": 3.0554789789112385e-05,
      "loss": 0.869,
      "step": 2500
    },
    {
      "epoch": 0.7712735249971135,
      "grad_norm": 0.05192202385416962,
      "learning_rate": 3.0168988411644205e-05,
      "loss": 0.8893,
      "step": 2505
    },
    {
      "epoch": 0.7728129931108801,
      "grad_norm": 0.04953362668159417,
      "learning_rate": 2.9785205059530263e-05,
      "loss": 0.9051,
      "step": 2510
    },
    {
      "epoch": 0.7743524612246468,
      "grad_norm": 0.04996622536563532,
      "learning_rate": 2.940345082360997e-05,
      "loss": 0.8986,
      "step": 2515
    },
    {
      "epoch": 0.7758919293384136,
      "grad_norm": 0.04836884568396388,
      "learning_rate": 2.9023736736083872e-05,
      "loss": 0.922,
      "step": 2520
    },
    {
      "epoch": 0.7774313974521803,
      "grad_norm": 0.050672635000901835,
      "learning_rate": 2.864607377019498e-05,
      "loss": 0.91,
      "step": 2525
    },
    {
      "epoch": 0.778970865565947,
      "grad_norm": 0.05373116940028323,
      "learning_rate": 2.82704728399115e-05,
      "loss": 0.9027,
      "step": 2530
    },
    {
      "epoch": 0.7805103336797137,
      "grad_norm": 0.05002900156096495,
      "learning_rate": 2.789694479961147e-05,
      "loss": 0.8951,
      "step": 2535
    },
    {
      "epoch": 0.7820498017934804,
      "grad_norm": 0.051641225857498176,
      "learning_rate": 2.7525500443769136e-05,
      "loss": 0.9025,
      "step": 2540
    },
    {
      "epoch": 0.7835892699072471,
      "grad_norm": 0.051090919816045334,
      "learning_rate": 2.715615050664294e-05,
      "loss": 0.9014,
      "step": 2545
    },
    {
      "epoch": 0.7851287380210138,
      "grad_norm": 0.04911341784983559,
      "learning_rate": 2.6788905661965458e-05,
      "loss": 0.8971,
      "step": 2550
    },
    {
      "epoch": 0.7866682061347804,
      "grad_norm": 0.049758629868399795,
      "learning_rate": 2.64237765226347e-05,
      "loss": 0.8912,
      "step": 2555
    },
    {
      "epoch": 0.7882076742485471,
      "grad_norm": 0.05110036066107192,
      "learning_rate": 2.606077364040762e-05,
      "loss": 0.891,
      "step": 2560
    },
    {
      "epoch": 0.7897471423623138,
      "grad_norm": 0.04918620440845424,
      "learning_rate": 2.5699907505595068e-05,
      "loss": 0.853,
      "step": 2565
    },
    {
      "epoch": 0.7912866104760805,
      "grad_norm": 0.04974068641972544,
      "learning_rate": 2.5341188546758688e-05,
      "loss": 0.8757,
      "step": 2570
    },
    {
      "epoch": 0.7928260785898472,
      "grad_norm": 0.0517642543172432,
      "learning_rate": 2.4984627130409577e-05,
      "loss": 0.8944,
      "step": 2575
    },
    {
      "epoch": 0.7943655467036139,
      "grad_norm": 0.051194030170498506,
      "learning_rate": 2.4630233560708615e-05,
      "loss": 0.8778,
      "step": 2580
    },
    {
      "epoch": 0.7959050148173806,
      "grad_norm": 0.048819792222408974,
      "learning_rate": 2.427801807916874e-05,
      "loss": 0.905,
      "step": 2585
    },
    {
      "epoch": 0.7974444829311473,
      "grad_norm": 0.05285747034364325,
      "learning_rate": 2.3927990864358984e-05,
      "loss": 0.8932,
      "step": 2590
    },
    {
      "epoch": 0.7989839510449139,
      "grad_norm": 0.05177880418117189,
      "learning_rate": 2.358016203161031e-05,
