{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 27.069486404833835,
  "eval_steps": 100,
  "global_step": 4200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06445115810674723,
      "grad_norm": 8.751441955566406,
      "learning_rate": 5.154639175257732e-07,
      "loss": 8.7908,
      "step": 10
    },
    {
      "epoch": 0.12890231621349446,
      "grad_norm": 5.980831146240234,
      "learning_rate": 1.0309278350515464e-06,
      "loss": 8.4297,
      "step": 20
    },
    {
      "epoch": 0.1933534743202417,
      "grad_norm": 3.555551528930664,
      "learning_rate": 1.5463917525773197e-06,
      "loss": 7.9847,
      "step": 30
    },
    {
      "epoch": 0.2578046324269889,
      "grad_norm": 2.1481900215148926,
      "learning_rate": 2.061855670103093e-06,
      "loss": 7.5452,
      "step": 40
    },
    {
      "epoch": 0.32225579053373615,
      "grad_norm": 1.3654990196228027,
      "learning_rate": 2.577319587628866e-06,
      "loss": 7.1644,
      "step": 50
    },
    {
      "epoch": 0.3867069486404834,
      "grad_norm": 1.294124960899353,
      "learning_rate": 3.0927835051546395e-06,
      "loss": 6.8552,
      "step": 60
    },
    {
      "epoch": 0.4511581067472306,
      "grad_norm": 1.3459229469299316,
      "learning_rate": 3.6082474226804126e-06,
      "loss": 6.5137,
      "step": 70
    },
    {
      "epoch": 0.5156092648539778,
      "grad_norm": 1.345466136932373,
      "learning_rate": 4.123711340206186e-06,
      "loss": 6.2217,
      "step": 80
    },
    {
      "epoch": 0.5800604229607251,
      "grad_norm": 1.0641965866088867,
      "learning_rate": 4.639175257731959e-06,
      "loss": 5.999,
      "step": 90
    },
    {
      "epoch": 0.6445115810674723,
      "grad_norm": 1.5514322519302368,
      "learning_rate": 5.154639175257732e-06,
      "loss": 5.8344,
      "step": 100
    },
    {
      "epoch": 0.6445115810674723,
      "eval_loss": 5.791144847869873,
      "eval_runtime": 20.2064,
      "eval_samples_per_second": 79.43,
      "eval_steps_per_second": 4.998,
      "step": 100
    },
    {
      "epoch": 0.7089627391742196,
      "grad_norm": 9.654533386230469,
      "learning_rate": 5.670103092783505e-06,
      "loss": 5.7513,
      "step": 110
    },
    {
      "epoch": 0.7734138972809668,
      "grad_norm": 2.767559289932251,
      "learning_rate": 6.185567010309279e-06,
      "loss": 5.6479,
      "step": 120
    },
    {
      "epoch": 0.837865055387714,
      "grad_norm": 1.2786972522735596,
      "learning_rate": 6.701030927835052e-06,
      "loss": 5.5542,
      "step": 130
    },
    {
      "epoch": 0.9023162134944612,
      "grad_norm": 1.3562631607055664,
      "learning_rate": 7.216494845360825e-06,
      "loss": 5.4406,
      "step": 140
    },
    {
      "epoch": 0.9667673716012085,
      "grad_norm": 3.51896333694458,
      "learning_rate": 7.731958762886599e-06,
      "loss": 5.323,
      "step": 150
    },
    {
      "epoch": 1.0312185297079557,
      "grad_norm": 1.808701753616333,
      "learning_rate": 8.247422680412371e-06,
      "loss": 5.6701,
      "step": 160
    },
    {
      "epoch": 1.095669687814703,
      "grad_norm": 3.2831215858459473,
      "learning_rate": 8.762886597938146e-06,
      "loss": 5.1268,
      "step": 170
    },
    {
