{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9999069680900549,
  "eval_steps": 100,
  "global_step": 2687,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018606381989022234,
      "grad_norm": 0.025198739022016525,
      "learning_rate": 3.7174721189591085e-07,
      "loss": 0.7637,
      "step": 5
    },
    {
      "epoch": 0.003721276397804447,
      "grad_norm": 0.017377199605107307,
      "learning_rate": 7.434944237918217e-07,
      "loss": 0.7786,
      "step": 10
    },
    {
      "epoch": 0.0055819145967066705,
      "grad_norm": 0.019743537530303,
      "learning_rate": 1.1152416356877324e-06,
      "loss": 0.7174,
      "step": 15
    },
    {
      "epoch": 0.007442552795608894,
      "grad_norm": 0.017805561423301697,
      "learning_rate": 1.4869888475836434e-06,
      "loss": 0.724,
      "step": 20
    },
    {
      "epoch": 0.009303190994511117,
      "grad_norm": 0.017246991395950317,
      "learning_rate": 1.858736059479554e-06,
      "loss": 0.7668,
      "step": 25
    },
    {
      "epoch": 0.011163829193413341,
      "grad_norm": 0.023152414709329605,
      "learning_rate": 2.2304832713754648e-06,
      "loss": 0.781,
      "step": 30
    },
    {
      "epoch": 0.013024467392315565,
      "grad_norm": 0.019485710188746452,
      "learning_rate": 2.6022304832713758e-06,
      "loss": 0.7593,
      "step": 35
    },
    {
      "epoch": 0.014885105591217787,
      "grad_norm": 0.01784881390631199,
      "learning_rate": 2.973977695167287e-06,
      "loss": 0.762,
      "step": 40
    },
    {
      "epoch": 0.01674574379012001,
      "grad_norm": 0.01972021535038948,
      "learning_rate": 3.3457249070631974e-06,
      "loss": 0.7489,
      "step": 45
    },
    {
      "epoch": 0.018606381989022234,
      "grad_norm": 0.019333072006702423,
      "learning_rate": 3.717472118959108e-06,
      "loss": 0.7461,
      "step": 50
    },
    {
      "epoch": 0.02046702018792446,
      "grad_norm": 0.015675414353609085,
      "learning_rate": 4.089219330855019e-06,
      "loss": 0.7621,
      "step": 55
    },
    {
      "epoch": 0.022327658386826682,
      "grad_norm": 0.021361010149121284,
      "learning_rate": 4.4609665427509296e-06,
      "loss": 0.7677,
      "step": 60
    },
    {
      "epoch": 0.024188296585728904,
      "grad_norm": 0.021466689184308052,
      "learning_rate": 4.83271375464684e-06,
      "loss": 0.7857,
      "step": 65
    },
    {
      "epoch": 0.02604893478463113,
      "grad_norm": 0.019599348306655884,
      "learning_rate": 5.2044609665427516e-06,
      "loss": 0.7308,
      "step": 70
    },
    {
      "epoch": 0.027909572983533353,
      "grad_norm": 0.015855278819799423,
      "learning_rate": 5.576208178438662e-06,
      "loss": 0.7423,
      "step": 75
    },
    {
      "epoch": 0.029770211182435575,
      "grad_norm": 0.019443219527602196,
      "learning_rate": 5.947955390334574e-06,
      "loss": 0.7782,
      "step": 80
    },
    {
      "epoch": 0.0316308493813378,
      "grad_norm": 0.01575319655239582,
      "learning_rate": 6.319702602230484e-06,
      "loss": 0.7494,
      "step": 85
    },
    {
      "epoch": 0.03349148758024002,
      "grad_norm": 0.0159482192248106,
      "learning_rate": 6.691449814126395e-06,
      "loss": 0.6999,
      "step": 90
    },
    {
      "epoch": 0.03535212577914225,
      "grad_norm": 0.016392122954130173,
      "learning_rate": 7.063197026022306e-06,
      "loss": 0.6966,
      "step": 95
    },
    {
      "epoch": 0.03721276397804447,
      "grad_norm": 0.017619600519537926,
      "learning_rate": 7.434944237918216e-06,
      "loss": 0.7659,
      "step": 100
    },
    {
      "epoch": 0.03721276397804447,
      "eval_loss": 0.760661780834198,
      "eval_runtime": 26.8914,
      "eval_samples_per_second": 4.723,
      "eval_steps_per_second": 4.723,
      "step": 100
    },
    {
      "epoch": 0.039073402176946694,
      "grad_norm": 0.02020377479493618,
      "learning_rate": 7.806691449814127e-06,
      "loss": 0.7475,
      "step": 105
    },
    {
      "epoch": 0.04093404037584892,
      "grad_norm": 0.018723690882325172,
      "learning_rate": 8.178438661710038e-06,
      "loss": 0.7233,
      "step": 110
    },
    {
      "epoch": 0.04279467857475114,
      "grad_norm": 0.017194446176290512,
      "learning_rate": 8.550185873605949e-06,
      "loss": 0.7655,
      "step": 115
    },
    {
      "epoch": 0.044655316773653364,
      "grad_norm": 0.013910962268710136,
      "learning_rate": 8.921933085501859e-06,
      "loss": 0.7493,
      "step": 120
    },
    {
      "epoch": 0.04651595497255559,
      "grad_norm": 0.01600288413465023,
      "learning_rate": 9.29368029739777e-06,
      "loss": 0.7528,
      "step": 125
    },
    {
      "epoch": 0.04837659317145781,
      "grad_norm": 0.01554879080504179,
      "learning_rate": 9.66542750929368e-06,
      "loss": 0.6947,
      "step": 130
    },
    {
      "epoch": 0.050237231370360035,
      "grad_norm": 0.015109645202755928,
      "learning_rate": 1.0037174721189591e-05,
      "loss": 0.6647,
      "step": 135
    },
    {
      "epoch": 0.05209786956926226,
      "grad_norm": 0.015620779246091843,
      "learning_rate": 1.0408921933085503e-05,
      "loss": 0.7298,
      "step": 140
    },
    {
      "epoch": 0.05395850776816448,
      "grad_norm": 0.013970567844808102,
      "learning_rate": 1.0780669144981412e-05,
      "loss": 0.7485,
      "step": 145
    },
    {
      "epoch": 0.055819145967066705,
      "grad_norm": 0.014776401221752167,
      "learning_rate": 1.1152416356877324e-05,
      "loss": 0.6767,
      "step": 150
    },
    {
      "epoch": 0.057679784165968924,
      "grad_norm": 0.014368101954460144,
      "learning_rate": 1.1524163568773235e-05,
      "loss": 0.7037,
      "step": 155
    },
    {
      "epoch": 0.05954042236487115,
      "grad_norm": 0.01449244562536478,
      "learning_rate": 1.1895910780669147e-05,
      "loss": 0.7305,
      "step": 160
    },
    {
      "epoch": 0.061401060563773376,
      "grad_norm": 0.015350298024713993,
      "learning_rate": 1.2267657992565056e-05,
      "loss": 0.6804,
      "step": 165
    },
    {
      "epoch": 0.0632616987626756,
      "grad_norm": 0.014395203441381454,
      "learning_rate": 1.2639405204460968e-05,
      "loss": 0.6832,
      "step": 170
    },
    {
      "epoch": 0.06512233696157782,
      "grad_norm": 0.015541068278253078,
      "learning_rate": 1.3011152416356879e-05,
      "loss": 0.7168,
      "step": 175
    },
    {
      "epoch": 0.06698297516048005,
      "grad_norm": 0.012579885311424732,
      "learning_rate": 1.338289962825279e-05,
      "loss": 0.693,
      "step": 180
    },
    {
      "epoch": 0.06884361335938227,
      "grad_norm": 0.014168789610266685,
      "learning_rate": 1.37546468401487e-05,
      "loss": 0.7056,
      "step": 185
    },
    {
      "epoch": 0.0707042515582845,
      "grad_norm": 0.012087655253708363,
      "learning_rate": 1.4126394052044612e-05,
      "loss": 0.7006,
      "step": 190
    },
    {
      "epoch": 0.07256488975718671,
      "grad_norm": 0.014433121308684349,
      "learning_rate": 1.4498141263940521e-05,
      "loss": 0.7016,
      "step": 195
    },
    {
      "epoch": 0.07442552795608894,
      "grad_norm": 0.015074139460921288,
      "learning_rate": 1.4869888475836432e-05,
      "loss": 0.6912,
      "step": 200
    },
    {
      "epoch": 0.07442552795608894,
      "eval_loss": 0.7027233242988586,
      "eval_runtime": 26.5611,
      "eval_samples_per_second": 4.781,
      "eval_steps_per_second": 4.781,
      "step": 200
    },
    {
      "epoch": 0.07628616615499116,
      "grad_norm": 0.013617471791803837,
      "learning_rate": 1.5241635687732344e-05,
      "loss": 0.7139,
      "step": 205
    },
    {
      "epoch": 0.07814680435389339,
      "grad_norm": 0.013274065218865871,
      "learning_rate": 1.5613382899628255e-05,
      "loss": 0.6955,
      "step": 210
    },
    {
      "epoch": 0.08000744255279561,
      "grad_norm": 0.013781987130641937,
      "learning_rate": 1.5985130111524165e-05,
      "loss": 0.65,
      "step": 215
    },
    {
      "epoch": 0.08186808075169784,
      "grad_norm": 0.01373015996068716,
      "learning_rate": 1.6356877323420076e-05,
      "loss": 0.6681,
      "step": 220
    },
    {
      "epoch": 0.08372871895060005,
      "grad_norm": 0.01403126772493124,
      "learning_rate": 1.6728624535315986e-05,
      "loss": 0.6981,
      "step": 225
    },
    {
      "epoch": 0.08558935714950228,
      "grad_norm": 0.01332685723900795,
      "learning_rate": 1.7100371747211897e-05,
      "loss": 0.6517,
      "step": 230
    },
    {
      "epoch": 0.0874499953484045,
      "grad_norm": 0.013317782431840897,
      "learning_rate": 1.7472118959107808e-05,
      "loss": 0.6835,
      "step": 235
    },
    {
      "epoch": 0.08931063354730673,
      "grad_norm": 0.015096917748451233,
      "learning_rate": 1.7843866171003718e-05,
      "loss": 0.7138,
      "step": 240
    },
    {
      "epoch": 0.09117127174620895,
      "grad_norm": 0.013502717949450016,
      "learning_rate": 1.8215613382899632e-05,
      "loss": 0.6722,
      "step": 245
    },
    {
      "epoch": 0.09303190994511118,
      "grad_norm": 0.01279054582118988,
      "learning_rate": 1.858736059479554e-05,
      "loss": 0.6323,
      "step": 250
    },
    {
      "epoch": 0.09489254814401339,
      "grad_norm": 0.011777005158364773,
      "learning_rate": 1.8959107806691453e-05,
      "loss": 0.6332,
      "step": 255
    },
    {
      "epoch": 0.09675318634291562,
      "grad_norm": 0.01320689544081688,
      "learning_rate": 1.933085501858736e-05,
      "loss": 0.7034,
      "step": 260
    },
    {
      "epoch": 0.09861382454181784,
      "grad_norm": 0.012140162289142609,
      "learning_rate": 1.970260223048327e-05,
      "loss": 0.6154,
      "step": 265
    },
    {
      "epoch": 0.10047446274072007,
      "grad_norm": 0.01772845722734928,
      "learning_rate": 1.9999991559715313e-05,
      "loss": 0.7119,
      "step": 270
    },
    {
      "epoch": 0.1023351009396223,
      "grad_norm": 0.012948377057909966,
      "learning_rate": 1.999969615124717e-05,
      "loss": 0.6554,
      "step": 275
    },
    {
      "epoch": 0.10419573913852452,
      "grad_norm": 0.013341420330107212,
      "learning_rate": 1.9998978742792098e-05,
      "loss": 0.636,
      "step": 280
    },
    {
      "epoch": 0.10605637733742673,
      "grad_norm": 0.012272336520254612,
      "learning_rate": 1.999783936462566e-05,
      "loss": 0.6182,
      "step": 285
    },
    {
      "epoch": 0.10791701553632896,
      "grad_norm": 0.012329615652561188,
      "learning_rate": 1.999627806483107e-05,
      "loss": 0.601,
      "step": 290
    },
    {
      "epoch": 0.10977765373523118,
      "grad_norm": 0.01313008088618517,
      "learning_rate": 1.999429490929718e-05,
      "loss": 0.7002,
      "step": 295
    },
    {
      "epoch": 0.11163829193413341,
      "grad_norm": 0.01170238945633173,
      "learning_rate": 1.9991889981715696e-05,
      "loss": 0.6784,
      "step": 300
    },
    {
      "epoch": 0.11163829193413341,
      "eval_loss": 0.676296055316925,
      "eval_runtime": 26.7211,
      "eval_samples_per_second": 4.753,
      "eval_steps_per_second": 4.753,
      "step": 300
    },
    {
      "epoch": 0.11349893013303564,
      "grad_norm": 0.012631393037736416,
      "learning_rate": 1.9989063383577644e-05,
      "loss": 0.6332,
      "step": 305
    },
    {
      "epoch": 0.11535956833193785,
      "grad_norm": 0.012392008677124977,
      "learning_rate": 1.998581523416908e-05,
      "loss": 0.713,
      "step": 310
    },
    {
      "epoch": 0.11722020653084007,
      "grad_norm": 0.012342042289674282,
      "learning_rate": 1.998214567056607e-05,
      "loss": 0.6072,
      "step": 315
    },
    {
      "epoch": 0.1190808447297423,
      "grad_norm": 0.011801215820014477,
      "learning_rate": 1.9978054847628908e-05,
      "loss": 0.6553,
      "step": 320
    },
    {
      "epoch": 0.12094148292864453,
      "grad_norm": 0.013543626293540001,
      "learning_rate": 1.997354293799555e-05,
      "loss": 0.638,
      "step": 325
    },
    {
      "epoch": 0.12280212112754675,
      "grad_norm": 0.012367943301796913,
      "learning_rate": 1.9968610132074372e-05,
      "loss": 0.6867,
      "step": 330
    },
    {
      "epoch": 0.12466275932644898,
      "grad_norm": 0.01382706593722105,
      "learning_rate": 1.99632566380361e-05,
      "loss": 0.67,
      "step": 335
    },
    {