      "loss": 0.9105,
      "step": 2595
    },
    {
      "epoch": 0.8005234191586806,
      "grad_norm": 0.050125842765204626,
      "learning_rate": 2.3234541632723272e-05,
      "loss": 0.8956,
      "step": 2600
    },
    {
      "epoch": 0.8020628872724473,
      "grad_norm": 0.05005110838381569,
      "learning_rate": 2.2891139655677673e-05,
      "loss": 0.9097,
      "step": 2605
    },
    {
      "epoch": 0.803602355386214,
      "grad_norm": 0.04996589310236205,
      "learning_rate": 2.2549966024343682e-05,
      "loss": 0.8967,
      "step": 2610
    },
    {
      "epoch": 0.8051418234999808,
      "grad_norm": 0.05164799583829708,
      "learning_rate": 2.2211030598195247e-05,
      "loss": 0.9096,
      "step": 2615
    },
    {
      "epoch": 0.8066812916137475,
      "grad_norm": 0.04921635852838881,
      "learning_rate": 2.187434317202508e-05,
      "loss": 0.8928,
      "step": 2620
    },
    {
      "epoch": 0.8082207597275142,
      "grad_norm": 0.05100977521284688,
      "learning_rate": 2.1539913475661576e-05,
      "loss": 0.8868,
      "step": 2625
    },
    {
      "epoch": 0.8097602278412809,
      "grad_norm": 0.052267814845576244,
      "learning_rate": 2.1207751173687785e-05,
      "loss": 0.9449,
      "step": 2630
    },
    {
      "epoch": 0.8112996959550476,
      "grad_norm": 0.055390619856773446,
      "learning_rate": 2.0877865865161915e-05,
      "loss": 0.896,
      "step": 2635
    },
    {
      "epoch": 0.8128391640688142,
      "grad_norm": 0.0542325705617693,
      "learning_rate": 2.0550267083340068e-05,
      "loss": 0.8914,
      "step": 2640
    },
    {
      "epoch": 0.8143786321825809,
      "grad_norm": 0.055418743810201794,
      "learning_rate": 2.0224964295400682e-05,
      "loss": 0.8848,
      "step": 2645
    },
    {
      "epoch": 0.8159181002963476,
      "grad_norm": 0.05089971022933999,
      "learning_rate": 1.9901966902170944e-05,
      "loss": 0.8805,
      "step": 2650
    },
    {
      "epoch": 0.8174575684101143,
      "grad_norm": 0.050847263932470396,
      "learning_rate": 1.95812842378552e-05,
      "loss": 0.8937,
      "step": 2655
    },
    {
      "epoch": 0.818997036523881,
      "grad_norm": 0.05157053531068126,
      "learning_rate": 1.9262925569765087e-05,
      "loss": 0.9221,
      "step": 2660
    },
    {
      "epoch": 0.8205365046376477,
      "grad_norm": 0.0499397348100615,
      "learning_rate": 1.8946900098051778e-05,
      "loss": 0.9412,
      "step": 2665
    },
    {
      "epoch": 0.8220759727514144,
      "grad_norm": 0.05111992135361657,
      "learning_rate": 1.8633216955440137e-05,
      "loss": 0.9242,
      "step": 2670
    },
    {
      "epoch": 0.8236154408651811,
      "grad_norm": 0.05339912401311365,
      "learning_rate": 1.832188520696472e-05,
      "loss": 0.9097,
      "step": 2675
    },
    {
      "epoch": 0.8251549089789477,
      "grad_norm": 0.051772005101857047,
      "learning_rate": 1.8012913849707868e-05,
      "loss": 0.8916,
      "step": 2680
    },
    {
      "epoch": 0.8266943770927144,
      "grad_norm": 0.050106594186993446,
      "learning_rate": 1.7706311812539757e-05,
      "loss": 0.893,
      "step": 2685
    },
    {
      "epoch": 0.8282338452064811,
      "grad_norm": 0.047870354702240756,
      "learning_rate": 1.7402087955860193e-05,
      "loss": 0.9003,