      "epoch": 1.1601208459214503,
      "grad_norm": 2.606133460998535,
      "learning_rate": 9.278350515463918e-06,
      "loss": 5.0758,
      "step": 180
    },
    {
      "epoch": 1.2245720040281973,
      "grad_norm": 1.7396525144577026,
      "learning_rate": 9.793814432989691e-06,
      "loss": 5.0134,
      "step": 190
    },
    {
      "epoch": 1.2890231621349446,
      "grad_norm": 0.7748392820358276,
      "learning_rate": 1.0309278350515464e-05,
      "loss": 4.9703,
      "step": 200
    },
    {
      "epoch": 1.2890231621349446,
      "eval_loss": 4.946072578430176,
      "eval_runtime": 19.5278,
      "eval_samples_per_second": 82.19,
      "eval_steps_per_second": 5.172,
      "step": 200
    },
    {
      "epoch": 1.353474320241692,
      "grad_norm": 2.377145290374756,
      "learning_rate": 1.0824742268041238e-05,
      "loss": 4.874,
      "step": 210
    },
    {
      "epoch": 1.417925478348439,
      "grad_norm": 1.2470804452896118,
      "learning_rate": 1.134020618556701e-05,
      "loss": 4.8473,
      "step": 220
    },
    {
      "epoch": 1.4823766364551862,
      "grad_norm": 2.26425838470459,
      "learning_rate": 1.1855670103092785e-05,
      "loss": 4.7739,
      "step": 230
    },
    {
      "epoch": 1.5468277945619335,
      "grad_norm": 1.238208532333374,
      "learning_rate": 1.2371134020618558e-05,
      "loss": 4.7524,
      "step": 240
    },
    {
      "epoch": 1.6112789526686808,
      "grad_norm": 0.9819815158843994,
      "learning_rate": 1.2886597938144332e-05,
      "loss": 4.6977,
      "step": 250
    },
    {
      "epoch": 1.675730110775428,
      "grad_norm": 1.37124502658844,
      "learning_rate": 1.3402061855670103e-05,
      "loss": 4.5812,
      "step": 260
    },
    {
      "epoch": 1.7401812688821754,
      "grad_norm": 1.0728330612182617,
      "learning_rate": 1.3917525773195878e-05,
      "loss": 4.5445,
      "step": 270
    },
    {
      "epoch": 1.8046324269889225,
      "grad_norm": 1.4190095663070679,
      "learning_rate": 1.443298969072165e-05,
      "loss": 4.4797,
      "step": 280
    },
    {
      "epoch": 1.8690835850956697,
      "grad_norm": 1.3823643922805786,
      "learning_rate": 1.4948453608247425e-05,
      "loss": 4.3833,
      "step": 290
    },
    {
      "epoch": 1.9335347432024168,
      "grad_norm": 1.6940747499465942,
      "learning_rate": 1.5463917525773197e-05,
      "loss": 4.2822,
      "step": 300
    },
    {
      "epoch": 1.9335347432024168,
      "eval_loss": 4.213192462921143,
      "eval_runtime": 19.7585,
      "eval_samples_per_second": 81.231,
      "eval_steps_per_second": 5.112,
      "step": 300
    },
    {
      "epoch": 1.997985901309164,
      "grad_norm": 1.6004289388656616,
      "learning_rate": 1.597938144329897e-05,
      "loss": 4.1608,
      "step": 310
    },
    {
      "epoch": 2.0624370594159114,
      "grad_norm": 2.5717203617095947,
      "learning_rate": 1.6494845360824743e-05,
      "loss": 4.392,
      "step": 320
    },
    {
      "epoch": 2.1268882175226587,
      "grad_norm": 1.9171743392944336,
      "learning_rate": 1.7010309278350517e-05,
      "loss": 3.9273,
      "step": 330
    },
    {
      "epoch": 2.191339375629406,
      "grad_norm": 2.1511971950531006,
      "learning_rate": 1.752577319587629e-05,