      "epoch": 0.1265233975253512,
      "grad_norm": 0.01336714904755354,
      "learning_rate": 1.9957482681805036e-05,
      "loss": 0.6615,
      "step": 340
    },
    {
      "epoch": 0.12838403572425341,
      "grad_norm": 0.012786686420440674,
      "learning_rate": 1.9951288507049532e-05,
      "loss": 0.6343,
      "step": 345
    },
    {
      "epoch": 0.13024467392315564,
      "grad_norm": 0.014750728383660316,
      "learning_rate": 1.9944674375171697e-05,
      "loss": 0.6478,
      "step": 350
    },
    {
      "epoch": 0.13210531212205787,
      "grad_norm": 0.01330367662012577,
      "learning_rate": 1.9937640565296372e-05,
      "loss": 0.6844,
      "step": 355
    },
    {
      "epoch": 0.1339659503209601,
      "grad_norm": 0.011332959868013859,
      "learning_rate": 1.9930187374259338e-05,
      "loss": 0.6188,
      "step": 360
    },
    {
      "epoch": 0.13582658851986232,
      "grad_norm": 0.013507647439837456,
      "learning_rate": 1.992231511659481e-05,
      "loss": 0.6844,
      "step": 365
    },
    {
      "epoch": 0.13768722671876454,
      "grad_norm": 0.01494019664824009,
      "learning_rate": 1.991402412452214e-05,
      "loss": 0.6616,
      "step": 370
    },
    {
      "epoch": 0.13954786491766677,
      "grad_norm": 0.012762832455337048,
      "learning_rate": 1.9905314747931816e-05,
      "loss": 0.6797,
      "step": 375
    },
    {
      "epoch": 0.141408503116569,
      "grad_norm": 0.014913683757185936,
      "learning_rate": 1.989618735437069e-05,
      "loss": 0.6268,
      "step": 380
    },
    {
      "epoch": 0.1432691413154712,
      "grad_norm": 0.012221734039485455,
      "learning_rate": 1.9886642329026457e-05,
      "loss": 0.6587,
      "step": 385
    },
    {
      "epoch": 0.14512977951437342,
      "grad_norm": 0.01553855836391449,
      "learning_rate": 1.9876680074711417e-05,
      "loss": 0.6403,
      "step": 390
    },
    {
      "epoch": 0.14699041771327565,
      "grad_norm": 0.013682518154382706,
      "learning_rate": 1.986630101184546e-05,
      "loss": 0.6287,
      "step": 395
    },
    {
      "epoch": 0.14885105591217787,
      "grad_norm": 0.013481502421200275,
      "learning_rate": 1.9855505578438343e-05,
      "loss": 0.6757,
      "step": 400
    },
    {
      "epoch": 0.14885105591217787,
      "eval_loss": 0.6647208333015442,
      "eval_runtime": 27.129,
      "eval_samples_per_second": 4.681,
      "eval_steps_per_second": 4.681,
      "step": 400
    },
    {
      "epoch": 0.1507116941110801,
      "grad_norm": 0.01228385604918003,
      "learning_rate": 1.984429423007117e-05,
      "loss": 0.6277,
      "step": 405
    },
    {
      "epoch": 0.15257233230998232,
      "grad_norm": 0.014119746163487434,
      "learning_rate": 1.9832667439877217e-05,
      "loss": 0.615,
      "step": 410
    },
    {
      "epoch": 0.15443297050888455,
      "grad_norm": 0.014395875856280327,
      "learning_rate": 1.9820625698521918e-05,
      "loss": 0.6417,
      "step": 415
    },
    {
      "epoch": 0.15629360870778677,
      "grad_norm": 0.013175971806049347,
      "learning_rate": 1.9808169514182182e-05,
      "loss": 0.6509,
      "step": 420
    },
    {
      "epoch": 0.158154246906689,
      "grad_norm": 0.015295376069843769,
      "learning_rate": 1.9795299412524948e-05,
      "loss": 0.6275,
      "step": 425
    },
    {
      "epoch": 0.16001488510559123,
      "grad_norm": 0.014611025340855122,
      "learning_rate": 1.9782015936684987e-05,
      "loss": 0.6627,
      "step": 430
    },
    {
      "epoch": 0.16187552330449345,
      "grad_norm": 0.01412207167595625,
      "learning_rate": 1.9768319647242e-05,
      "loss": 0.6362,
      "step": 435
    },
    {
      "epoch": 0.16373616150339568,
      "grad_norm": 0.012070410884916782,
      "learning_rate": 1.9754211122196945e-05,
      "loss": 0.6429,
      "step": 440
    },
    {
      "epoch": 0.16559679970229788,
      "grad_norm": 0.013232079334557056,
      "learning_rate": 1.9739690956947652e-05,
      "loss": 0.6941,
      "step": 445
    },
    {
      "epoch": 0.1674574379012001,
      "grad_norm": 0.012606708332896233,
      "learning_rate": 1.972475976426369e-05,
      "loss": 0.6554,
      "step": 450
    },
    {
      "epoch": 0.16931807610010233,
      "grad_norm": 0.012638423591852188,
      "learning_rate": 1.9709418174260523e-05,
      "loss": 0.645,
      "step": 455
    },
    {
      "epoch": 0.17117871429900455,
      "grad_norm": 0.013119902461767197,
      "learning_rate": 1.9693666834372896e-05,
      "loss": 0.6128,
      "step": 460
    },
    {
      "epoch": 0.17303935249790678,
      "grad_norm": 0.011363113299012184,
      "learning_rate": 1.9677506409327532e-05,
      "loss": 0.6294,
      "step": 465
    },
    {
      "epoch": 0.174899990696809,
      "grad_norm": 0.014238959178328514,
      "learning_rate": 1.9660937581115073e-05,
      "loss": 0.6647,
      "step": 470
    },
    {
      "epoch": 0.17676062889571123,
      "grad_norm": 0.013214629143476486,
      "learning_rate": 1.9643961048961283e-05,
      "loss": 0.6037,
      "step": 475
    },
    {
      "epoch": 0.17862126709461346,
      "grad_norm": 0.012312485836446285,
      "learning_rate": 1.9626577529297573e-05,
      "loss": 0.6703,
      "step": 480
    },
    {
      "epoch": 0.18048190529351568,
      "grad_norm": 0.012318914756178856,
      "learning_rate": 1.9608787755730746e-05,
      "loss": 0.6141,
      "step": 485
    },
    {
      "epoch": 0.1823425434924179,
      "grad_norm": 0.01374764647334814,
      "learning_rate": 1.9590592479012022e-05,
      "loss": 0.673,
      "step": 490
    },
    {
      "epoch": 0.18420318169132013,
      "grad_norm": 0.012634415179491043,
      "learning_rate": 1.9571992467005395e-05,
      "loss": 0.6135,
      "step": 495
    },
    {
      "epoch": 0.18606381989022236,
      "grad_norm": 0.012813772074878216,
      "learning_rate": 1.9552988504655194e-05,
      "loss": 0.6648,
      "step": 500
    },
    {
      "epoch": 0.18606381989022236,
      "eval_loss": 0.6582558751106262,
      "eval_runtime": 26.9621,
      "eval_samples_per_second": 4.71,
      "eval_steps_per_second": 4.71,
      "step": 500
    },
    {
      "epoch": 0.18792445808912456,
      "grad_norm": 0.012235240079462528,
      "learning_rate": 1.9533581393952978e-05,
      "loss": 0.6108,
      "step": 505
    },
    {
      "epoch": 0.18978509628802678,
      "grad_norm": 0.012589952908456326,
      "learning_rate": 1.951377195390367e-05,
      "loss": 0.6218,
      "step": 510
    },
    {
      "epoch": 0.191645734486929,
      "grad_norm": 0.012783786281943321,
      "learning_rate": 1.9493561020491024e-05,
      "loss": 0.6668,
      "step": 515
    },
    {
      "epoch": 0.19350637268583123,
      "grad_norm": 0.013828632421791553,
      "learning_rate": 1.9472949446642318e-05,
      "loss": 0.6081,
      "step": 520
    },
    {
      "epoch": 0.19536701088473346,
      "grad_norm": 0.011929171159863472,
      "learning_rate": 1.945193810219237e-05,
      "loss": 0.6329,
      "step": 525
    },
    {
      "epoch": 0.1972276490836357,
      "grad_norm": 0.014584473334252834,
      "learning_rate": 1.9430527873846826e-05,
      "loss": 0.7017,
      "step": 530
    },
    {
      "epoch": 0.1990882872825379,
      "grad_norm": 0.01474926806986332,
      "learning_rate": 1.9408719665144756e-05,
      "loss": 0.632,
      "step": 535
    },
    {
      "epoch": 0.20094892548144014,
      "grad_norm": 0.015552829019725323,
      "learning_rate": 1.9386514396420503e-05,
      "loss": 0.6757,
      "step": 540
    },
    {
      "epoch": 0.20280956368034236,
      "grad_norm": 0.013232480734586716,
      "learning_rate": 1.9363913004764847e-05,
      "loss": 0.6722,
      "step": 545
    },
    {
      "epoch": 0.2046702018792446,
      "grad_norm": 0.012858827598392963,
      "learning_rate": 1.9340916443985465e-05,
      "loss": 0.6231,
      "step": 550
    },
    {
      "epoch": 0.20653084007814682,
      "grad_norm": 0.012365566566586494,
      "learning_rate": 1.9317525684566686e-05,
      "loss": 0.5986,
      "step": 555
    },
    {
      "epoch": 0.20839147827704904,
      "grad_norm": 0.01528852991759777,
      "learning_rate": 1.9293741713628518e-05,
      "loss": 0.6537,
      "step": 560
    },
    {
      "epoch": 0.21025211647595124,
      "grad_norm": 0.014512522146105766,
      "learning_rate": 1.9269565534885003e-05,
      "loss": 0.6527,
      "step": 565
    },
    {
      "epoch": 0.21211275467485347,
      "grad_norm": 0.013798325322568417,
      "learning_rate": 1.9244998168601848e-05,
      "loss": 0.6148,
      "step": 570
    },
    {
      "epoch": 0.2139733928737557,
      "grad_norm": 0.013186248019337654,
      "learning_rate": 1.9220040651553388e-05,
      "loss": 0.6106,
      "step": 575
    },
    {
      "epoch": 0.21583403107265792,
      "grad_norm": 0.013986771926283836,
      "learning_rate": 1.9194694036978807e-05,
      "loss": 0.6654,
      "step": 580
    },
    {
      "epoch": 0.21769466927156014,
      "grad_norm": 0.016201818361878395,
      "learning_rate": 1.9168959394537708e-05,
      "loss": 0.6306,
      "step": 585
    },
    {
      "epoch": 0.21955530747046237,
      "grad_norm": 0.013889294117689133,
      "learning_rate": 1.9142837810264972e-05,
      "loss": 0.6749,
      "step": 590
    },
    {
      "epoch": 0.2214159456693646,
      "grad_norm": 0.013025142252445221,
      "learning_rate": 1.911633038652491e-05,
      "loss": 0.6075,
      "step": 595
    },
    {
      "epoch": 0.22327658386826682,
      "grad_norm": 0.013716059736907482,
      "learning_rate": 1.9089438241964764e-05,
      "loss": 0.6516,
      "step": 600
    },
    {
      "epoch": 0.22327658386826682,
      "eval_loss": 0.6547934412956238,
      "eval_runtime": 26.7928,
      "eval_samples_per_second": 4.74,
      "eval_steps_per_second": 4.74,
      "step": 600
    },
    {
      "epoch": 0.22513722206716905,
      "grad_norm": 0.016251811757683754,
      "learning_rate": 1.906216251146748e-05,
      "loss": 0.6265,
      "step": 605
    },
    {
      "epoch": 0.22699786026607127,
      "grad_norm": 0.013359563425183296,
      "learning_rate": 1.9034504346103825e-05,
      "loss": 0.6052,
      "step": 610
    },
    {
      "epoch": 0.2288584984649735,
      "grad_norm": 0.012794552370905876,
      "learning_rate": 1.9006464913083807e-05,
      "loss": 0.613,
      "step": 615
    },
    {
      "epoch": 0.2307191366638757,
      "grad_norm": 0.012533072382211685,
      "learning_rate": 1.897804539570742e-05,
      "loss": 0.6735,
      "step": 620
    },
    {
      "epoch": 0.23257977486277792,
      "grad_norm": 0.013286220841109753,
      "learning_rate": 1.8949246993314694e-05,
      "loss": 0.6692,
      "step": 625
    },
    {
      "epoch": 0.23444041306168015,
      "grad_norm": 0.013466808013617992,
      "learning_rate": 1.892007092123511e-05,
      "loss": 0.6513,
      "step": 630
    },
    {
      "epoch": 0.23630105126058237,
      "grad_norm": 0.012991335242986679,
      "learning_rate": 1.8890518410736275e-05,
      "loss": 0.6405,
      "step": 635
    },
    {
      "epoch": 0.2381616894594846,
      "grad_norm": 0.013223089277744293,
      "learning_rate": 1.8860590708971997e-05,
      "loss": 0.6488,
      "step": 640
    },
    {
      "epoch": 0.24002232765838682,
      "grad_norm": 0.012394067831337452,
      "learning_rate": 1.8830289078929618e-05,
      "loss": 0.6131,
      "step": 645
    },
    {
      "epoch": 0.24188296585728905,
      "grad_norm": 0.013721502386033535,
      "learning_rate": 1.8799614799376743e-05,
      "loss": 0.681,
      "step": 650
    },
    {
      "epoch": 0.24374360405619128,
      "grad_norm": 0.012726777233183384,
      "learning_rate": 1.8768569164807272e-05,
      "loss": 0.6837,
      "step": 655
    },
    {
      "epoch": 0.2456042422550935,
      "grad_norm": 0.013720668852329254,
      "learning_rate": 1.8737153485386737e-05,
      "loss": 0.6007,
      "step": 660
    },
    {
      "epoch": 0.24746488045399573,
      "grad_norm": 0.012646087445318699,
      "learning_rate": 1.8705369086897063e-05,
      "loss": 0.6545,
      "step": 665
    },
    {
      "epoch": 0.24932551865289795,
      "grad_norm": 0.013658811338245869,
      "learning_rate": 1.8673217310680578e-05,
      "loss": 0.6379,
      "step": 670
    },
    {
      "epoch": 0.2511861568518002,
      "grad_norm": 0.012248256243765354,
      "learning_rate": 1.864069951358342e-05,
      "loss": 0.6356,