      "step": 2690
    },
    {
      "epoch": 0.8297733133202478,
      "grad_norm": 0.05082697115153644,
      "learning_rate": 1.710025107134272e-05,
      "loss": 0.8903,
      "step": 2695
    },
    {
      "epoch": 0.8313127814340145,
      "grad_norm": 0.07972819984168905,
      "learning_rate": 1.680080988168049e-05,
      "loss": 0.8958,
      "step": 2700
    },
    {
      "epoch": 0.8328522495477813,
      "grad_norm": 0.04588401947701594,
      "learning_rate": 1.6503773040334126e-05,
      "loss": 0.8714,
      "step": 2705
    },
    {
      "epoch": 0.834391717661548,
      "grad_norm": 0.05193910001685307,
      "learning_rate": 1.620914913128184e-05,
      "loss": 0.9155,
      "step": 2710
    },
    {
      "epoch": 0.8359311857753147,
      "grad_norm": 0.04988441946676993,
      "learning_rate": 1.591694666877114e-05,
      "loss": 0.8812,
      "step": 2715
    },
    {
      "epoch": 0.8374706538890814,
      "grad_norm": 0.05238058487707531,
      "learning_rate": 1.5627174097072904e-05,
      "loss": 0.8731,
      "step": 2720
    },
    {
      "epoch": 0.839010122002848,
      "grad_norm": 0.05198017123832165,
      "learning_rate": 1.533983979023733e-05,
      "loss": 0.9285,
      "step": 2725
    },
    {
      "epoch": 0.8405495901166147,
      "grad_norm": 0.050178169987826704,
      "learning_rate": 1.5054952051851934e-05,
      "loss": 0.8987,
      "step": 2730
    },
    {
      "epoch": 0.8420890582303814,
      "grad_norm": 0.048234332694294745,
      "learning_rate": 1.477251911480162e-05,
      "loss": 0.9023,
      "step": 2735
    },
    {
      "epoch": 0.8436285263441481,
      "grad_norm": 0.05161636752265006,
      "learning_rate": 1.4492549141030687e-05,
      "loss": 0.9216,
      "step": 2740
    },
    {
      "epoch": 0.8451679944579148,
      "grad_norm": 0.05376436585199748,
      "learning_rate": 1.4215050221307002e-05,
      "loss": 0.8949,
      "step": 2745
    },
    {
      "epoch": 0.8467074625716815,
      "grad_norm": 0.049090105672334974,
      "learning_rate": 1.394003037498821e-05,
      "loss": 0.8991,
      "step": 2750
    },
    {
      "epoch": 0.8482469306854482,
      "grad_norm": 0.04938979818296241,
      "learning_rate": 1.3667497549789932e-05,
      "loss": 0.8902,
      "step": 2755
    },
    {
      "epoch": 0.8497863987992149,
      "grad_norm": 0.05110926192690603,
      "learning_rate": 1.339745962155613e-05,
      "loss": 0.905,
      "step": 2760
    },
    {
      "epoch": 0.8513258669129815,
      "grad_norm": 0.052370426487849654,
      "learning_rate": 1.3129924394031535e-05,
      "loss": 0.8813,
      "step": 2765
    },
    {
      "epoch": 0.8528653350267482,
      "grad_norm": 0.05580153374668592,
      "learning_rate": 1.2864899598636004e-05,
      "loss": 0.8941,
      "step": 2770
    },
    {
      "epoch": 0.8544048031405149,
      "grad_norm": 0.04874411693031306,
      "learning_rate": 1.2602392894241222e-05,
      "loss": 0.909,
      "step": 2775
    },
    {
      "epoch": 0.8559442712542816,
      "grad_norm": 0.051902977581340555,
      "learning_rate": 1.234241186694931e-05,
      "loss": 0.8861,
      "step": 2780
    },
    {
      "epoch": 0.8574837393680483,
      "grad_norm": 0.046190369417241875,
      "learning_rate": 1.208496402987358e-05,
      "loss": 0.901,
      "step": 2785