      "loss": 3.8527,
      "step": 340
    },
    {
      "epoch": 2.2557905337361532,
      "grad_norm": 1.949204683303833,
      "learning_rate": 1.8041237113402062e-05,
      "loss": 3.7843,
      "step": 350
    },
    {
      "epoch": 2.3202416918429005,
      "grad_norm": 1.6854971647262573,
      "learning_rate": 1.8556701030927837e-05,
      "loss": 3.7476,
      "step": 360
    },
    {
      "epoch": 2.3846928499496474,
      "grad_norm": 1.9099489450454712,
      "learning_rate": 1.907216494845361e-05,
      "loss": 3.6868,
      "step": 370
    },
    {
      "epoch": 2.4491440080563947,
      "grad_norm": 1.9068998098373413,
      "learning_rate": 1.9587628865979382e-05,
      "loss": 3.6684,
      "step": 380
    },
    {
      "epoch": 2.513595166163142,
      "grad_norm": 1.3064167499542236,
      "learning_rate": 1.9999996358015542e-05,
      "loss": 3.6028,
      "step": 390
    },
    {
      "epoch": 2.5780463242698892,
      "grad_norm": 1.7267900705337524,
      "learning_rate": 1.9999868888837957e-05,
      "loss": 3.5599,
      "step": 400
    },
    {
      "epoch": 2.5780463242698892,
      "eval_loss": 3.5509181022644043,
      "eval_runtime": 20.1957,
      "eval_samples_per_second": 79.472,
      "eval_steps_per_second": 5.001,
      "step": 400
    },
    {
      "epoch": 2.6424974823766365,
      "grad_norm": 2.2463932037353516,
      "learning_rate": 1.9999559323090132e-05,
      "loss": 3.5059,
      "step": 410
    },
    {
      "epoch": 2.706948640483384,
      "grad_norm": 2.046381950378418,
      "learning_rate": 1.9999067666409225e-05,
      "loss": 3.4585,
      "step": 420
    },
    {
      "epoch": 2.771399798590131,
      "grad_norm": 1.8293402194976807,
      "learning_rate": 1.9998393927748257e-05,
      "loss": 3.4393,
      "step": 430
    },
    {
      "epoch": 2.835850956696878,
      "grad_norm": 1.701651930809021,
      "learning_rate": 1.9997538119375938e-05,
      "loss": 3.3986,
      "step": 440
    },
    {
      "epoch": 2.900302114803625,
      "grad_norm": 1.8549060821533203,
      "learning_rate": 1.9996500256876447e-05,
      "loss": 3.3568,
      "step": 450
    },
    {
      "epoch": 2.9647532729103725,
      "grad_norm": 2.38415789604187,
      "learning_rate": 1.999528035914915e-05,
      "loss": 3.3115,
      "step": 460
    },
    {
      "epoch": 3.02920443101712,
      "grad_norm": 3.2878332138061523,
      "learning_rate": 1.9993878448408263e-05,
      "loss": 3.5941,
      "step": 470
    },
    {
      "epoch": 3.093655589123867,
      "grad_norm": 2.5062038898468018,
      "learning_rate": 1.999229455018243e-05,
      "loss": 3.2588,
      "step": 480
    },
    {
      "epoch": 3.1581067472306144,
      "grad_norm": 2.2001876831054688,
      "learning_rate": 1.9990528693314273e-05,
      "loss": 3.2381,
      "step": 490
    },
    {
      "epoch": 3.2225579053373616,
      "grad_norm": 2.3333001136779785,
      "learning_rate": 1.9988580909959864e-05,
      "loss": 3.2271,
      "step": 500
    },
    {
      "epoch": 3.2225579053373616,
      "eval_loss": 3.236539840698242,
      "eval_runtime": 19.5719,
      "eval_samples_per_second": 82.005,
      "eval_steps_per_second": 5.16,
      "step": 500
    },
    {
      "epoch": 3.287009063444109,