      "step": 675
    },
    {
      "epoch": 0.2530467950507024,
      "grad_norm": 0.0133894681930542,
      "learning_rate": 1.860781706789829e-05,
      "loss": 0.6737,
      "step": 680
    },
    {
      "epoch": 0.25490743324960463,
      "grad_norm": 0.01430124044418335,
      "learning_rate": 1.857457136130651e-05,
      "loss": 0.6169,
      "step": 685
    },
    {
      "epoch": 0.25676807144850683,
      "grad_norm": 0.013437042012810707,
      "learning_rate": 1.854096379681949e-05,
      "loss": 0.6021,
      "step": 690
    },
    {
      "epoch": 0.2586287096474091,
      "grad_norm": 0.011497768573462963,
      "learning_rate": 1.8506995792719498e-05,
      "loss": 0.6119,
      "step": 695
    },
    {
      "epoch": 0.2604893478463113,
      "grad_norm": 0.01319235097616911,
      "learning_rate": 1.8472668782499817e-05,
      "loss": 0.627,
      "step": 700
    },
    {
      "epoch": 0.2604893478463113,
      "eval_loss": 0.6522720456123352,
      "eval_runtime": 26.7201,
      "eval_samples_per_second": 4.753,
      "eval_steps_per_second": 4.753,
      "step": 700
    },
    {
      "epoch": 0.26234998604521353,
      "grad_norm": 0.014044429175555706,
      "learning_rate": 1.843798421480426e-05,
      "loss": 0.6244,
      "step": 705
    },
    {
      "epoch": 0.26421062424411573,
      "grad_norm": 0.011311609297990799,
      "learning_rate": 1.8402943553365998e-05,
      "loss": 0.5975,
      "step": 710
    },
    {
      "epoch": 0.26607126244301793,
      "grad_norm": 0.012355692684650421,
      "learning_rate": 1.8367548276945846e-05,
      "loss": 0.6009,
      "step": 715
    },
    {
      "epoch": 0.2679319006419202,
      "grad_norm": 0.012264437042176723,
      "learning_rate": 1.83317998792698e-05,
      "loss": 0.6103,
      "step": 720
    },
    {
      "epoch": 0.2697925388408224,
      "grad_norm": 0.012037084437906742,
      "learning_rate": 1.8295699868966038e-05,
      "loss": 0.5602,
      "step": 725
    },
    {
      "epoch": 0.27165317703972464,
      "grad_norm": 0.012773050926625729,
      "learning_rate": 1.8259249769501237e-05,
      "loss": 0.6215,
      "step": 730
    },
    {
      "epoch": 0.27351381523862683,
      "grad_norm": 0.012415550649166107,
      "learning_rate": 1.8222451119116288e-05,
      "loss": 0.6364,
      "step": 735
    },
    {
      "epoch": 0.2753744534375291,
      "grad_norm": 0.01302468404173851,
      "learning_rate": 1.8185305470761366e-05,
      "loss": 0.5994,
      "step": 740
    },
    {
      "epoch": 0.2772350916364313,
      "grad_norm": 0.013447316363453865,
      "learning_rate": 1.814781439203043e-05,
      "loss": 0.6458,
      "step": 745
    },
    {
      "epoch": 0.27909572983533354,
      "grad_norm": 0.012098093517124653,
      "learning_rate": 1.8109979465095014e-05,
      "loss": 0.6357,
      "step": 750
    },
    {
      "epoch": 0.28095636803423574,
      "grad_norm": 0.012987499125301838,
      "learning_rate": 1.8071802286637505e-05,
      "loss": 0.6248,
      "step": 755
    },
    {
      "epoch": 0.282817006233138,
      "grad_norm": 0.011747024022042751,
      "learning_rate": 1.8033284467783742e-05,
      "loss": 0.6202,
      "step": 760
    },
    {
      "epoch": 0.2846776444320402,
      "grad_norm": 0.01332057174295187,
      "learning_rate": 1.7994427634035016e-05,
      "loss": 0.6347,
      "step": 765
    },
    {
      "epoch": 0.2865382826309424,
      "grad_norm": 0.013383504003286362,
      "learning_rate": 1.795523342519948e-05,
      "loss": 0.6001,
      "step": 770
    },
    {
      "epoch": 0.28839892082984464,
      "grad_norm": 0.013648821040987968,
      "learning_rate": 1.7915703495322967e-05,
      "loss": 0.6399,
      "step": 775
    },
    {
      "epoch": 0.29025955902874684,
      "grad_norm": 0.012947522103786469,
      "learning_rate": 1.7875839512619148e-05,
      "loss": 0.6298,
      "step": 780
    },
    {
      "epoch": 0.2921201972276491,
      "grad_norm": 0.01334394421428442,
      "learning_rate": 1.7835643159399156e-05,
      "loss": 0.6418,
      "step": 785
    },
    {
      "epoch": 0.2939808354265513,
      "grad_norm": 0.014045110903680325,
      "learning_rate": 1.7795116132000587e-05,
      "loss": 0.6403,
      "step": 790
    },
    {
      "epoch": 0.29584147362545354,
      "grad_norm": 0.015219368040561676,
      "learning_rate": 1.7754260140715918e-05,
      "loss": 0.6277,
      "step": 795
    },
    {
      "epoch": 0.29770211182435574,
      "grad_norm": 0.01307649165391922,
      "learning_rate": 1.771307690972031e-05,
      "loss": 0.6271,
      "step": 800
    },
    {
      "epoch": 0.29770211182435574,
      "eval_loss": 0.6502260565757751,
      "eval_runtime": 26.7042,
      "eval_samples_per_second": 4.756,
      "eval_steps_per_second": 4.756,
      "step": 800
    },
    {
      "epoch": 0.299562750023258,
      "grad_norm": 0.013835963793098927,
      "learning_rate": 1.7671568176998865e-05,
      "loss": 0.6286,
      "step": 805
    },
    {
      "epoch": 0.3014233882221602,
      "grad_norm": 0.013574733398854733,
      "learning_rate": 1.762973569427328e-05,
      "loss": 0.6462,
      "step": 810
    },
    {
      "epoch": 0.30328402642106245,
      "grad_norm": 0.01131366565823555,
      "learning_rate": 1.758758122692791e-05,
      "loss": 0.6167,
      "step": 815
    },
    {
      "epoch": 0.30514466461996465,
      "grad_norm": 0.013482702895998955,
      "learning_rate": 1.7545106553935277e-05,
      "loss": 0.6413,
      "step": 820
    },
    {
      "epoch": 0.3070053028188669,
      "grad_norm": 0.012861824594438076,
      "learning_rate": 1.7502313467780988e-05,
      "loss": 0.6027,
      "step": 825
    },
    {
      "epoch": 0.3088659410177691,
      "grad_norm": 0.012556380592286587,
      "learning_rate": 1.7459203774388097e-05,
      "loss": 0.6603,
      "step": 830
    },
    {
      "epoch": 0.3107265792166713,
      "grad_norm": 0.012278062291443348,
      "learning_rate": 1.7415779293040887e-05,
      "loss": 0.5803,
      "step": 835
    },
    {
      "epoch": 0.31258721741557355,
      "grad_norm": 0.012213567271828651,
      "learning_rate": 1.7372041856308098e-05,
      "loss": 0.6624,
      "step": 840
    },
    {
      "epoch": 0.31444785561447575,
      "grad_norm": 0.0131307952105999,
      "learning_rate": 1.7327993309965583e-05,
      "loss": 0.6447,
      "step": 845
    },
    {
      "epoch": 0.316308493813378,
      "grad_norm": 0.0121999466791749,
      "learning_rate": 1.7283635512918423e-05,
      "loss": 0.6451,
      "step": 850
    },
    {
      "epoch": 0.3181691320122802,
      "grad_norm": 0.012872702442109585,
      "learning_rate": 1.7238970337122484e-05,
      "loss": 0.5724,
      "step": 855
    },
    {
      "epoch": 0.32002977021118245,
      "grad_norm": 0.013137550093233585,
      "learning_rate": 1.7193999667505387e-05,
      "loss": 0.6459,
      "step": 860
    },
    {
      "epoch": 0.32189040841008465,
      "grad_norm": 0.013948196545243263,
      "learning_rate": 1.7148725401887002e-05,
      "loss": 0.651,
      "step": 865
    },
    {
      "epoch": 0.3237510466089869,
      "grad_norm": 0.012517811730504036,
      "learning_rate": 1.710314945089933e-05,
      "loss": 0.6114,
      "step": 870
    },
    {
      "epoch": 0.3256116848078891,
      "grad_norm": 0.014199101366102695,
      "learning_rate": 1.7057273737905887e-05,
      "loss": 0.6405,
      "step": 875
    },
    {
      "epoch": 0.32747232300679135,
      "grad_norm": 0.012591714970767498,
      "learning_rate": 1.7011100198920528e-05,
      "loss": 0.6767,
      "step": 880
    },
    {
      "epoch": 0.32933296120569355,
      "grad_norm": 0.012114683166146278,
      "learning_rate": 1.6964630782525743e-05,
      "loss": 0.6037,
      "step": 885
    },
    {
      "epoch": 0.33119359940459575,
      "grad_norm": 0.013139299117028713,
      "learning_rate": 1.6917867449790432e-05,
      "loss": 0.643,
      "step": 890
    },
    {
      "epoch": 0.333054237603498,
      "grad_norm": 0.012391779571771622,
      "learning_rate": 1.6870812174187136e-05,
      "loss": 0.647,
      "step": 895
    },
    {
      "epoch": 0.3349148758024002,
      "grad_norm": 0.013360656797885895,
      "learning_rate": 1.6823466941508762e-05,
      "loss": 0.725,
      "step": 900
    },
    {
      "epoch": 0.3349148758024002,
      "eval_loss": 0.6488396525382996,
      "eval_runtime": 27.3774,
      "eval_samples_per_second": 4.639,
      "eval_steps_per_second": 4.639,
      "step": 900
    },
    {
      "epoch": 0.33677551400130246,
      "grad_norm": 0.013424807228147984,
      "learning_rate": 1.677583374978478e-05,
      "loss": 0.6342,
      "step": 905
    },
    {
      "epoch": 0.33863615220020465,
      "grad_norm": 0.01573541946709156,
      "learning_rate": 1.6727914609196895e-05,
      "loss": 0.6562,
      "step": 910
    },
    {
      "epoch": 0.3404967903991069,
      "grad_norm": 0.011363287456333637,
      "learning_rate": 1.6679711541994227e-05,
      "loss": 0.6492,
      "step": 915
    },
    {
      "epoch": 0.3423574285980091,
      "grad_norm": 0.012154373340308666,
      "learning_rate": 1.6631226582407954e-05,
      "loss": 0.602,
      "step": 920
    },
    {
      "epoch": 0.34421806679691136,
      "grad_norm": 0.01744014024734497,
      "learning_rate": 1.658246177656548e-05,
      "loss": 0.6318,
      "step": 925
    },
    {
      "epoch": 0.34607870499581356,
      "grad_norm": 0.012465902604162693,
      "learning_rate": 1.6533419182404078e-05,
      "loss": 0.6522,
      "step": 930
    },
    {
      "epoch": 0.3479393431947158,
      "grad_norm": 0.015055039897561073,
      "learning_rate": 1.6484100869584044e-05,
      "loss": 0.6376,
      "step": 935
    },
    {
      "epoch": 0.349799981393618,
      "grad_norm": 0.012351201847195625,
      "learning_rate": 1.6434508919401357e-05,
      "loss": 0.6206,
      "step": 940
    },
    {
      "epoch": 0.3516606195925202,
      "grad_norm": 0.012793191708624363,
      "learning_rate": 1.6384645424699835e-05,
      "loss": 0.6182,
      "step": 945
    },
    {
      "epoch": 0.35352125779142246,
      "grad_norm": 0.012946651317179203,
      "learning_rate": 1.6334512489782833e-05,
      "loss": 0.5839,
      "step": 950
    },
    {
      "epoch": 0.35538189599032466,
      "grad_norm": 0.012998082675039768,
      "learning_rate": 1.628411223032442e-05,
      "loss": 0.6517,
      "step": 955
    },
    {
      "epoch": 0.3572425341892269,
      "grad_norm": 0.012614963576197624,
      "learning_rate": 1.6233446773280113e-05,
      "loss": 0.6235,
      "step": 960
    },
    {
      "epoch": 0.3591031723881291,
      "grad_norm": 0.012318151071667671,
      "learning_rate": 1.6182518256797095e-05,
      "loss": 0.664,
      "step": 965
    },
    {
      "epoch": 0.36096381058703136,
      "grad_norm": 0.012551162391901016,
      "learning_rate": 1.6131328830123997e-05,
      "loss": 0.6317,
      "step": 970
    },
    {
      "epoch": 0.36282444878593356,
      "grad_norm": 0.013372802175581455,
      "learning_rate": 1.60798806535202e-05,
      "loss": 0.6418,
      "step": 975
    },
    {
      "epoch": 0.3646850869848358,
      "grad_norm": 0.011675640940666199,
      "learning_rate": 1.6028175898164665e-05,
      "loss": 0.6118,
      "step": 980
    },
    {
      "epoch": 0.366545725183738,
      "grad_norm": 0.013295911252498627,
      "learning_rate": 1.5976216746064294e-05,
      "loss": 0.6217,
      "step": 985
    },
    {
      "epoch": 0.36840636338264027,
      "grad_norm": 0.012895721010863781,
      "learning_rate": 1.5924005389961866e-05,
      "loss": 0.6436,
      "step": 990
    },
    {
      "epoch": 0.37026700158154247,
      "grad_norm": 0.014572090469300747,
      "learning_rate": 1.5871544033243488e-05,
      "loss": 0.6342,
      "step": 995
    },
    {
      "epoch": 0.3721276397804447,
      "grad_norm": 0.012393898330628872,
      "learning_rate": 1.581883488984562e-05,
      "loss": 0.6218,
      "step": 1000
    },
    {
      "epoch": 0.3721276397804447,
      "eval_loss": 0.6476932168006897,
      "eval_runtime": 27.1488,
      "eval_samples_per_second": 4.678,
      "eval_steps_per_second": 4.678,
      "step": 1000
    },
    {
      "epoch": 0.3739882779793469,
      "grad_norm": 0.014692210592329502,
      "learning_rate": 1.5765880184161625e-05,
      "loss": 0.6216,
      "step": 1005
    },
    {
      "epoch": 0.3758489161782491,
      "grad_norm": 0.012152746319770813,