    },
    {
      "epoch": 0.859023207481815,
      "grad_norm": 0.049371890786887626,
      "learning_rate": 1.1830056822921521e-05,
      "loss": 0.8802,
      "step": 2790
    },
    {
      "epoch": 0.8605626755955817,
      "grad_norm": 0.05247959814051246,
      "learning_rate": 1.1577697612579641e-05,
      "loss": 0.9007,
      "step": 2795
    },
    {
      "epoch": 0.8621021437093485,
      "grad_norm": 0.048667585045453814,
      "learning_rate": 1.1327893691700698e-05,
      "loss": 0.8676,
      "step": 2800
    },
    {
      "epoch": 0.8636416118231152,
      "grad_norm": 0.050643478515145925,
      "learning_rate": 1.1080652279292891e-05,
      "loss": 0.8941,
      "step": 2805
    },
    {
      "epoch": 0.8651810799368818,
      "grad_norm": 0.0540254490096542,
      "learning_rate": 1.0835980520311251e-05,
      "loss": 0.8912,
      "step": 2810
    },
    {
      "epoch": 0.8667205480506485,
      "grad_norm": 0.04782094798408078,
      "learning_rate": 1.0593885485451237e-05,
      "loss": 0.8907,
      "step": 2815
    },
    {
      "epoch": 0.8682600161644152,
      "grad_norm": 0.053906572458732704,
      "learning_rate": 1.0354374170944258e-05,
      "loss": 0.9131,
      "step": 2820
    },
    {
      "epoch": 0.8697994842781819,
      "grad_norm": 0.05055406257947271,
      "learning_rate": 1.011745349835559e-05,
      "loss": 0.8901,
      "step": 2825
    },
    {
      "epoch": 0.8713389523919486,
      "grad_norm": 0.04921638188981594,
      "learning_rate": 9.883130314384348e-06,
      "loss": 0.8986,
      "step": 2830
    },
    {
      "epoch": 0.8728784205057153,
      "grad_norm": 0.04989204793182651,
      "learning_rate": 9.651411390665577e-06,
      "loss": 0.8741,
      "step": 2835
    },
    {
      "epoch": 0.874417888619482,
      "grad_norm": 0.048798752921894346,
      "learning_rate": 9.422303423574596e-06,
      "loss": 0.8919,
      "step": 2840
    },
    {
      "epoch": 0.8759573567332487,
      "grad_norm": 0.05284836395696706,
      "learning_rate": 9.195813034033508e-06,
      "loss": 0.8926,
      "step": 2845
    },
    {
      "epoch": 0.8774968248470154,
      "grad_norm": 0.05168276964659464,
      "learning_rate": 8.971946767319805e-06,
      "loss": 0.8794,
      "step": 2850
    },
    {
      "epoch": 0.879036292960782,
      "grad_norm": 0.05122404064787664,
      "learning_rate": 8.75071109287724e-06,
      "loss": 0.8831,
      "step": 2855
    },
    {
      "epoch": 0.8805757610745487,
      "grad_norm": 0.04747450340823261,
      "learning_rate": 8.532112404128877e-06,
      "loss": 0.8855,
      "step": 2860
    },
    {
      "epoch": 0.8821152291883154,
      "grad_norm": 0.04813099729864012,
      "learning_rate": 8.316157018292326e-06,
      "loss": 0.9194,
      "step": 2865
    },
    {
      "epoch": 0.8836546973020821,
      "grad_norm": 0.046778198065101924,
      "learning_rate": 8.102851176197201e-06,
      "loss": 0.8914,
      "step": 2870
    },
    {
      "epoch": 0.8851941654158488,
      "grad_norm": 0.0505389117652261,
      "learning_rate": 7.892201042104718e-06,
      "loss": 0.9076,
      "step": 2875
    },
    {
      "epoch": 0.8867336335296155,
      "grad_norm": 0.05155069773693662,
      "learning_rate": 7.684212703529624e-06,
      "loss": 0.9105,
      "step": 2880
    },
    {