      "grad_norm": 2.109981060028076,
      "learning_rate": 1.9986451235588135e-05,
      "loss": 3.2038,
      "step": 510
    },
    {
      "epoch": 3.351460221550856,
      "grad_norm": 2.376877546310425,
      "learning_rate": 1.9984139708980228e-05,
      "loss": 3.1678,
      "step": 520
    },
    {
      "epoch": 3.415911379657603,
      "grad_norm": 2.0894505977630615,
      "learning_rate": 1.9981646372228813e-05,
      "loss": 3.1701,
      "step": 530
    },
    {
      "epoch": 3.4803625377643503,
      "grad_norm": 1.9505690336227417,
      "learning_rate": 1.997897127073728e-05,
      "loss": 3.1902,
      "step": 540
    },
    {
      "epoch": 3.5448136958710976,
      "grad_norm": 3.200566530227661,
      "learning_rate": 1.997611445321896e-05,
      "loss": 3.1286,
      "step": 550
    },
    {
      "epoch": 3.609264853977845,
      "grad_norm": 2.295381784439087,
      "learning_rate": 1.9973075971696195e-05,
      "loss": 3.1308,
      "step": 560
    },
    {
      "epoch": 3.673716012084592,
      "grad_norm": 2.323788642883301,
      "learning_rate": 1.9969855881499413e-05,
      "loss": 3.1093,
      "step": 570
    },
    {
      "epoch": 3.7381671701913395,
      "grad_norm": 1.9054023027420044,
      "learning_rate": 1.996645424126613e-05,
      "loss": 3.1029,
      "step": 580
    },
    {
      "epoch": 3.8026183282980868,
      "grad_norm": 2.854268789291382,
      "learning_rate": 1.996287111293986e-05,
      "loss": 3.0843,
      "step": 590
    },
    {
      "epoch": 3.8670694864048336,
      "grad_norm": 2.066882848739624,
      "learning_rate": 1.9959106561768988e-05,
      "loss": 3.0301,
      "step": 600
    },
    {
      "epoch": 3.8670694864048336,
      "eval_loss": 3.056886672973633,
      "eval_runtime": 20.2056,
      "eval_samples_per_second": 79.433,
      "eval_steps_per_second": 4.999,
      "step": 600
    },
    {
      "epoch": 3.931520644511581,
      "grad_norm": 2.220766544342041,
      "learning_rate": 1.9955160656305606e-05,
      "loss": 3.02,
      "step": 610
    },
    {
      "epoch": 3.995971802618328,
      "grad_norm": 2.461122751235962,
      "learning_rate": 1.995103346840424e-05,
      "loss": 3.0121,
      "step": 620
    },
    {
      "epoch": 4.0604229607250755,
      "grad_norm": 1.9937182664871216,
      "learning_rate": 1.9946725073220542e-05,
      "loss": 3.249,
      "step": 630
    },
    {
      "epoch": 4.124874118831823,
      "grad_norm": 2.3354651927948,
      "learning_rate": 1.9942235549209955e-05,
      "loss": 2.9879,
      "step": 640
    },
    {
      "epoch": 4.18932527693857,
      "grad_norm": 2.059208393096924,
      "learning_rate": 1.9937564978126233e-05,
      "loss": 2.987,
      "step": 650
    },
    {
      "epoch": 4.253776435045317,
      "grad_norm": 2.804398775100708,
      "learning_rate": 1.9932713445019993e-05,
      "loss": 2.9377,
      "step": 660
    },
    {
      "epoch": 4.318227593152065,
      "grad_norm": 2.1567623615264893,
      "learning_rate": 1.992768103823714e-05,
      "loss": 2.9478,
      "step": 670
    },
    {
      "epoch": 4.382678751258812,
      "grad_norm": 2.021939992904663,
      "learning_rate": 1.9922467849417288e-05,
      "loss": 2.9119,
      "step": 680
    },
    {
      "epoch": 4.447129909365559,