      "learning_rate": 1.5712682150947926e-05,
      "loss": 0.6243,
      "step": 1010
    },
    {
      "epoch": 0.37770955437715137,
      "grad_norm": 0.012929155491292477,
      "learning_rate": 1.5659243035229657e-05,
      "loss": 0.6493,
      "step": 1015
    },
    {
      "epoch": 0.37957019257605357,
      "grad_norm": 0.0136475944891572,
      "learning_rate": 1.5605565092205973e-05,
      "loss": 0.6506,
      "step": 1020
    },
    {
      "epoch": 0.3814308307749558,
      "grad_norm": 0.014008302241563797,
      "learning_rate": 1.5551650587154815e-05,
      "loss": 0.6429,
      "step": 1025
    },
    {
      "epoch": 0.383291468973858,
      "grad_norm": 0.014000017195940018,
      "learning_rate": 1.5497501795337366e-05,
      "loss": 0.6277,
      "step": 1030
    },
    {
      "epoch": 0.38515210717276027,
      "grad_norm": 0.012146887369453907,
      "learning_rate": 1.5443121001901994e-05,
      "loss": 0.635,
      "step": 1035
    },
    {
      "epoch": 0.38701274537166247,
      "grad_norm": 0.013878699392080307,
      "learning_rate": 1.5388510501787855e-05,
      "loss": 0.6416,
      "step": 1040
    },
    {
      "epoch": 0.3888733835705647,
      "grad_norm": 0.011823480948805809,
      "learning_rate": 1.5333672599628005e-05,
      "loss": 0.637,
      "step": 1045
    },
    {
      "epoch": 0.3907340217694669,
      "grad_norm": 0.012524113990366459,
      "learning_rate": 1.527860960965216e-05,
      "loss": 0.6763,
      "step": 1050
    },
    {
      "epoch": 0.3925946599683692,
      "grad_norm": 0.013192784041166306,
      "learning_rate": 1.5223323855589027e-05,
      "loss": 0.6501,
      "step": 1055
    },
    {
      "epoch": 0.3944552981672714,
      "grad_norm": 0.012439992278814316,
      "learning_rate": 1.5167817670568253e-05,
      "loss": 0.5886,
      "step": 1060
    },
    {
      "epoch": 0.39631593636617357,
      "grad_norm": 0.013605669140815735,
      "learning_rate": 1.5112093397021945e-05,
      "loss": 0.5925,
      "step": 1065
    },
    {
      "epoch": 0.3981765745650758,
      "grad_norm": 0.01384530495852232,
      "learning_rate": 1.5056153386585828e-05,
      "loss": 0.6607,
      "step": 1070
    },
    {
      "epoch": 0.400037212763978,
      "grad_norm": 0.014060786925256252,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.6458,
      "step": 1075
    },
    {
      "epoch": 0.4018978509628803,
      "grad_norm": 0.012853951193392277,
      "learning_rate": 1.494363560700931e-05,
      "loss": 0.6028,
      "step": 1080
    },
    {
      "epoch": 0.4037584891617825,
      "grad_norm": 0.012846381403505802,
      "learning_rate": 1.4887062586263334e-05,
      "loss": 0.6543,
      "step": 1085
    },
    {
      "epoch": 0.40561912736068473,
      "grad_norm": 0.013001542538404465,
      "learning_rate": 1.4830283325216026e-05,
      "loss": 0.5654,
      "step": 1090
    },
    {
      "epoch": 0.4074797655595869,
      "grad_norm": 0.012333175167441368,
      "learning_rate": 1.477330022002493e-05,
      "loss": 0.6465,
      "step": 1095
    },
    {
      "epoch": 0.4093404037584892,
      "grad_norm": 0.01342159602791071,
      "learning_rate": 1.4716115675450078e-05,
      "loss": 0.6168,
      "step": 1100
    },
    {
      "epoch": 0.4093404037584892,
      "eval_loss": 0.6468775272369385,
      "eval_runtime": 26.8263,
      "eval_samples_per_second": 4.734,
      "eval_steps_per_second": 4.734,
      "step": 1100
    },
    {
      "epoch": 0.4112010419573914,
      "grad_norm": 0.0139292748644948,
      "learning_rate": 1.4658732104752507e-05,
      "loss": 0.634,
      "step": 1105
    },
    {
      "epoch": 0.41306168015629363,
      "grad_norm": 0.013169731944799423,
      "learning_rate": 1.4601151929592403e-05,
      "loss": 0.6227,
      "step": 1110
    },
    {
      "epoch": 0.41492231835519583,
      "grad_norm": 0.013705245219171047,
      "learning_rate": 1.4543377579926915e-05,
      "loss": 0.6441,
      "step": 1115
    },
    {
      "epoch": 0.4167829565540981,
      "grad_norm": 0.013035726733505726,
      "learning_rate": 1.4485411493907617e-05,
      "loss": 0.6498,
      "step": 1120
    },
    {
      "epoch": 0.4186435947530003,
      "grad_norm": 0.01190096139907837,
      "learning_rate": 1.442725611777758e-05,
      "loss": 0.6285,
      "step": 1125
    },
    {
      "epoch": 0.4205042329519025,
      "grad_norm": 0.013753347098827362,
      "learning_rate": 1.4368913905768178e-05,
      "loss": 0.6541,
      "step": 1130
    },
    {
      "epoch": 0.42236487115080473,
      "grad_norm": 0.012330746278166771,
      "learning_rate": 1.4310387319995492e-05,
      "loss": 0.6721,
      "step": 1135
    },
    {
      "epoch": 0.42422550934970693,
      "grad_norm": 0.01278294064104557,
      "learning_rate": 1.4251678830356408e-05,
      "loss": 0.6589,
      "step": 1140
    },
    {
      "epoch": 0.4260861475486092,
      "grad_norm": 0.012772184796631336,
      "learning_rate": 1.41927909144244e-05,
      "loss": 0.6411,
      "step": 1145
    },
    {
      "epoch": 0.4279467857475114,
      "grad_norm": 0.012047790922224522,
      "learning_rate": 1.413372605734495e-05,
      "loss": 0.5759,
      "step": 1150
    },
    {
      "epoch": 0.42980742394641364,
      "grad_norm": 0.014543715864419937,
      "learning_rate": 1.4074486751730687e-05,
      "loss": 0.6578,
      "step": 1155
    },
    {
      "epoch": 0.43166806214531583,
      "grad_norm": 0.013436605222523212,
      "learning_rate": 1.4015075497556193e-05,
      "loss": 0.5876,
      "step": 1160
    },
    {
      "epoch": 0.4335287003442181,
      "grad_norm": 0.011584432795643806,
      "learning_rate": 1.3955494802052498e-05,
      "loss": 0.656,
      "step": 1165
    },
    {
      "epoch": 0.4353893385431203,
      "grad_norm": 0.012196795083582401,
      "learning_rate": 1.3895747179601275e-05,
      "loss": 0.6562,
      "step": 1170
    },
    {
      "epoch": 0.43724997674202254,
      "grad_norm": 0.011762428097426891,
      "learning_rate": 1.3835835151628728e-05,
      "loss": 0.5918,
      "step": 1175
    },
    {
      "epoch": 0.43911061494092474,
      "grad_norm": 0.014100808650255203,
      "learning_rate": 1.3775761246499177e-05,
      "loss": 0.6216,
      "step": 1180
    },
    {
      "epoch": 0.44097125313982694,
      "grad_norm": 0.011724433861672878,
      "learning_rate": 1.3715527999408376e-05,
      "loss": 0.6434,
      "step": 1185
    },
    {
      "epoch": 0.4428318913387292,
      "grad_norm": 0.013045977801084518,
      "learning_rate": 1.365513795227651e-05,
      "loss": 0.5915,
      "step": 1190
    },
    {
      "epoch": 0.4446925295376314,
      "grad_norm": 0.013339078053832054,
      "learning_rate": 1.359459365364092e-05,
      "loss": 0.6148,
      "step": 1195
    },
    {
      "epoch": 0.44655316773653364,
      "grad_norm": 0.011851554736495018,
      "learning_rate": 1.3533897658548571e-05,
      "loss": 0.6294,
      "step": 1200
    },
    {
      "epoch": 0.44655316773653364,
      "eval_loss": 0.6459712982177734,
      "eval_runtime": 27.1858,
      "eval_samples_per_second": 4.672,
      "eval_steps_per_second": 4.672,
      "step": 1200
    },
    {
      "epoch": 0.44841380593543584,
      "grad_norm": 0.014516811817884445,
      "learning_rate": 1.3473052528448203e-05,
      "loss": 0.6052,
      "step": 1205
    },
    {
      "epoch": 0.4502744441343381,
      "grad_norm": 0.013740907423198223,
      "learning_rate": 1.341206083108225e-05,
      "loss": 0.6035,
      "step": 1210
    },
    {
      "epoch": 0.4521350823332403,
      "grad_norm": 0.013699905015528202,
      "learning_rate": 1.3350925140378465e-05,
      "loss": 0.64,
      "step": 1215
    },
    {
      "epoch": 0.45399572053214254,
      "grad_norm": 0.013219136744737625,
      "learning_rate": 1.328964803634131e-05,
      "loss": 0.6102,
      "step": 1220
    },
    {
      "epoch": 0.45585635873104474,
      "grad_norm": 0.012603058479726315,
      "learning_rate": 1.3228232104943073e-05,
      "loss": 0.5452,
      "step": 1225
    },
    {
      "epoch": 0.457716996929947,
      "grad_norm": 0.012072132900357246,
      "learning_rate": 1.3166679938014728e-05,
      "loss": 0.5864,
      "step": 1230
    },
    {
      "epoch": 0.4595776351288492,
      "grad_norm": 0.01093310210853815,
      "learning_rate": 1.3104994133136563e-05,
      "loss": 0.6122,
      "step": 1235
    },
    {
      "epoch": 0.4614382733277514,
      "grad_norm": 0.01245942059904337,
      "learning_rate": 1.3043177293528571e-05,
      "loss": 0.5889,
      "step": 1240
    },
    {
      "epoch": 0.46329891152665365,
      "grad_norm": 0.013127041980624199,
      "learning_rate": 1.2981232027940562e-05,
      "loss": 0.6225,
      "step": 1245
    },
    {
      "epoch": 0.46515954972555584,
      "grad_norm": 0.018618909642100334,
      "learning_rate": 1.2919160950542095e-05,
      "loss": 0.6189,
      "step": 1250
    },
    {
      "epoch": 0.4670201879244581,
      "grad_norm": 0.013042682781815529,
      "learning_rate": 1.2856966680812148e-05,
      "loss": 0.674,
      "step": 1255
    },
    {
      "epoch": 0.4688808261233603,
      "grad_norm": 0.013208975084125996,
      "learning_rate": 1.2794651843428575e-05,
      "loss": 0.6084,
      "step": 1260
    },
    {
      "epoch": 0.47074146432226255,
      "grad_norm": 0.012473770417273045,
      "learning_rate": 1.2732219068157335e-05,
      "loss": 0.5748,
      "step": 1265
    },
    {
      "epoch": 0.47260210252116475,
      "grad_norm": 0.013629582710564137,
      "learning_rate": 1.2669670989741519e-05,
      "loss": 0.6358,
      "step": 1270
    },
    {
      "epoch": 0.474462740720067,
      "grad_norm": 0.014257396571338177,
      "learning_rate": 1.2607010247790158e-05,
      "loss": 0.6794,
      "step": 1275
    },
    {
      "epoch": 0.4763233789189692,
      "grad_norm": 0.0164741612970829,
      "learning_rate": 1.2544239486666831e-05,
      "loss": 0.6647,
      "step": 1280
    },
    {
      "epoch": 0.47818401711787145,
      "grad_norm": 0.012896180152893066,
      "learning_rate": 1.2481361355378066e-05,
      "loss": 0.6413,
      "step": 1285
    },
    {
      "epoch": 0.48004465531677365,
      "grad_norm": 0.012296337634325027,
      "learning_rate": 1.2418378507461544e-05,
      "loss": 0.62,
      "step": 1290
    },
    {
      "epoch": 0.4819052935156759,
      "grad_norm": 0.01213445421308279,
      "learning_rate": 1.2355293600874132e-05,
      "loss": 0.6611,
      "step": 1295
    },
    {
      "epoch": 0.4837659317145781,
      "grad_norm": 0.01198558695614338,
      "learning_rate": 1.229210929787969e-05,
      "loss": 0.6438,
      "step": 1300
    },
    {
      "epoch": 0.4837659317145781,
      "eval_loss": 0.6453238725662231,
      "eval_runtime": 26.6037,
      "eval_samples_per_second": 4.774,
      "eval_steps_per_second": 4.774,
      "step": 1300
    },
    {
      "epoch": 0.4856265699134803,
      "grad_norm": 0.012105841189622879,
      "learning_rate": 1.2228828264936755e-05,
      "loss": 0.675,
      "step": 1305
    },
    {
      "epoch": 0.48748720811238255,
      "grad_norm": 0.012657279148697853,
      "learning_rate": 1.2165453172585964e-05,
      "loss": 0.6066,
      "step": 1310
    },
    {
      "epoch": 0.48934784631128475,
      "grad_norm": 0.01320530753582716,
      "learning_rate": 1.2101986695337407e-05,
      "loss": 0.6578,
      "step": 1315
    },
    {
      "epoch": 0.491208484510187,
      "grad_norm": 0.012736879289150238,
      "learning_rate": 1.2038431511557715e-05,
      "loss": 0.6596,
      "step": 1320
    },
    {
      "epoch": 0.4930691227090892,
      "grad_norm": 0.013182558119297028,
      "learning_rate": 1.197479030335706e-05,
      "loss": 0.595,
      "step": 1325
    },
    {
      "epoch": 0.49492976090799146,
      "grad_norm": 0.013970241881906986,
      "learning_rate": 1.1911065756475953e-05,
      "loss": 0.6525,
      "step": 1330
    },
    {
      "epoch": 0.49679039910689365,
      "grad_norm": 0.012158108875155449,
      "learning_rate": 1.1847260560171895e-05,
      "loss": 0.576,
      "step": 1335
    },
    {
      "epoch": 0.4986510373057959,
      "grad_norm": 0.012398924678564072,
      "learning_rate": 1.1783377407105907e-05,
      "loss": 0.6039,
      "step": 1340
    },
    {
      "epoch": 0.5005116755046981,
      "grad_norm": 0.013791786506772041,
      "learning_rate": 1.1719418993228883e-05,
      "loss": 0.6585,
      "step": 1345
    },
    {