      "epoch": 0.8882731016433822,
      "grad_norm": 0.047673629117401105,
      "learning_rate": 7.4788921710642e-06,
      "loss": 0.8957,
      "step": 2885
    },
    {
      "epoch": 0.889812569757149,
      "grad_norm": 0.05265601437978253,
      "learning_rate": 7.276245378204616e-06,
      "loss": 0.9044,
      "step": 2890
    },
    {
      "epoch": 0.8913520378709155,
      "grad_norm": 0.05163264368672346,
      "learning_rate": 7.076278181179485e-06,
      "loss": 0.9073,
      "step": 2895
    },
    {
      "epoch": 0.8928915059846823,
      "grad_norm": 0.05024106977313982,
      "learning_rate": 6.878996358780532e-06,
      "loss": 0.9091,
      "step": 2900
    },
    {
      "epoch": 0.894430974098449,
      "grad_norm": 0.04810053039047795,
      "learning_rate": 6.684405612195688e-06,
      "loss": 0.8896,
      "step": 2905
    },
    {
      "epoch": 0.8959704422122157,
      "grad_norm": 0.04830929316220151,
      "learning_rate": 6.492511564844273e-06,
      "loss": 0.8761,
      "step": 2910
    },
    {
      "epoch": 0.8975099103259824,
      "grad_norm": 0.05050911259620937,
      "learning_rate": 6.303319762214499e-06,
      "loss": 0.9073,
      "step": 2915
    },
    {
      "epoch": 0.8990493784397491,
      "grad_norm": 0.05061057110700013,
      "learning_rate": 6.11683567170328e-06,
      "loss": 0.9201,
      "step": 2920
    },
    {
      "epoch": 0.9005888465535158,
      "grad_norm": 0.05195265187043874,
      "learning_rate": 5.933064682458122e-06,
      "loss": 0.8862,
      "step": 2925
    },
    {
      "epoch": 0.9021283146672825,
      "grad_norm": 0.05146600570379482,
      "learning_rate": 5.7520121052214275e-06,
      "loss": 0.8822,
      "step": 2930
    },
    {
      "epoch": 0.9036677827810492,
      "grad_norm": 0.05230001510322826,
      "learning_rate": 5.57368317217708e-06,
      "loss": 0.9265,
      "step": 2935
    },
    {
      "epoch": 0.9052072508948158,
      "grad_norm": 0.051934624209648686,
      "learning_rate": 5.398083036799129e-06,
      "loss": 0.8534,
      "step": 2940
    },
    {
      "epoch": 0.9067467190085825,
      "grad_norm": 0.05044527476112878,
      "learning_rate": 5.225216773702968e-06,
      "loss": 0.9018,
      "step": 2945
    },
    {
      "epoch": 0.9082861871223492,
      "grad_norm": 0.05201869092317567,
      "learning_rate": 5.055089378498634e-06,
      "loss": 0.8926,
      "step": 2950
    },
    {
      "epoch": 0.9098256552361159,
      "grad_norm": 0.0475536468466368,
      "learning_rate": 4.887705767646434e-06,
      "loss": 0.8786,
      "step": 2955
    },
    {
      "epoch": 0.9113651233498826,
      "grad_norm": 0.05211001458842346,
      "learning_rate": 4.7230707783148864e-06,
      "loss": 0.885,
      "step": 2960
    },
    {
      "epoch": 0.9129045914636493,
      "grad_norm": 0.05131430855578592,
      "learning_rate": 4.561189168240909e-06,
      "loss": 0.9396,
      "step": 2965
    },
    {
      "epoch": 0.914444059577416,
      "grad_norm": 0.04842782742971858,
      "learning_rate": 4.402065615592344e-06,
      "loss": 0.878,
      "step": 2970
    },
    {
      "epoch": 0.9159835276911827,
      "grad_norm": 0.048784500669142396,
      "learning_rate": 4.245704718832811e-06,
      "loss": 0.9005,
      "step": 2975
    },
    {
      "epoch": 0.9175229958049493,