      "grad_norm": 1.5279889106750488,
      "learning_rate": 1.9917073973492055e-05,
      "loss": 2.9033,
      "step": 690
    },
    {
      "epoch": 4.5115810674723065,
      "grad_norm": 1.7887712717056274,
      "learning_rate": 1.991149950868336e-05,
      "loss": 2.8944,
      "step": 700
    },
    {
      "epoch": 4.5115810674723065,
      "eval_loss": 2.8965914249420166,
      "eval_runtime": 21.0307,
      "eval_samples_per_second": 76.317,
      "eval_steps_per_second": 4.802,
      "step": 700
    },
    {
      "epoch": 4.576032225579054,
      "grad_norm": 1.7073420286178589,
      "learning_rate": 1.9905744556501627e-05,
      "loss": 2.8471,
      "step": 710
    },
    {
      "epoch": 4.640483383685801,
      "grad_norm": 1.862641453742981,
      "learning_rate": 1.989980922174394e-05,
      "loss": 2.8432,
      "step": 720
    },
    {
      "epoch": 4.704934541792548,
      "grad_norm": 1.9594634771347046,
      "learning_rate": 1.9893693612492116e-05,
      "loss": 2.8482,
      "step": 730
    },
    {
      "epoch": 4.769385699899295,
      "grad_norm": 1.6772149801254272,
      "learning_rate": 1.988739784011077e-05,
      "loss": 2.8538,
      "step": 740
    },
    {
      "epoch": 4.833836858006042,
      "grad_norm": 1.9590495824813843,
      "learning_rate": 1.9880922019245258e-05,
      "loss": 2.8703,
      "step": 750
    },
    {
      "epoch": 4.898288016112789,
      "grad_norm": 1.4671125411987305,
      "learning_rate": 1.9874266267819604e-05,
      "loss": 2.8078,
      "step": 760
    },
    {
      "epoch": 4.962739174219537,
      "grad_norm": 1.8790684938430786,
      "learning_rate": 1.986743070703435e-05,
      "loss": 2.779,
      "step": 770
    },
    {
      "epoch": 5.027190332326284,
      "grad_norm": 1.6177160739898682,
      "learning_rate": 1.9860415461364343e-05,
      "loss": 3.0088,
      "step": 780
    },
    {
      "epoch": 5.091641490433031,
      "grad_norm": 1.774277687072754,
      "learning_rate": 1.9853220658556474e-05,
      "loss": 2.7841,
      "step": 790
    },
    {
      "epoch": 5.1560926485397784,
      "grad_norm": 1.4572412967681885,
      "learning_rate": 1.984584642962735e-05,
      "loss": 2.7748,
      "step": 800
    },
    {
      "epoch": 5.1560926485397784,
      "eval_loss": 2.7890875339508057,
      "eval_runtime": 19.598,
      "eval_samples_per_second": 81.896,
      "eval_steps_per_second": 5.154,
      "step": 800
    },
    {
      "epoch": 5.220543806646526,
      "grad_norm": 1.4425498247146606,
      "learning_rate": 1.9838292908860922e-05,
      "loss": 2.7712,
      "step": 810
    },
    {
      "epoch": 5.284994964753273,
      "grad_norm": 1.4506257772445679,
      "learning_rate": 1.9830560233806006e-05,
      "loss": 2.758,
      "step": 820
    },
    {
      "epoch": 5.34944612286002,
      "grad_norm": 1.3948203325271606,
      "learning_rate": 1.982264854527381e-05,
      "loss": 2.7487,
      "step": 830
    },
    {
      "epoch": 5.413897280966768,
      "grad_norm": 1.170486330986023,
      "learning_rate": 1.9814557987335363e-05,
      "loss": 2.7558,
      "step": 840
    },
    {
      "epoch": 5.478348439073515,
      "grad_norm": 1.3575382232666016,
      "learning_rate": 1.980628870731888e-05,
      "loss": 2.7266,