      "epoch": 0.5023723137036004,
      "grad_norm": 0.011280239559710026,
      "learning_rate": 1.1655388017667812e-05,
      "loss": 0.5919,
      "step": 1350
    },
    {
      "epoch": 0.5042329519025026,
      "grad_norm": 0.015436794608831406,
      "learning_rate": 1.159128718261189e-05,
      "loss": 0.6632,
      "step": 1355
    },
    {
      "epoch": 0.5060935901014048,
      "grad_norm": 0.015739573165774345,
      "learning_rate": 1.1527119193198466e-05,
      "loss": 0.6384,
      "step": 1360
    },
    {
      "epoch": 0.507954228300307,
      "grad_norm": 0.01318281702697277,
      "learning_rate": 1.146288675739889e-05,
      "loss": 0.6312,
      "step": 1365
    },
    {
      "epoch": 0.5098148664992093,
      "grad_norm": 0.01480270829051733,
      "learning_rate": 1.1398592585904234e-05,
      "loss": 0.6453,
      "step": 1370
    },
    {
      "epoch": 0.5116755046981114,
      "grad_norm": 0.011967113241553307,
      "learning_rate": 1.133423939201089e-05,
      "loss": 0.6335,
      "step": 1375
    },
    {
      "epoch": 0.5135361428970137,
      "grad_norm": 0.013129732571542263,
      "learning_rate": 1.1269829891506081e-05,
      "loss": 0.5852,
      "step": 1380
    },
    {
      "epoch": 0.5153967810959159,
      "grad_norm": 0.0132956113666296,
      "learning_rate": 1.1205366802553231e-05,
      "loss": 0.6477,
      "step": 1385
    },
    {
      "epoch": 0.5172574192948182,
      "grad_norm": 0.01314165536314249,
      "learning_rate": 1.1140852845577273e-05,
      "loss": 0.6441,
      "step": 1390
    },
    {
      "epoch": 0.5191180574937203,
      "grad_norm": 0.01178740430623293,
      "learning_rate": 1.1076290743149827e-05,
      "loss": 0.6035,
      "step": 1395
    },
    {
      "epoch": 0.5209786956926226,
      "grad_norm": 0.013940893113613129,
      "learning_rate": 1.1011683219874324e-05,
      "loss": 0.6492,
      "step": 1400
    },
    {
      "epoch": 0.5209786956926226,
      "eval_loss": 0.644782543182373,
      "eval_runtime": 26.5697,
      "eval_samples_per_second": 4.78,
      "eval_steps_per_second": 4.78,
      "step": 1400
    },
    {
      "epoch": 0.5228393338915248,
      "grad_norm": 0.013468287885189056,
      "learning_rate": 1.0947033002271001e-05,
      "loss": 0.6135,
      "step": 1405
    },
    {
      "epoch": 0.5246999720904271,
      "grad_norm": 0.015101495198905468,
      "learning_rate": 1.0882342818661859e-05,
      "loss": 0.6449,
      "step": 1410
    },
    {
      "epoch": 0.5265606102893292,
      "grad_norm": 0.013072527013719082,
      "learning_rate": 1.0817615399055513e-05,
      "loss": 0.6252,
      "step": 1415
    },
    {
      "epoch": 0.5284212484882315,
      "grad_norm": 0.01259040180593729,
      "learning_rate": 1.075285347503198e-05,
      "loss": 0.6274,
      "step": 1420
    },
    {
      "epoch": 0.5302818866871337,
      "grad_norm": 0.014084907248616219,
      "learning_rate": 1.0688059779627417e-05,
      "loss": 0.6298,
      "step": 1425
    },
    {
      "epoch": 0.5321425248860359,
      "grad_norm": 0.015118095092475414,
      "learning_rate": 1.0623237047218771e-05,
      "loss": 0.6638,
      "step": 1430
    },
    {
      "epoch": 0.5340031630849381,
      "grad_norm": 0.013168774545192719,
      "learning_rate": 1.0558388013408378e-05,
      "loss": 0.6134,
      "step": 1435
    },
    {
      "epoch": 0.5358638012838404,
      "grad_norm": 0.011875185184180737,
      "learning_rate": 1.0493515414908542e-05,
      "loss": 0.6396,
      "step": 1440
    },
    {
      "epoch": 0.5377244394827426,
      "grad_norm": 0.013676362112164497,
      "learning_rate": 1.0428621989426016e-05,
      "loss": 0.6286,
      "step": 1445
    },
    {
      "epoch": 0.5395850776816448,
      "grad_norm": 0.012775209732353687,
      "learning_rate": 1.0363710475546483e-05,
      "loss": 0.6156,
      "step": 1450
    },
    {
      "epoch": 0.541445715880547,
      "grad_norm": 0.014763396233320236,
      "learning_rate": 1.0298783612618977e-05,
      "loss": 0.6713,
      "step": 1455
    },
    {
      "epoch": 0.5433063540794493,
      "grad_norm": 0.014012233354151249,
      "learning_rate": 1.0233844140640287e-05,
      "loss": 0.5887,
      "step": 1460
    },
    {
      "epoch": 0.5451669922783515,
      "grad_norm": 0.013994456268846989,
      "learning_rate": 1.0168894800139311e-05,
      "loss": 0.6509,
      "step": 1465
    },
    {
      "epoch": 0.5470276304772537,
      "grad_norm": 0.014792009256780148,
      "learning_rate": 1.0103938332061422e-05,
      "loss": 0.6434,
      "step": 1470
    },
    {
      "epoch": 0.5488882686761559,
      "grad_norm": 0.01255644578486681,
      "learning_rate": 1.0038977477652779e-05,
      "loss": 0.6407,
      "step": 1475
    },
    {
      "epoch": 0.5507489068750582,
      "grad_norm": 0.013431582599878311,
      "learning_rate": 9.974014978344646e-06,
      "loss": 0.6528,
      "step": 1480
    },
    {
      "epoch": 0.5526095450739603,
      "grad_norm": 0.0123568931594491,
      "learning_rate": 9.909053575637717e-06,
      "loss": 0.602,
      "step": 1485
    },
    {
      "epoch": 0.5544701832728626,
      "grad_norm": 0.014569776132702827,
      "learning_rate": 9.844096010986392e-06,
      "loss": 0.6268,
      "step": 1490
    },
    {
      "epoch": 0.5563308214717648,
      "grad_norm": 0.012518184259533882,
      "learning_rate": 9.779145025683114e-06,
      "loss": 0.5936,
      "step": 1495
    },
    {
      "epoch": 0.5581914596706671,
      "grad_norm": 0.011615007184445858,
      "learning_rate": 9.714203360742666e-06,
      "loss": 0.6275,
      "step": 1500
    },
    {
      "epoch": 0.5581914596706671,
      "eval_loss": 0.6442868709564209,
      "eval_runtime": 26.5597,
      "eval_samples_per_second": 4.782,
      "eval_steps_per_second": 4.782,
      "step": 1500
    },
    {
      "epoch": 0.5600520978695692,
      "grad_norm": 0.013426556251943111,
      "learning_rate": 9.649273756786486e-06,
      "loss": 0.6291,
      "step": 1505
    },
    {
      "epoch": 0.5619127360684715,
      "grad_norm": 0.01306986529380083,
      "learning_rate": 9.584358953927043e-06,
      "loss": 0.6211,
      "step": 1510
    },
    {
      "epoch": 0.5637733742673737,
      "grad_norm": 0.012594708241522312,
      "learning_rate": 9.519461691652169e-06,
      "loss": 0.5803,
      "step": 1515
    },
    {
      "epoch": 0.565634012466276,
      "grad_norm": 0.014830299653112888,
      "learning_rate": 9.454584708709462e-06,
      "loss": 0.5976,
      "step": 1520
    },
    {
      "epoch": 0.5674946506651781,
      "grad_norm": 0.013333864510059357,
      "learning_rate": 9.389730742990714e-06,
      "loss": 0.6154,
      "step": 1525
    },
    {
      "epoch": 0.5693552888640804,
      "grad_norm": 0.013150627724826336,
      "learning_rate": 9.324902531416348e-06,
      "loss": 0.581,
      "step": 1530
    },
    {
      "epoch": 0.5712159270629826,
      "grad_norm": 0.012874056585133076,
      "learning_rate": 9.260102809819939e-06,
      "loss": 0.6224,
      "step": 1535
    },
    {
      "epoch": 0.5730765652618848,
      "grad_norm": 0.012678616680204868,
      "learning_rate": 9.195334312832742e-06,
      "loss": 0.6705,
      "step": 1540
    },
    {
      "epoch": 0.574937203460787,
      "grad_norm": 0.011814710684120655,
      "learning_rate": 9.1305997737683e-06,
      "loss": 0.6103,
      "step": 1545
    },
    {
      "epoch": 0.5767978416596893,
      "grad_norm": 0.013010908849537373,
      "learning_rate": 9.065901924507085e-06,
      "loss": 0.655,
      "step": 1550
    },
    {
      "epoch": 0.5786584798585915,
      "grad_norm": 0.013622297905385494,
      "learning_rate": 9.001243495381207e-06,
      "loss": 0.5961,
      "step": 1555
    },
    {
      "epoch": 0.5805191180574937,
      "grad_norm": 0.013876644894480705,
      "learning_rate": 8.936627215059206e-06,
      "loss": 0.6789,
      "step": 1560
    },
    {
      "epoch": 0.5823797562563959,
      "grad_norm": 0.01281541958451271,
      "learning_rate": 8.872055810430881e-06,
      "loss": 0.6567,
      "step": 1565
    },
    {
      "epoch": 0.5842403944552982,
      "grad_norm": 0.012182512320578098,
      "learning_rate": 8.80753200649222e-06,
      "loss": 0.6338,
      "step": 1570
    },
    {
      "epoch": 0.5861010326542004,
      "grad_norm": 0.012718496844172478,
      "learning_rate": 8.743058526230409e-06,
      "loss": 0.6151,
      "step": 1575
    },
    {
      "epoch": 0.5879616708531026,
      "grad_norm": 0.01301120687276125,
      "learning_rate": 8.678638090508897e-06,
      "loss": 0.6147,
      "step": 1580
    },
    {
      "epoch": 0.5898223090520048,
      "grad_norm": 0.011715607717633247,
      "learning_rate": 8.614273417952593e-06,
      "loss": 0.5776,
      "step": 1585
    },
    {
      "epoch": 0.5916829472509071,
      "grad_norm": 0.013846187852323055,
      "learning_rate": 8.549967224833131e-06,
      "loss": 0.6604,
      "step": 1590
    },
    {
      "epoch": 0.5935435854498092,
      "grad_norm": 0.013312350027263165,
      "learning_rate": 8.485722224954237e-06,
      "loss": 0.6395,
      "step": 1595
    },
    {
      "epoch": 0.5954042236487115,
      "grad_norm": 0.014313463121652603,
      "learning_rate": 8.421541129537194e-06,
      "loss": 0.6848,
      "step": 1600
    },
    {
      "epoch": 0.5954042236487115,
      "eval_loss": 0.6439012885093689,
      "eval_runtime": 26.5935,
      "eval_samples_per_second": 4.776,
      "eval_steps_per_second": 4.776,
      "step": 1600
    },
    {
      "epoch": 0.5972648618476137,
      "grad_norm": 0.012196023017168045,
      "learning_rate": 8.357426647106451e-06,
      "loss": 0.6079,
      "step": 1605
    },
    {
      "epoch": 0.599125500046516,
      "grad_norm": 0.0135956397280097,
      "learning_rate": 8.293381483375293e-06,
      "loss": 0.6463,
      "step": 1610
    },
    {
      "epoch": 0.6009861382454181,
      "grad_norm": 0.011982797645032406,
      "learning_rate": 8.229408341131665e-06,
      "loss": 0.6113,
      "step": 1615
    },
    {
      "epoch": 0.6028467764443204,
      "grad_norm": 0.014127896167337894,
      "learning_rate": 8.165509920124125e-06,
      "loss": 0.6602,
      "step": 1620
    },
    {
      "epoch": 0.6047074146432226,
      "grad_norm": 0.012588880024850368,
      "learning_rate": 8.10168891694789e-06,
      "loss": 0.6376,
      "step": 1625
    },
    {
      "epoch": 0.6065680528421249,
      "grad_norm": 0.012559432536363602,
      "learning_rate": 8.037948024931039e-06,
      "loss": 0.6336,
      "step": 1630
    },
    {
      "epoch": 0.608428691041027,
      "grad_norm": 0.012455436401069164,
      "learning_rate": 7.974289934020879e-06,
      "loss": 0.6403,
      "step": 1635
    },
    {
      "epoch": 0.6102893292399293,
      "grad_norm": 0.01241991762071848,
      "learning_rate": 7.91071733067038e-06,
      "loss": 0.6518,
      "step": 1640
    },
    {
      "epoch": 0.6121499674388315,
      "grad_norm": 0.012328005395829678,
      "learning_rate": 7.84723289772484e-06,
      "loss": 0.6162,
      "step": 1645
    },
    {
      "epoch": 0.6140106056377338,
      "grad_norm": 0.012782512232661247,
      "learning_rate": 7.783839314308656e-06,
      "loss": 0.6624,
      "step": 1650
    },
    {
      "epoch": 0.6158712438366359,
      "grad_norm": 0.012050081044435501,
      "learning_rate": 7.720539255712252e-06,
      "loss": 0.6565,
      "step": 1655
    },
    {
      "epoch": 0.6177318820355382,
      "grad_norm": 0.013794245198369026,
      "learning_rate": 7.657335393279179e-06,
      "loss": 0.6475,
      "step": 1660
    },
    {
      "epoch": 0.6195925202344404,
      "grad_norm": 0.012430761009454727,
      "learning_rate": 7.594230394293404e-06,
      "loss": 0.5821,
      "step": 1665
    },
    {
      "epoch": 0.6214531584333426,
      "grad_norm": 0.013150406070053577,
      "learning_rate": 7.531226921866715e-06,
      "loss": 0.6023,
      "step": 1670
    },
    {
      "epoch": 0.6233137966322448,
      "grad_norm": 0.013466687873005867,
      "learning_rate": 7.468327634826354e-06,
      "loss": 0.637,
      "step": 1675
    },
    {
      "epoch": 0.6251744348311471,
      "grad_norm": 0.014000273309648037,
      "learning_rate": 7.405535187602809e-06,
      "loss": 0.6113,
      "step": 1680
    },
    {
      "epoch": 0.6270350730300494,
      "grad_norm": 0.013736380264163017,
      "learning_rate": 7.3428522301177894e-06,
      "loss": 0.5914,
      "step": 1685
    },