      "grad_norm": 0.045989168578578606,
      "learning_rate": 4.092110996588705e-06,
      "loss": 0.9104,
      "step": 2980
    },
    {
      "epoch": 0.919062463918716,
      "grad_norm": 0.04785800998672356,
      "learning_rate": 3.941288887518713e-06,
      "loss": 0.8858,
      "step": 2985
    },
    {
      "epoch": 0.9206019320324828,
      "grad_norm": 0.05196609915745631,
      "learning_rate": 3.7932427501854996e-06,
      "loss": 0.9067,
      "step": 2990
    },
    {
      "epoch": 0.9221414001462495,
      "grad_norm": 0.05048397127412001,
      "learning_rate": 3.647976862929747e-06,
      "loss": 0.9276,
      "step": 2995
    },
    {
      "epoch": 0.9236808682600162,
      "grad_norm": 0.04746994677124347,
      "learning_rate": 3.505495423746574e-06,
      "loss": 0.905,
      "step": 3000
    },
    {
      "epoch": 0.9252203363737829,
      "grad_norm": 0.04623520678007943,
      "learning_rate": 3.3658025501641323e-06,
      "loss": 0.924,
      "step": 3005
    },
    {
      "epoch": 0.9267598044875496,
      "grad_norm": 0.05915242583831865,
      "learning_rate": 3.228902279124657e-06,
      "loss": 0.9013,
      "step": 3010
    },
    {
      "epoch": 0.9282992726013163,
      "grad_norm": 0.05314450197725433,
      "learning_rate": 3.094798566867818e-06,
      "loss": 0.8891,
      "step": 3015
    },
    {
      "epoch": 0.929838740715083,
      "grad_norm": 0.04994737623939914,
      "learning_rate": 2.963495288816376e-06,
      "loss": 0.9393,
      "step": 3020
    },
    {
      "epoch": 0.9313782088288496,
      "grad_norm": 0.05558242936423758,
      "learning_rate": 2.8349962394641605e-06,
      "loss": 0.9367,
      "step": 3025
    },
    {
      "epoch": 0.9329176769426163,
      "grad_norm": 0.051634691252208646,
      "learning_rate": 2.709305132266493e-06,
      "loss": 0.9261,
      "step": 3030
    },
    {
      "epoch": 0.934457145056383,
      "grad_norm": 0.04954189779873653,
      "learning_rate": 2.5864255995327936e-06,
      "loss": 0.898,
      "step": 3035
    },
    {
      "epoch": 0.9359966131701497,
      "grad_norm": 0.050217713689466745,
      "learning_rate": 2.46636119232162e-06,
      "loss": 0.9051,
      "step": 3040
    },
    {
      "epoch": 0.9375360812839164,
      "grad_norm": 0.048010003704034514,
      "learning_rate": 2.349115380338096e-06,
      "loss": 0.9227,
      "step": 3045
    },
    {
      "epoch": 0.9390755493976831,
      "grad_norm": 0.050171020409517944,
      "learning_rate": 2.2346915518335786e-06,
      "loss": 0.8574,
      "step": 3050
    },
    {
      "epoch": 0.9406150175114498,
      "grad_norm": 0.04899312509832363,
      "learning_rate": 2.1230930135078373e-06,
      "loss": 0.8851,
      "step": 3055
    },
    {
      "epoch": 0.9421544856252165,
      "grad_norm": 0.049198496749408474,
      "learning_rate": 2.014322990413353e-06,
      "loss": 0.9028,
      "step": 3060
    },
    {
      "epoch": 0.9436939537389831,
      "grad_norm": 0.04937831083792284,
      "learning_rate": 1.9083846258622586e-06,
      "loss": 0.8901,
      "step": 3065
    },
    {
      "epoch": 0.9452334218527498,
      "grad_norm": 0.05060015921827992,
      "learning_rate": 1.8052809813354111e-06,
      "loss": 0.8851,
      "step": 3070
    },
    {
      "epoch": 0.9467728899665165,
      "grad_norm": 0.051074988313885475,