      "step": 850
    },
    {
      "epoch": 5.542799597180262,
      "grad_norm": 1.3390381336212158,
      "learning_rate": 1.979784085580708e-05,
      "loss": 2.7156,
      "step": 860
    },
    {
      "epoch": 5.6072507552870094,
      "grad_norm": 1.3795325756072998,
      "learning_rate": 1.978921458663447e-05,
      "loss": 2.7069,
      "step": 870
    },
    {
      "epoch": 5.671701913393756,
      "grad_norm": 1.2567253112792969,
      "learning_rate": 1.9780410056884505e-05,
      "loss": 2.7121,
      "step": 880
    },
    {
      "epoch": 5.736153071500503,
      "grad_norm": 1.6211382150650024,
      "learning_rate": 1.977142742688676e-05,
      "loss": 2.7006,
      "step": 890
    },
    {
      "epoch": 5.80060422960725,
      "grad_norm": 1.5447043180465698,
      "learning_rate": 1.9762266860213982e-05,
      "loss": 2.6635,
      "step": 900
    },
    {
      "epoch": 5.80060422960725,
      "eval_loss": 2.704418659210205,
      "eval_runtime": 20.2115,
      "eval_samples_per_second": 79.41,
      "eval_steps_per_second": 4.997,
      "step": 900
    },
    {
      "epoch": 5.865055387713998,
      "grad_norm": 1.28032648563385,
      "learning_rate": 1.9752928523679145e-05,
      "loss": 2.7062,
      "step": 910
    },
    {
      "epoch": 5.929506545820745,
      "grad_norm": 1.323886752128601,
      "learning_rate": 1.974341258733238e-05,
      "loss": 2.6955,
      "step": 920
    },
    {
      "epoch": 5.993957703927492,
      "grad_norm": 1.370737910270691,
      "learning_rate": 1.9733719224457896e-05,
      "loss": 2.6607,
      "step": 930
    },
    {
      "epoch": 6.05840886203424,
      "grad_norm": 1.2304973602294922,
      "learning_rate": 1.972384861157082e-05,
      "loss": 2.8645,
      "step": 940
    },
    {
      "epoch": 6.122860020140987,
      "grad_norm": 1.251336693763733,
      "learning_rate": 1.9713800928413987e-05,
      "loss": 2.678,
      "step": 950
    },
    {
      "epoch": 6.187311178247734,
      "grad_norm": 1.4331055879592896,
      "learning_rate": 1.9703576357954653e-05,
      "loss": 2.6339,
      "step": 960
    },
    {
      "epoch": 6.251762336354481,
      "grad_norm": 1.4404268264770508,
      "learning_rate": 1.969317508638119e-05,
      "loss": 2.6497,
      "step": 970
    },
    {
      "epoch": 6.316213494461229,
      "grad_norm": 1.2648017406463623,
      "learning_rate": 1.9682597303099663e-05,
      "loss": 2.6194,
      "step": 980
    },
    {
      "epoch": 6.380664652567976,
      "grad_norm": 1.464142084121704,
      "learning_rate": 1.9671843200730408e-05,
      "loss": 2.6342,
      "step": 990
    },
    {
      "epoch": 6.445115810674723,
      "grad_norm": 1.4572694301605225,
      "learning_rate": 1.96609129751045e-05,
      "loss": 2.6102,
      "step": 1000
    },
    {
      "epoch": 6.445115810674723,
      "eval_loss": 2.6344494819641113,
      "eval_runtime": 19.7801,
      "eval_samples_per_second": 81.142,
      "eval_steps_per_second": 5.106,
      "step": 1000
    },
    {
      "epoch": 6.509566968781471,
      "grad_norm": 1.3998351097106934,
      "learning_rate": 1.9649806825260215e-05,
      "loss": 2.6117,
      "step": 1010
    },
    {
      "epoch": 6.574018126888218,
      "grad_norm": 1.2323154211044312,