    {
      "epoch": 0.6288957112289515,
      "grad_norm": 0.013854081742465496,
      "learning_rate": 7.2802814076723896e-06,
      "loss": 0.6744,
      "step": 1690
    },
    {
      "epoch": 0.6307563494278537,
      "grad_norm": 0.012866591103374958,
      "learning_rate": 7.217825360835475e-06,
      "loss": 0.6209,
      "step": 1695
    },
    {
      "epoch": 0.632616987626756,
      "grad_norm": 0.012788123451173306,
      "learning_rate": 7.155486725332224e-06,
      "loss": 0.5764,
      "step": 1700
    },
    {
      "epoch": 0.632616987626756,
      "eval_loss": 0.6434745192527771,
      "eval_runtime": 26.5737,
      "eval_samples_per_second": 4.779,
      "eval_steps_per_second": 4.779,
      "step": 1700
    },
    {
      "epoch": 0.6344776258256583,
      "grad_norm": 0.013442011550068855,
      "learning_rate": 7.093268131932905e-06,
      "loss": 0.6522,
      "step": 1705
    },
    {
      "epoch": 0.6363382640245604,
      "grad_norm": 0.013659958727657795,
      "learning_rate": 7.03117220634187e-06,
      "loss": 0.5949,
      "step": 1710
    },
    {
      "epoch": 0.6381989022234626,
      "grad_norm": 0.013367819599807262,
      "learning_rate": 6.9692015690867135e-06,
      "loss": 0.5959,
      "step": 1715
    },
    {
      "epoch": 0.6400595404223649,
      "grad_norm": 0.013543561100959778,
      "learning_rate": 6.9073588354077125e-06,
      "loss": 0.6539,
      "step": 1720
    },
    {
      "epoch": 0.641920178621267,
      "grad_norm": 0.014200146310031414,
      "learning_rate": 6.845646615147445e-06,
      "loss": 0.6438,
      "step": 1725
    },
    {
      "epoch": 0.6437808168201693,
      "grad_norm": 0.012410931289196014,
      "learning_rate": 6.784067512640666e-06,
      "loss": 0.6035,
      "step": 1730
    },
    {
      "epoch": 0.6456414550190716,
      "grad_norm": 0.01299245934933424,
      "learning_rate": 6.7226241266043735e-06,
      "loss": 0.6507,
      "step": 1735
    },
    {
      "epoch": 0.6475020932179738,
      "grad_norm": 0.013046164996922016,
      "learning_rate": 6.661319050028167e-06,
      "loss": 0.6277,
      "step": 1740
    },
    {
      "epoch": 0.649362731416876,
      "grad_norm": 0.013011117465794086,
      "learning_rate": 6.600154870064812e-06,
      "loss": 0.6415,
      "step": 1745
    },
    {
      "epoch": 0.6512233696157782,
      "grad_norm": 0.012195507064461708,
      "learning_rate": 6.53913416792105e-06,
      "loss": 0.5718,
      "step": 1750
    },
    {
      "epoch": 0.6530840078146805,
      "grad_norm": 0.012789854779839516,
      "learning_rate": 6.478259518748675e-06,
      "loss": 0.5963,
      "step": 1755
    },
    {
      "epoch": 0.6549446460135827,
      "grad_norm": 0.01217294204980135,
      "learning_rate": 6.41753349153587e-06,
      "loss": 0.6361,
      "step": 1760
    },
    {
      "epoch": 0.6568052842124849,
      "grad_norm": 0.014139696955680847,
      "learning_rate": 6.356958648998762e-06,
      "loss": 0.6321,
      "step": 1765
    },
    {
      "epoch": 0.6586659224113871,
      "grad_norm": 0.014064906165003777,
      "learning_rate": 6.296537547473302e-06,
      "loss": 0.6519,
      "step": 1770
    },
    {
      "epoch": 0.6605265606102894,
      "grad_norm": 0.012787656858563423,
      "learning_rate": 6.236272736807378e-06,
      "loss": 0.6033,
      "step": 1775
    },
    {
      "epoch": 0.6623871988091915,
      "grad_norm": 0.012475831434130669,
      "learning_rate": 6.176166760253196e-06,
      "loss": 0.5947,
      "step": 1780
    },
    {
      "epoch": 0.6642478370080938,
      "grad_norm": 0.013439202681183815,
      "learning_rate": 6.116222154359952e-06,
      "loss": 0.636,
      "step": 1785
    },
    {
      "epoch": 0.666108475206996,
      "grad_norm": 0.01295219361782074,
      "learning_rate": 6.056441448866817e-06,
      "loss": 0.6203,
      "step": 1790
    },
    {
      "epoch": 0.6679691134058983,
      "grad_norm": 0.012275603599846363,
      "learning_rate": 5.996827166596129e-06,
      "loss": 0.6743,
      "step": 1795
    },
    {
      "epoch": 0.6698297516048004,
      "grad_norm": 0.01329441275447607,
      "learning_rate": 5.937381823346964e-06,
      "loss": 0.5975,
      "step": 1800
    },
    {
      "epoch": 0.6698297516048004,
      "eval_loss": 0.6431623101234436,
      "eval_runtime": 26.5644,
      "eval_samples_per_second": 4.781,
      "eval_steps_per_second": 4.781,
      "step": 1800
    },
    {
      "epoch": 0.6716903898037027,
      "grad_norm": 0.013007073663175106,
      "learning_rate": 5.878107927788962e-06,
      "loss": 0.6165,
      "step": 1805
    },
    {
      "epoch": 0.6735510280026049,
      "grad_norm": 0.013040522113442421,
      "learning_rate": 5.819007981356441e-06,
      "loss": 0.6107,
      "step": 1810
    },
    {
      "epoch": 0.6754116662015072,
      "grad_norm": 0.014542641118168831,
      "learning_rate": 5.760084478142842e-06,
      "loss": 0.6284,
      "step": 1815
    },
    {
      "epoch": 0.6772723044004093,
      "grad_norm": 0.01339266262948513,
      "learning_rate": 5.701339904795486e-06,
      "loss": 0.6228,
      "step": 1820
    },
    {
      "epoch": 0.6791329425993116,
      "grad_norm": 0.013091943226754665,
      "learning_rate": 5.642776740410618e-06,
      "loss": 0.5995,
      "step": 1825
    },
    {
      "epoch": 0.6809935807982138,
      "grad_norm": 0.013321259059011936,
      "learning_rate": 5.584397456428785e-06,
      "loss": 0.627,
      "step": 1830
    },
    {
      "epoch": 0.682854218997116,
      "grad_norm": 0.012838364578783512,
      "learning_rate": 5.5262045165305615e-06,
      "loss": 0.658,
      "step": 1835
    },
    {
      "epoch": 0.6847148571960182,
      "grad_norm": 0.01306592021137476,
      "learning_rate": 5.468200376532552e-06,
      "loss": 0.6756,
      "step": 1840
    },
    {
      "epoch": 0.6865754953949205,
      "grad_norm": 0.014314945787191391,
      "learning_rate": 5.410387484283767e-06,
      "loss": 0.6598,
      "step": 1845
    },
    {
      "epoch": 0.6884361335938227,
      "grad_norm": 0.013973386958241463,
      "learning_rate": 5.352768279562315e-06,
      "loss": 0.6535,
      "step": 1850
    },
    {
      "epoch": 0.6902967717927249,
      "grad_norm": 0.012864621356129646,
      "learning_rate": 5.295345193972445e-06,
      "loss": 0.6422,
      "step": 1855
    },
    {
      "epoch": 0.6921574099916271,
      "grad_norm": 0.014392906799912453,
      "learning_rate": 5.238120650841925e-06,
      "loss": 0.664,
      "step": 1860
    },
    {
      "epoch": 0.6940180481905294,
      "grad_norm": 0.01327193807810545,
      "learning_rate": 5.18109706511978e-06,
      "loss": 0.6179,
      "step": 1865
    },
    {
      "epoch": 0.6958786863894316,
      "grad_norm": 0.01355487760156393,
      "learning_rate": 5.124276843274372e-06,
      "loss": 0.605,
      "step": 1870
    },
    {
      "epoch": 0.6977393245883338,
      "grad_norm": 0.012646518647670746,
      "learning_rate": 5.067662383191845e-06,
      "loss": 0.6608,
      "step": 1875
    },
    {
      "epoch": 0.699599962787236,
      "grad_norm": 0.013363865204155445,
      "learning_rate": 5.011256074074945e-06,
      "loss": 0.6975,
      "step": 1880
    },
    {
      "epoch": 0.7014606009861383,
      "grad_norm": 0.013032359071075916,
      "learning_rate": 4.955060296342163e-06,
      "loss": 0.6041,
      "step": 1885
    },
    {
      "epoch": 0.7033212391850404,
      "grad_norm": 0.012049228884279728,
      "learning_rate": 4.899077421527304e-06,
      "loss": 0.6078,
      "step": 1890
    },
    {
      "epoch": 0.7051818773839427,
      "grad_norm": 0.013416007161140442,
      "learning_rate": 4.843309812179405e-06,
      "loss": 0.6514,
      "step": 1895
    },
    {
      "epoch": 0.7070425155828449,
      "grad_norm": 0.013701760210096836,
      "learning_rate": 4.787759821763017e-06,
      "loss": 0.6606,
      "step": 1900
    },
    {
      "epoch": 0.7070425155828449,
      "eval_loss": 0.6429811716079712,
      "eval_runtime": 26.563,
      "eval_samples_per_second": 4.781,
      "eval_steps_per_second": 4.781,
      "step": 1900
    },
    {
      "epoch": 0.7089031537817472,
      "grad_norm": 0.013226517476141453,
      "learning_rate": 4.732429794558887e-06,
      "loss": 0.6391,
      "step": 1905
    },
    {
      "epoch": 0.7107637919806493,
      "grad_norm": 0.012964668683707714,
      "learning_rate": 4.677322065565039e-06,
      "loss": 0.6768,
      "step": 1910
    },
    {
      "epoch": 0.7126244301795516,
      "grad_norm": 0.01224041823297739,
      "learning_rate": 4.622438960398234e-06,
      "loss": 0.645,
      "step": 1915
    },
    {
      "epoch": 0.7144850683784538,
      "grad_norm": 0.013205330818891525,
      "learning_rate": 4.567782795195816e-06,
      "loss": 0.6051,
      "step": 1920
    },
    {
      "epoch": 0.7163457065773561,
      "grad_norm": 0.01323428563773632,
      "learning_rate": 4.5133558765179576e-06,
      "loss": 0.6113,
      "step": 1925
    },
    {
      "epoch": 0.7182063447762582,
      "grad_norm": 0.012786868028342724,
      "learning_rate": 4.459160501250358e-06,
      "loss": 0.6677,
      "step": 1930
    },
    {
      "epoch": 0.7200669829751605,
      "grad_norm": 0.010466697625815868,
      "learning_rate": 4.405198956507272e-06,
      "loss": 0.586,
      "step": 1935
    },
    {
      "epoch": 0.7219276211740627,
      "grad_norm": 0.012410039082169533,
      "learning_rate": 4.35147351953501e-06,
      "loss": 0.6111,
      "step": 1940
    },
    {
      "epoch": 0.723788259372965,
      "grad_norm": 0.013155271299183369,
      "learning_rate": 4.297986457615836e-06,
      "loss": 0.6404,
      "step": 1945
    },
    {
      "epoch": 0.7256488975718671,
      "grad_norm": 0.01472384575754404,
      "learning_rate": 4.244740027972275e-06,
      "loss": 0.6553,
      "step": 1950
    },
    {
      "epoch": 0.7275095357707694,
      "grad_norm": 0.012972739525139332,
      "learning_rate": 4.191736477671864e-06,
      "loss": 0.6613,
      "step": 1955
    },
    {
      "epoch": 0.7293701739696716,
      "grad_norm": 0.013699792325496674,
      "learning_rate": 4.138978043532332e-06,
      "loss": 0.6178,
      "step": 1960
    },
    {
      "epoch": 0.7312308121685738,
      "grad_norm": 0.012072841636836529,
      "learning_rate": 4.086466952027171e-06,
      "loss": 0.5865,
      "step": 1965
    },
    {
      "epoch": 0.733091450367476,
      "grad_norm": 0.013510013930499554,
      "learning_rate": 4.034205419191709e-06,
      "loss": 0.6387,
      "step": 1970
    },
    {
      "epoch": 0.7349520885663783,
      "grad_norm": 0.01405192632228136,
      "learning_rate": 3.982195650529583e-06,
      "loss": 0.5677,
      "step": 1975
    },
    {
      "epoch": 0.7368127267652805,
      "grad_norm": 0.012013067491352558,
      "learning_rate": 3.930439840919652e-06,
      "loss": 0.5868,
      "step": 1980
    },
    {
      "epoch": 0.7386733649641827,
      "grad_norm": 0.01283415500074625,
      "learning_rate": 3.878940174523371e-06,
      "loss": 0.597,
      "step": 1985
    },
    {
      "epoch": 0.7405340031630849,
      "grad_norm": 0.013413486070930958,
      "learning_rate": 3.827698824692643e-06,
      "loss": 0.6074,
      "step": 1990
    },
    {
      "epoch": 0.7423946413619872,
      "grad_norm": 0.013137887232005596,
      "learning_rate": 3.776717953878064e-06,
      "loss": 0.6599,
      "step": 1995
    },
    {
      "epoch": 0.7442552795608894,
      "grad_norm": 0.013435564935207367,
      "learning_rate": 3.725999713537689e-06,
      "loss": 0.6191,
      "step": 2000
    },
    {
      "epoch": 0.7442552795608894,
      "eval_loss": 0.6428595185279846,
      "eval_runtime": 26.6061,
      "eval_samples_per_second": 4.773,
      "eval_steps_per_second": 4.773,
      "step": 2000
    },
    {
      "epoch": 0.7461159177597916,
      "grad_norm": 0.013417122885584831,
      "learning_rate": 3.6755462440462288e-06,
      "loss": 0.6012,
      "step": 2005
    },
    {
      "epoch": 0.7479765559586938,
      "grad_norm": 0.01378430612385273,
      "learning_rate": 3.625359674604725e-06,
      "loss": 0.607,
      "step": 2010
    },
    {
      "epoch": 0.7498371941575961,
      "grad_norm": 0.012614194303750992,
      "learning_rate": 3.5754421231506953e-06,
      "loss": 0.6364,
      "step": 2015
    },
    {
      "epoch": 0.7516978323564982,
      "grad_norm": 0.013113941997289658,
      "learning_rate": 3.5257956962687545e-06,