      "learning_rate": 1.705015036393931e-06,
      "loss": 0.9242,
      "step": 3075
    },
    {
      "epoch": 0.9483123580802832,
      "grad_norm": 0.051491649099328136,
      "learning_rate": 1.6075896885931807e-06,
      "loss": 0.8978,
      "step": 3080
    },
    {
      "epoch": 0.94985182619405,
      "grad_norm": 0.05031617030138609,
      "learning_rate": 1.5130077533988873e-06,
      "loss": 0.9118,
      "step": 3085
    },
    {
      "epoch": 0.9513912943078167,
      "grad_norm": 0.05073702344732129,
      "learning_rate": 1.421271964105908e-06,
      "loss": 0.9345,
      "step": 3090
    },
    {
      "epoch": 0.9529307624215834,
      "grad_norm": 0.05394201570580873,
      "learning_rate": 1.3323849717591376e-06,
      "loss": 0.8932,
      "step": 3095
    },
    {
      "epoch": 0.9544702305353501,
      "grad_norm": 0.05288876825736752,
      "learning_rate": 1.2463493450769915e-06,
      "loss": 0.9176,
      "step": 3100
    },
    {
      "epoch": 0.9560096986491168,
      "grad_norm": 0.052706852259209895,
      "learning_rate": 1.1631675703771105e-06,
      "loss": 0.9136,
      "step": 3105
    },
    {
      "epoch": 0.9575491667628834,
      "grad_norm": 0.050101686356315255,
      "learning_rate": 1.0828420515045178e-06,
      "loss": 0.8997,
      "step": 3110
    },
    {
      "epoch": 0.9590886348766501,
      "grad_norm": 0.046158354718340906,
      "learning_rate": 1.0053751097621856e-06,
      "loss": 0.8741,
      "step": 3115
    },
    {
      "epoch": 0.9606281029904168,
      "grad_norm": 0.04793073226919364,
      "learning_rate": 9.307689838439104e-07,
      "loss": 0.9213,
      "step": 3120
    },
    {
      "epoch": 0.9621675711041835,
      "grad_norm": 0.0480250612155145,
      "learning_rate": 8.590258297696108e-07,
      "loss": 0.8592,
      "step": 3125
    },
    {
      "epoch": 0.9637070392179502,
      "grad_norm": 0.05079308974694697,
      "learning_rate": 7.901477208230979e-07,
      "loss": 0.8949,
      "step": 3130
    },
    {
      "epoch": 0.9652465073317169,
      "grad_norm": 0.049530271547413275,
      "learning_rate": 7.241366474920797e-07,
      "loss": 0.8813,
      "step": 3135
    },
    {
      "epoch": 0.9667859754454836,
      "grad_norm": 0.04804929713391988,
      "learning_rate": 6.609945174106402e-07,
      "loss": 0.9129,
      "step": 3140
    },
    {
      "epoch": 0.9683254435592503,
      "grad_norm": 0.051257812589911356,
      "learning_rate": 6.007231553041837e-07,
      "loss": 0.8839,
      "step": 3145
    },
    {
      "epoch": 0.969864911673017,
      "grad_norm": 0.04955517862562011,
      "learning_rate": 5.43324302936643e-07,
      "loss": 0.9571,
      "step": 3150
    },
    {
      "epoch": 0.9714043797867836,
      "grad_norm": 0.0491228255053995,
      "learning_rate": 4.887996190601318e-07,
      "loss": 0.9139,
      "step": 3155
    },
    {
      "epoch": 0.9729438479005503,
      "grad_norm": 0.050192881863201706,
      "learning_rate": 4.3715067936705987e-07,
      "loss": 0.8927,
      "step": 3160
    },
    {
      "epoch": 0.974483316014317,
      "grad_norm": 0.050063559948920915,
      "learning_rate": 3.8837897644457e-07,
      "loss": 0.9188,
      "step": 3165
    },
    {
      "epoch": 0.9760227841280837,
      "grad_norm": 0.047101809428331816,