      "learning_rate": 1.9638524953439385e-05,
      "loss": 2.5907,
      "step": 1020
    },
    {
      "epoch": 6.638469284994965,
      "grad_norm": 1.5451122522354126,
      "learning_rate": 1.9627067565083716e-05,
      "loss": 2.5988,
      "step": 1030
    },
    {
      "epoch": 6.702920443101712,
      "grad_norm": 1.4288865327835083,
      "learning_rate": 1.9615434868831057e-05,
      "loss": 2.6088,
      "step": 1040
    },
    {
      "epoch": 6.76737160120846,
      "grad_norm": 1.4163669347763062,
      "learning_rate": 1.9603627076511595e-05,
      "loss": 2.5787,
      "step": 1050
    },
    {
      "epoch": 6.831822759315206,
      "grad_norm": 1.4155839681625366,
      "learning_rate": 1.9591644403143997e-05,
      "loss": 2.5802,
      "step": 1060
    },
    {
      "epoch": 6.896273917421953,
      "grad_norm": 1.2025197744369507,
      "learning_rate": 1.9579487066931495e-05,
      "loss": 2.56,
      "step": 1070
    },
    {
      "epoch": 6.960725075528701,
      "grad_norm": 1.331957459449768,
      "learning_rate": 1.956715528925792e-05,
      "loss": 2.5797,
      "step": 1080
    },
    {
      "epoch": 7.025176233635448,
      "grad_norm": 1.543091893196106,
      "learning_rate": 1.955464929468365e-05,
      "loss": 2.7584,
      "step": 1090
    },
    {
      "epoch": 7.089627391742195,
      "grad_norm": 1.4929845333099365,
      "learning_rate": 1.954196931094155e-05,
      "loss": 2.5724,
      "step": 1100
    },
    {
      "epoch": 7.089627391742195,
      "eval_loss": 2.5758652687072754,
      "eval_runtime": 19.7829,
      "eval_samples_per_second": 81.131,
      "eval_steps_per_second": 5.105,
      "step": 1100
    },
    {
      "epoch": 7.1540785498489425,
      "grad_norm": 1.3239967823028564,
      "learning_rate": 1.9529115568932796e-05,
      "loss": 2.5215,
      "step": 1110
    },
    {
      "epoch": 7.21852970795569,
      "grad_norm": 1.2997671365737915,
      "learning_rate": 1.9516088302722696e-05,
      "loss": 2.5256,
      "step": 1120
    },
    {
      "epoch": 7.282980866062437,
      "grad_norm": 1.4063647985458374,
      "learning_rate": 1.9502887749536406e-05,
      "loss": 2.5286,
      "step": 1130
    },
    {
      "epoch": 7.347432024169184,
      "grad_norm": 1.625854730606079,
      "learning_rate": 1.9489514149754624e-05,
      "loss": 2.5245,
      "step": 1140
    },
    {
      "epoch": 7.411883182275932,
      "grad_norm": 1.2255531549453735,
      "learning_rate": 1.9475967746909212e-05,
      "loss": 2.5373,
      "step": 1150
    },
    {
      "epoch": 7.476334340382679,
      "grad_norm": 1.3738540410995483,
      "learning_rate": 1.946224878767875e-05,
      "loss": 2.5108,
      "step": 1160
    },
    {
      "epoch": 7.540785498489426,
      "grad_norm": 1.7884373664855957,
      "learning_rate": 1.9448357521884057e-05,
      "loss": 2.512,
      "step": 1170
    },
    {
      "epoch": 7.6052366565961735,
      "grad_norm": 1.4379594326019287,
      "learning_rate": 1.9434294202483634e-05,
      "loss": 2.5219,
      "step": 1180
    },
    {
      "epoch": 7.669687814702921,
      "grad_norm": 1.5573759078979492,
      "learning_rate": 1.9420059085569062e-05,
      "loss": 2.4937,
      "step": 1190
    },
    {
      "epoch": 7.734138972809667,
      "grad_norm": 1.5050828456878662,