      "loss": 0.6148,
      "step": 2020
    },
    {
      "epoch": 0.7535584705554005,
      "grad_norm": 0.012877865694463253,
      "learning_rate": 3.476422489101713e-06,
      "loss": 0.6369,
      "step": 2025
    },
    {
      "epoch": 0.7554191087543027,
      "grad_norm": 0.013343026861548424,
      "learning_rate": 3.427324585262156e-06,
      "loss": 0.6257,
      "step": 2030
    },
    {
      "epoch": 0.757279746953205,
      "grad_norm": 0.01313395518809557,
      "learning_rate": 3.3785040567445282e-06,
      "loss": 0.6301,
      "step": 2035
    },
    {
      "epoch": 0.7591403851521071,
      "grad_norm": 0.013584131374955177,
      "learning_rate": 3.329962963837661e-06,
      "loss": 0.6244,
      "step": 2040
    },
    {
      "epoch": 0.7610010233510094,
      "grad_norm": 0.011685586534440517,
      "learning_rate": 3.281703355037854e-06,
      "loss": 0.5818,
      "step": 2045
    },
    {
      "epoch": 0.7628616615499116,
      "grad_norm": 0.013454140163958073,
      "learning_rate": 3.233727266962425e-06,
      "loss": 0.6151,
      "step": 2050
    },
    {
      "epoch": 0.7647222997488139,
      "grad_norm": 0.012276149354875088,
      "learning_rate": 3.186036724263748e-06,
      "loss": 0.621,
      "step": 2055
    },
    {
      "epoch": 0.766582937947716,
      "grad_norm": 0.013796020299196243,
      "learning_rate": 3.138633739543805e-06,
      "loss": 0.6466,
      "step": 2060
    },
    {
      "epoch": 0.7684435761466183,
      "grad_norm": 0.013075289316475391,
      "learning_rate": 3.0915203132692805e-06,
      "loss": 0.6116,
      "step": 2065
    },
    {
      "epoch": 0.7703042143455205,
      "grad_norm": 0.014096383936703205,
      "learning_rate": 3.0446984336871144e-06,
      "loss": 0.5877,
      "step": 2070
    },
    {
      "epoch": 0.7721648525444227,
      "grad_norm": 0.013074836693704128,
      "learning_rate": 2.998170076740601e-06,
      "loss": 0.5829,
      "step": 2075
    },
    {
      "epoch": 0.7740254907433249,
      "grad_norm": 0.01671191304922104,
      "learning_rate": 2.951937205986004e-06,
      "loss": 0.6439,
      "step": 2080
    },
    {
      "epoch": 0.7758861289422272,
      "grad_norm": 0.013207321055233479,
      "learning_rate": 2.9060017725096943e-06,
      "loss": 0.621,
      "step": 2085
    },
    {
      "epoch": 0.7777467671411294,
      "grad_norm": 0.014768741093575954,
      "learning_rate": 2.8603657148458053e-06,
      "loss": 0.6272,
      "step": 2090
    },
    {
      "epoch": 0.7796074053400316,
      "grad_norm": 0.015293040312826633,
      "learning_rate": 2.8150309588944304e-06,
      "loss": 0.6388,
      "step": 2095
    },
    {
      "epoch": 0.7814680435389338,
      "grad_norm": 0.01271377969533205,
      "learning_rate": 2.769999417840341e-06,
      "loss": 0.6249,
      "step": 2100
    },
    {
      "epoch": 0.7814680435389338,
      "eval_loss": 0.6426796913146973,
      "eval_runtime": 26.6067,
      "eval_samples_per_second": 4.773,
      "eval_steps_per_second": 4.773,
      "step": 2100
    },
    {
      "epoch": 0.7833286817378361,
      "grad_norm": 0.013834511861205101,
      "learning_rate": 2.7252729920722564e-06,
      "loss": 0.6254,
      "step": 2105
    },
    {
      "epoch": 0.7851893199367384,
      "grad_norm": 0.020366957411170006,
      "learning_rate": 2.680853569102633e-06,
      "loss": 0.632,
      "step": 2110
    },
    {
      "epoch": 0.7870499581356405,
      "grad_norm": 0.012656195089221,
      "learning_rate": 2.6367430234880286e-06,
      "loss": 0.6274,
      "step": 2115
    },
    {
      "epoch": 0.7889105963345427,
      "grad_norm": 0.01308165118098259,
      "learning_rate": 2.5929432167499658e-06,
      "loss": 0.6457,
      "step": 2120
    },
    {
      "epoch": 0.790771234533445,
      "grad_norm": 0.012838170863687992,
      "learning_rate": 2.5494559972963928e-06,
      "loss": 0.6436,
      "step": 2125
    },
    {
      "epoch": 0.7926318727323471,
      "grad_norm": 0.012966095469892025,
      "learning_rate": 2.5062832003436833e-06,
      "loss": 0.6449,
      "step": 2130
    },
    {
      "epoch": 0.7944925109312494,
      "grad_norm": 0.013781096786260605,
      "learning_rate": 2.463426647839173e-06,
      "loss": 0.584,
      "step": 2135
    },
    {
      "epoch": 0.7963531491301516,
      "grad_norm": 0.012445746921002865,
      "learning_rate": 2.420888148384265e-06,
      "loss": 0.6397,
      "step": 2140
    },
    {
      "epoch": 0.7982137873290539,
      "grad_norm": 0.012599524110555649,
      "learning_rate": 2.378669497158138e-06,
      "loss": 0.5974,
      "step": 2145
    },
    {
      "epoch": 0.800074425527956,
      "grad_norm": 0.011499716900289059,
      "learning_rate": 2.3367724758419495e-06,
      "loss": 0.5552,
      "step": 2150
    },
    {
      "epoch": 0.8019350637268583,
      "grad_norm": 0.012771397829055786,
      "learning_rate": 2.2951988525436695e-06,
      "loss": 0.666,
      "step": 2155
    },
    {
      "epoch": 0.8037957019257606,
      "grad_norm": 0.013215843588113785,
      "learning_rate": 2.2539503817234553e-06,
      "loss": 0.5925,
      "step": 2160
    },
    {
      "epoch": 0.8056563401246628,
      "grad_norm": 0.014011417515575886,
      "learning_rate": 2.2130288041196135e-06,
      "loss": 0.6216,
      "step": 2165
    },
    {
      "epoch": 0.807516978323565,
      "grad_norm": 0.014169846661388874,
      "learning_rate": 2.1724358466751394e-06,
      "loss": 0.625,
      "step": 2170
    },
    {
      "epoch": 0.8093776165224672,
      "grad_norm": 0.013908619992434978,
      "learning_rate": 2.132173222464834e-06,
      "loss": 0.6641,
      "step": 2175
    },
    {
      "epoch": 0.8112382547213695,
      "grad_norm": 0.01255475077778101,
      "learning_rate": 2.092242630623016e-06,
      "loss": 0.6135,
      "step": 2180
    },
    {
      "epoch": 0.8130988929202716,
      "grad_norm": 0.01320129819214344,
      "learning_rate": 2.0526457562718074e-06,
      "loss": 0.5893,
      "step": 2185
    },
    {
      "epoch": 0.8149595311191739,
      "grad_norm": 0.012163982726633549,
      "learning_rate": 2.013384270450036e-06,
      "loss": 0.6552,
      "step": 2190
    },
    {
      "epoch": 0.8168201693180761,
      "grad_norm": 0.012990654446184635,
      "learning_rate": 1.974459830042691e-06,
      "loss": 0.6309,
      "step": 2195
    },
    {
      "epoch": 0.8186808075169784,
      "grad_norm": 0.015373739413917065,
      "learning_rate": 1.9358740777110154e-06,
      "loss": 0.6761,
      "step": 2200
    },
    {
      "epoch": 0.8186808075169784,
      "eval_loss": 0.6425639390945435,
      "eval_runtime": 26.5817,
      "eval_samples_per_second": 4.778,
      "eval_steps_per_second": 4.778,
      "step": 2200
    },
    {
      "epoch": 0.8205414457158805,
      "grad_norm": 0.012748058885335922,
      "learning_rate": 1.8976286418231916e-06,
      "loss": 0.6313,
      "step": 2205
    },
    {
      "epoch": 0.8224020839147828,
      "grad_norm": 0.014108446426689625,
      "learning_rate": 1.8597251363856061e-06,
      "loss": 0.5997,
      "step": 2210
    },
    {
      "epoch": 0.824262722113685,
      "grad_norm": 0.0133186811581254,
      "learning_rate": 1.8221651609747337e-06,
      "loss": 0.6732,
      "step": 2215
    },
    {
      "epoch": 0.8261233603125873,
      "grad_norm": 0.012968887574970722,
      "learning_rate": 1.7849503006696566e-06,
      "loss": 0.6129,
      "step": 2220
    },
    {
      "epoch": 0.8279839985114894,
      "grad_norm": 0.012473355047404766,
      "learning_rate": 1.7480821259851488e-06,
      "loss": 0.6131,
      "step": 2225
    },
    {
      "epoch": 0.8298446367103917,
      "grad_norm": 0.012497167102992535,
      "learning_rate": 1.7115621928054105e-06,
      "loss": 0.6199,
      "step": 2230
    },
    {
      "epoch": 0.8317052749092939,
      "grad_norm": 0.014741111546754837,
      "learning_rate": 1.6753920423184022e-06,
      "loss": 0.6159,
      "step": 2235
    },
    {
      "epoch": 0.8335659131081962,
      "grad_norm": 0.013057067058980465,
      "learning_rate": 1.6395732009508058e-06,
      "loss": 0.5898,
      "step": 2240
    },
    {
      "epoch": 0.8354265513070983,
      "grad_norm": 0.014299440197646618,
      "learning_rate": 1.60410718030361e-06,
      "loss": 0.6486,
      "step": 2245
    },
    {
      "epoch": 0.8372871895060006,
      "grad_norm": 0.013178296387195587,
      "learning_rate": 1.568995477088323e-06,
      "loss": 0.6254,
      "step": 2250
    },
    {
      "epoch": 0.8391478277049028,
      "grad_norm": 0.013072814792394638,
      "learning_rate": 1.5342395730637904e-06,
      "loss": 0.6385,
      "step": 2255
    },
    {
      "epoch": 0.841008465903805,
      "grad_norm": 0.014187455177307129,
      "learning_rate": 1.4998409349736841e-06,
      "loss": 0.6458,
      "step": 2260
    },
    {
      "epoch": 0.8428691041027072,
      "grad_norm": 0.011844666674733162,
      "learning_rate": 1.4658010144846001e-06,
      "loss": 0.6556,
      "step": 2265
    },
    {
      "epoch": 0.8447297423016095,
      "grad_norm": 0.013700997456908226,
      "learning_rate": 1.432121248124786e-06,
      "loss": 0.6381,
      "step": 2270
    },
    {
      "epoch": 0.8465903805005117,
      "grad_norm": 0.014204096049070358,
      "learning_rate": 1.3988030572235212e-06,
      "loss": 0.624,
      "step": 2275
    },
    {
      "epoch": 0.8484510186994139,
      "grad_norm": 0.013039126060903072,
      "learning_rate": 1.3658478478511416e-06,
      "loss": 0.593,
      "step": 2280
    },
    {
      "epoch": 0.8503116568983161,
      "grad_norm": 0.014153816737234592,
      "learning_rate": 1.333257010759702e-06,
      "loss": 0.5909,
      "step": 2285
    },
    {
      "epoch": 0.8521722950972184,
      "grad_norm": 0.012623702175915241,
      "learning_rate": 1.3010319213242762e-06,
      "loss": 0.6006,
      "step": 2290
    },
    {
      "epoch": 0.8540329332961206,
      "grad_norm": 0.013684497214853764,
      "learning_rate": 1.2691739394849089e-06,
      "loss": 0.6112,
      "step": 2295
    },
    {
      "epoch": 0.8558935714950228,
      "grad_norm": 0.012923210859298706,
      "learning_rate": 1.2376844096892526e-06,
      "loss": 0.6239,
      "step": 2300
    },
    {
      "epoch": 0.8558935714950228,
      "eval_loss": 0.6425209641456604,
      "eval_runtime": 26.5582,
      "eval_samples_per_second": 4.782,
      "eval_steps_per_second": 4.782,
      "step": 2300
    },
    {
      "epoch": 0.857754209693925,
      "grad_norm": 0.01325127761811018,
      "learning_rate": 1.2065646608357972e-06,
      "loss": 0.6537,
      "step": 2305
    },
    {
      "epoch": 0.8596148478928273,
      "grad_norm": 0.01215402316302061,
      "learning_rate": 1.1758160062178093e-06,
      "loss": 0.634,
      "step": 2310
    },
    {
      "epoch": 0.8614754860917294,
      "grad_norm": 0.015918321907520294,
      "learning_rate": 1.1454397434679022e-06,
      "loss": 0.6296,
      "step": 2315
    },
    {
      "epoch": 0.8633361242906317,
      "grad_norm": 0.011971482075750828,
      "learning_rate": 1.1154371545032738e-06,
      "loss": 0.5983,
      "step": 2320
    },
    {
      "epoch": 0.8651967624895339,
      "grad_norm": 0.01542737614363432,
      "learning_rate": 1.0858095054716111e-06,
      "loss": 0.6468,
      "step": 2325
    },
    {
      "epoch": 0.8670574006884362,
      "grad_norm": 0.012823596596717834,
      "learning_rate": 1.0565580466976566e-06,
      "loss": 0.6222,
      "step": 2330
    },
    {
      "epoch": 0.8689180388873383,
      "grad_norm": 0.012822597287595272,
      "learning_rate": 1.027684012630441e-06,
      "loss": 0.6385,
      "step": 2335
    },
    {
      "epoch": 0.8707786770862406,
      "grad_norm": 0.014365943148732185,
      "learning_rate": 9.991886217911851e-07,
      "loss": 0.6541,
      "step": 2340
    },
    {
      "epoch": 0.8726393152851428,
      "grad_norm": 0.013635417446494102,
      "learning_rate": 9.710730767218913e-07,
      "loss": 0.6323,
      "step": 2345
    },
    {
      "epoch": 0.8744999534840451,
      "grad_norm": 0.012839280068874359,
      "learning_rate": 9.433385639345705e-07,
      "loss": 0.5944,
      "step": 2350
    },
    {
      "epoch": 0.8763605916829472,
      "grad_norm": 0.0141257019713521,
      "learning_rate": 9.159862538611908e-07,
      "loss": 0.6632,
      "step": 2355
    },
    {
      "epoch": 0.8782212298818495,
      "grad_norm": 0.013571621850132942,