      "learning_rate": 3.4248591973140566e-07,
      "loss": 0.9134,
      "step": 3170
    },
    {
      "epoch": 0.9775622522418504,
      "grad_norm": 0.05350592485136906,
      "learning_rate": 2.994728354771659e-07,
      "loss": 0.9353,
      "step": 3175
    },
    {
      "epoch": 0.9791017203556172,
      "grad_norm": 0.04855844715305267,
      "learning_rate": 2.593409667040247e-07,
      "loss": 0.8648,
      "step": 3180
    },
    {
      "epoch": 0.9806411884693839,
      "grad_norm": 0.048812751788986294,
      "learning_rate": 2.2209147317074908e-07,
      "loss": 0.8787,
      "step": 3185
    },
    {
      "epoch": 0.9821806565831506,
      "grad_norm": 0.051356351052386,
      "learning_rate": 1.8772543133922515e-07,
      "loss": 0.9042,
      "step": 3190
    },
    {
      "epoch": 0.9837201246969172,
      "grad_norm": 0.047884395981010776,
      "learning_rate": 1.5624383434333923e-07,
      "loss": 0.8822,
      "step": 3195
    },
    {
      "epoch": 0.9852595928106839,
      "grad_norm": 0.05094435499033899,
      "learning_rate": 1.276475919602671e-07,
      "loss": 0.8962,
      "step": 3200
    },
    {
      "epoch": 0.9867990609244506,
      "grad_norm": 0.05551428987367575,
      "learning_rate": 1.019375305842063e-07,
      "loss": 0.887,
      "step": 3205
    },
    {
      "epoch": 0.9883385290382173,
      "grad_norm": 0.04546637336940152,
      "learning_rate": 7.911439320247294e-08,
      "loss": 0.888,
      "step": 3210
    },
    {
      "epoch": 0.989877997151984,
      "grad_norm": 0.051725181634757214,
      "learning_rate": 5.9178839374018914e-08,
      "loss": 0.9079,
      "step": 3215
    },
    {
      "epoch": 0.9914174652657507,
      "grad_norm": 0.054748326673771164,
      "learning_rate": 4.213144521042489e-08,
      "loss": 0.9068,
      "step": 3220
    },
    {
      "epoch": 0.9929569333795174,
      "grad_norm": 0.05377608385855954,
      "learning_rate": 2.797270335916924e-08,
      "loss": 0.8806,
      "step": 3225
    },
    {
      "epoch": 0.9944964014932841,
      "grad_norm": 0.04798606701698547,
      "learning_rate": 1.6703022989494887e-08,
      "loss": 0.9099,
      "step": 3230
    },
    {
      "epoch": 0.9960358696070508,
      "grad_norm": 0.0472421438494704,
      "learning_rate": 8.322729780474436e-09,
      "loss": 0.8932,
      "step": 3235
    },
    {
      "epoch": 0.9975753377208174,
      "grad_norm": 0.048793466284065876,
      "learning_rate": 2.8320659116953806e-09,
      "loss": 0.8917,
      "step": 3240
    },
    {
      "epoch": 0.9991148058345841,
      "grad_norm": 0.05220366875224664,
      "learning_rate": 2.311900561768887e-10,
      "loss": 0.8864,
      "step": 3245
    },
    {
      "epoch": 0.9997305930800908,
      "eval_loss": 0.8975253105163574,
      "eval_runtime": 1716.3616,
      "eval_samples_per_second": 5.826,
      "eval_steps_per_second": 0.364,
      "step": 3247
    },
    {
      "epoch": 0.9997305930800908,
      "step": 3247,
      "total_flos": 5.273312589722419e+16,
      "train_loss": 0.9348164240617108,
      "train_runtime": 152106.1857,
      "train_samples_per_second": 1.367,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 5,
  "max_steps": 3247,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 5.273312589722419e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}