      "learning_rate": 8.890173008042768e-07,
      "loss": 0.6086,
      "step": 2360
    },
    {
      "epoch": 0.8800818680807517,
      "grad_norm": 0.01433682069182396,
      "learning_rate": 8.624328428881945e-07,
      "loss": 0.6482,
      "step": 2365
    },
    {
      "epoch": 0.8819425062796539,
      "grad_norm": 0.014090972021222115,
      "learning_rate": 8.36234002011117e-07,
      "loss": 0.6335,
      "step": 2370
    },
    {
      "epoch": 0.8838031444785561,
      "grad_norm": 0.013004067353904247,
      "learning_rate": 8.10421883797694e-07,
      "loss": 0.6145,
      "step": 2375
    },
    {
      "epoch": 0.8856637826774584,
      "grad_norm": 0.01391025260090828,
      "learning_rate": 7.849975775523777e-07,
      "loss": 0.6415,
      "step": 2380
    },
    {
      "epoch": 0.8875244208763606,
      "grad_norm": 0.012305272743105888,
      "learning_rate": 7.599621562134596e-07,
      "loss": 0.6462,
      "step": 2385
    },
    {
      "epoch": 0.8893850590752628,
      "grad_norm": 0.013947544619441032,
      "learning_rate": 7.35316676307789e-07,
      "loss": 0.6266,
      "step": 2390
    },
    {
      "epoch": 0.891245697274165,
      "grad_norm": 0.016055511310696602,
      "learning_rate": 7.110621779061889e-07,
      "loss": 0.6501,
      "step": 2395
    },
    {
      "epoch": 0.8931063354730673,
      "grad_norm": 0.01238927897065878,
      "learning_rate": 6.871996845795581e-07,
      "loss": 0.5895,
      "step": 2400
    },
    {
      "epoch": 0.8931063354730673,
      "eval_loss": 0.6424703598022461,
      "eval_runtime": 26.6049,
      "eval_samples_per_second": 4.774,
      "eval_steps_per_second": 4.774,
      "step": 2400
    },
    {
      "epoch": 0.8949669736719695,
      "grad_norm": 0.012343904934823513,
      "learning_rate": 6.637302033556891e-07,
      "loss": 0.6105,
      "step": 2405
    },
    {
      "epoch": 0.8968276118708717,
      "grad_norm": 0.01289752684533596,
      "learning_rate": 6.40654724676748e-07,
      "loss": 0.6265,
      "step": 2410
    },
    {
      "epoch": 0.8986882500697739,
      "grad_norm": 0.013334677554666996,
      "learning_rate": 6.179742223574936e-07,
      "loss": 0.6261,
      "step": 2415
    },
    {
      "epoch": 0.9005488882686762,
      "grad_norm": 0.013512643985450268,
      "learning_rate": 5.956896535441803e-07,
      "loss": 0.5797,
      "step": 2420
    },
    {
      "epoch": 0.9024095264675783,
      "grad_norm": 0.01403987966477871,
      "learning_rate": 5.738019586741573e-07,
      "loss": 0.616,
      "step": 2425
    },
    {
      "epoch": 0.9042701646664806,
      "grad_norm": 0.013224196620285511,
      "learning_rate": 5.523120614361821e-07,
      "loss": 0.608,
      "step": 2430
    },
    {
      "epoch": 0.9061308028653828,
      "grad_norm": 0.01317799836397171,
      "learning_rate": 5.312208687314502e-07,
      "loss": 0.6206,
      "step": 2435
    },
    {
      "epoch": 0.9079914410642851,
      "grad_norm": 0.01357815321534872,
      "learning_rate": 5.105292706353093e-07,
      "loss": 0.6759,
      "step": 2440
    },
    {
      "epoch": 0.9098520792631872,
      "grad_norm": 0.012360199354588985,
      "learning_rate": 4.902381403597046e-07,
      "loss": 0.6182,
      "step": 2445
    },
    {
      "epoch": 0.9117127174620895,
      "grad_norm": 0.012978832237422466,
      "learning_rate": 4.703483342163262e-07,
      "loss": 0.5892,
      "step": 2450
    },
    {
      "epoch": 0.9135733556609917,
      "grad_norm": 0.013323403894901276,
      "learning_rate": 4.5086069158047143e-07,
      "loss": 0.636,
      "step": 2455
    },
    {
      "epoch": 0.915433993859894,
      "grad_norm": 0.013127562589943409,
      "learning_rate": 4.3177603485562327e-07,
      "loss": 0.5847,
      "step": 2460
    },
    {
      "epoch": 0.9172946320587961,
      "grad_norm": 0.012560434639453888,
      "learning_rate": 4.1309516943874196e-07,
      "loss": 0.6073,
      "step": 2465
    },
    {
      "epoch": 0.9191552702576984,
      "grad_norm": 0.01229447964578867,
      "learning_rate": 3.9481888368627764e-07,
      "loss": 0.5846,
      "step": 2470
    },
    {
      "epoch": 0.9210159084566006,
      "grad_norm": 0.012554515153169632,
      "learning_rate": 3.7694794888090025e-07,
      "loss": 0.6186,
      "step": 2475
    },
    {
      "epoch": 0.9228765466555028,
      "grad_norm": 0.011958773247897625,
      "learning_rate": 3.594831191989523e-07,
      "loss": 0.6217,
      "step": 2480
    },
    {
      "epoch": 0.924737184854405,
      "grad_norm": 0.01367896143347025,
      "learning_rate": 3.424251316786165e-07,
      "loss": 0.6572,
      "step": 2485
    },
    {
      "epoch": 0.9265978230533073,
      "grad_norm": 0.013063084334135056,
      "learning_rate": 3.2577470618881726e-07,
      "loss": 0.5973,
      "step": 2490
    },
    {
      "epoch": 0.9284584612522095,
      "grad_norm": 0.012596881948411465,
      "learning_rate": 3.095325453988385e-07,
      "loss": 0.6153,
      "step": 2495
    },
    {
      "epoch": 0.9303190994511117,
      "grad_norm": 0.013333328068256378,
      "learning_rate": 2.9369933474867496e-07,
      "loss": 0.6337,
      "step": 2500
    },
    {
      "epoch": 0.9303190994511117,
      "eval_loss": 0.6424322128295898,
      "eval_runtime": 26.5954,
      "eval_samples_per_second": 4.775,
      "eval_steps_per_second": 4.775,
      "step": 2500
    },
    {
      "epoch": 0.9321797376500139,
      "grad_norm": 0.012263627722859383,
      "learning_rate": 2.7827574242009434e-07,
      "loss": 0.6258,
      "step": 2505
    },
    {
      "epoch": 0.9340403758489162,
      "grad_norm": 0.01295017171651125,
      "learning_rate": 2.632624193084499e-07,
      "loss": 0.6071,
      "step": 2510
    },
    {
      "epoch": 0.9359010140478184,
      "grad_norm": 0.013952870853245258,
      "learning_rate": 2.48659998995211e-07,
      "loss": 0.6594,
      "step": 2515
    },
    {
      "epoch": 0.9377616522467206,
      "grad_norm": 0.01351676881313324,
      "learning_rate": 2.344690977212205e-07,
      "loss": 0.6072,
      "step": 2520
    },
    {
      "epoch": 0.9396222904456228,
      "grad_norm": 0.012661050073802471,
      "learning_rate": 2.2069031436068643e-07,
      "loss": 0.6152,
      "step": 2525
    },
    {
      "epoch": 0.9414829286445251,
      "grad_norm": 0.013936568051576614,
      "learning_rate": 2.0732423039591998e-07,
      "loss": 0.6149,
      "step": 2530
    },
    {
      "epoch": 0.9433435668434274,
      "grad_norm": 0.012601389549672604,
      "learning_rate": 1.9437140989278624e-07,
      "loss": 0.6205,
      "step": 2535
    },
    {
      "epoch": 0.9452042050423295,
      "grad_norm": 0.011978083290159702,
      "learning_rate": 1.8183239947690112e-07,
      "loss": 0.6252,
      "step": 2540
    },
    {
      "epoch": 0.9470648432412317,
      "grad_norm": 0.012075323611497879,
      "learning_rate": 1.6970772831056637e-07,
      "loss": 0.6438,
      "step": 2545
    },
    {
      "epoch": 0.948925481440134,
      "grad_norm": 0.012837120331823826,
      "learning_rate": 1.5799790807043857e-07,
      "loss": 0.6304,
      "step": 2550
    },
    {
      "epoch": 0.9507861196390361,
      "grad_norm": 0.012269029393792152,
      "learning_rate": 1.467034329259287e-07,
      "loss": 0.5938,
      "step": 2555
    },
    {
      "epoch": 0.9526467578379384,
      "grad_norm": 0.01306453812867403,
      "learning_rate": 1.358247795183587e-07,
      "loss": 0.5981,
      "step": 2560
    },
    {
      "epoch": 0.9545073960368406,
      "grad_norm": 0.012171389535069466,
      "learning_rate": 1.2536240694083658e-07,
      "loss": 0.5599,
      "step": 2565
    },
    {
      "epoch": 0.9563680342357429,
      "grad_norm": 0.013338044285774231,
      "learning_rate": 1.1531675671888621e-07,
      "loss": 0.6366,
      "step": 2570
    },
    {
      "epoch": 0.958228672434645,
      "grad_norm": 0.012387475930154324,
      "learning_rate": 1.0568825279181572e-07,
      "loss": 0.6376,
      "step": 2575
    },
    {
      "epoch": 0.9600893106335473,
      "grad_norm": 0.013054094277322292,
      "learning_rate": 9.647730149482614e-08,
      "loss": 0.6386,
      "step": 2580
    },
    {
      "epoch": 0.9619499488324496,
      "grad_norm": 0.01319506112486124,
      "learning_rate": 8.768429154185853e-08,
      "loss": 0.6067,
      "step": 2585
    },
    {
      "epoch": 0.9638105870313518,
      "grad_norm": 0.013576803728938103,
      "learning_rate": 7.930959400919924e-08,
      "loss": 0.6258,
      "step": 2590
    },
    {
      "epoch": 0.965671225230254,
      "grad_norm": 0.013341645710170269,
      "learning_rate": 7.135356231981028e-08,
      "loss": 0.6344,
      "step": 2595
    },
    {
      "epoch": 0.9675318634291562,
      "grad_norm": 0.013685373589396477,
      "learning_rate": 6.381653222842011e-08,
      "loss": 0.6144,
      "step": 2600
    },
    {
      "epoch": 0.9675318634291562,
      "eval_loss": 0.642424464225769,
      "eval_runtime": 26.5716,
      "eval_samples_per_second": 4.78,
      "eval_steps_per_second": 4.78,
      "step": 2600
    },
    {
      "epoch": 0.9693925016280585,
      "grad_norm": 0.018206071108579636,
      "learning_rate": 5.6698821807354975e-08,
      "loss": 0.5716,
      "step": 2605
    },
    {
      "epoch": 0.9712531398269606,
      "grad_norm": 0.014082850888371468,
      "learning_rate": 5.000073143310969e-08,
      "loss": 0.641,
      "step": 2610
    },
    {
      "epoch": 0.9731137780258629,
      "grad_norm": 0.01310745719820261,
      "learning_rate": 4.3722543773681016e-08,
      "loss": 0.6206,
      "step": 2615
    },
    {
      "epoch": 0.9749744162247651,
      "grad_norm": 0.013652559369802475,
      "learning_rate": 3.7864523776628414e-08,
      "loss": 0.6405,
      "step": 2620
    },
    {
      "epoch": 0.9768350544236674,
      "grad_norm": 0.01430275198072195,
      "learning_rate": 3.242691865790071e-08,
      "loss": 0.6191,
      "step": 2625
    },
    {
      "epoch": 0.9786956926225695,
      "grad_norm": 0.013149100355803967,
      "learning_rate": 2.7409957891397775e-08,
      "loss": 0.6676,
      "step": 2630
    },
    {
      "epoch": 0.9805563308214718,
      "grad_norm": 0.01429106667637825,
      "learning_rate": 2.2813853199292745e-08,
      "loss": 0.656,
      "step": 2635
    },
    {
      "epoch": 0.982416969020374,
      "grad_norm": 0.012532561086118221,
      "learning_rate": 1.8638798543090253e-08,
      "loss": 0.6065,
      "step": 2640
    },
    {
      "epoch": 0.9842776072192763,
      "grad_norm": 0.01296111661940813,
      "learning_rate": 1.4884970115444097e-08,
      "loss": 0.604,
      "step": 2645
    },
    {
      "epoch": 0.9861382454181784,
      "grad_norm": 0.012804310768842697,
      "learning_rate": 1.1552526332723191e-08,
      "loss": 0.6236,
      "step": 2650
    },
    {
      "epoch": 0.9879988836170807,
      "grad_norm": 0.012863220646977425,
      "learning_rate": 8.641607828324682e-09,
      "loss": 0.6245,
      "step": 2655
    },
    {
      "epoch": 0.9898595218159829,
      "grad_norm": 0.013312343508005142,
      "learning_rate": 6.152337446736489e-09,
      "loss": 0.6561,
      "step": 2660
    },
    {
      "epoch": 0.991720160014885,
      "grad_norm": 0.011470803059637547,
      "learning_rate": 4.0848202383581e-09,
      "loss": 0.5847,
      "step": 2665
    },
    {
      "epoch": 0.9935807982137873,
      "grad_norm": 0.013984150253236294,
      "learning_rate": 2.4391434550652403e-09,
      "loss": 0.5881,
      "step": 2670
    },
    {
      "epoch": 0.9954414364126896,
      "grad_norm": 0.012489933520555496,
      "learning_rate": 1.2153765465250378e-09,
      "loss": 0.6653,
      "step": 2675
    },
    {
      "epoch": 0.9973020746115918,
      "grad_norm": 0.012180610559880733,
      "learning_rate": 4.1357115726947674e-10,
      "loss": 0.629,
      "step": 2680
    },
    {
      "epoch": 0.999162712810494,
      "grad_norm": 0.014413848519325256,
      "learning_rate": 3.376112451158875e-11,
      "loss": 0.6525,
      "step": 2685
    },
    {
      "epoch": 0.9999069680900549,
      "step": 2687,
      "total_flos": 8.247315480402985e+17,
      "train_loss": 0.638783668202896,
      "train_runtime": 14784.0018,
      "train_samples_per_second": 1.454,
      "train_steps_per_second": 0.182
    }
  ],
  "logging_steps": 5,
  "max_steps": 2687,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.247315480402985e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}