      "learning_rate": 1.940565243036034e-05,
      "loss": 2.5267,
      "step": 1200
    },
    {
      "epoch": 7.734138972809667,
      "eval_loss": 2.5181825160980225,
      "eval_runtime": 19.5726,
      "eval_samples_per_second": 82.002,
      "eval_steps_per_second": 5.16,
      "step": 1200
    },
    {
      "epoch": 7.7985901309164145,
      "grad_norm": 1.6680591106414795,
      "learning_rate": 1.9391074499201155e-05,
      "loss": 2.5034,
      "step": 1210
    },
    {
      "epoch": 7.863041289023162,
      "grad_norm": 1.1828731298446655,
      "learning_rate": 1.9376325557554113e-05,
      "loss": 2.493,
      "step": 1220
    },
    {
      "epoch": 7.927492447129909,
      "grad_norm": 1.5516493320465088,
      "learning_rate": 1.9361405873995904e-05,
      "loss": 2.4866,
      "step": 1230
    },
    {
      "epoch": 7.991943605236656,
      "grad_norm": 1.2538108825683594,
      "learning_rate": 1.9346315720212416e-05,
      "loss": 2.4935,
      "step": 1240
    },
    {
      "epoch": 8.056394763343404,
      "grad_norm": 1.4746509790420532,
      "learning_rate": 1.933105537099377e-05,
      "loss": 2.6784,
      "step": 1250
    },
    {
      "epoch": 8.120845921450151,
      "grad_norm": 1.491493821144104,
      "learning_rate": 1.9315625104229336e-05,
      "loss": 2.484,
      "step": 1260
    },
    {
      "epoch": 8.185297079556898,
      "grad_norm": 1.5967274904251099,
      "learning_rate": 1.9300025200902666e-05,
      "loss": 2.4746,
      "step": 1270
    },
    {
      "epoch": 8.249748237663646,
      "grad_norm": 1.477860689163208,
      "learning_rate": 1.928425594508637e-05,
      "loss": 2.4447,
      "step": 1280
    },
    {
      "epoch": 8.314199395770393,
      "grad_norm": 1.4457802772521973,
      "learning_rate": 1.9268317623936957e-05,
      "loss": 2.4438,
      "step": 1290
    },
    {
      "epoch": 8.37865055387714,
      "grad_norm": 1.1860662698745728,
      "learning_rate": 1.9252210527689596e-05,
      "loss": 2.4863,
      "step": 1300
    },
    {
      "epoch": 8.37865055387714,
      "eval_loss": 2.4702823162078857,
      "eval_runtime": 20.2158,
      "eval_samples_per_second": 79.393,
      "eval_steps_per_second": 4.996,
      "step": 1300
    },
    {
      "epoch": 8.443101711983887,
      "grad_norm": 1.36308753490448,
      "learning_rate": 1.9235934949652825e-05,
      "loss": 2.4382,
      "step": 1310
    },
    {
      "epoch": 8.507552870090635,
      "grad_norm": 1.0049257278442383,
      "learning_rate": 1.9219491186203222e-05,
      "loss": 2.4478,
      "step": 1320
    },
    {
      "epoch": 8.572004028197382,
      "grad_norm": 1.4377230405807495,
      "learning_rate": 1.9202879536780013e-05,
      "loss": 2.4276,
      "step": 1330
    },
    {
      "epoch": 8.63645518630413,
      "grad_norm": 1.3836121559143066,
      "learning_rate": 1.91861003038796e-05,
      "loss": 2.4464,
      "step": 1340
    },
    {
      "epoch": 8.700906344410877,
      "grad_norm": 1.3200100660324097,
      "learning_rate": 1.9169153793050065e-05,
      "loss": 2.438,
      "step": 1350
    },
    {
      "epoch": 8.765357502517624,
      "grad_norm": 1.4599627256393433,
      "learning_rate": 1.9152040312885604e-05,
      "loss": 2.4192,
      "step": 1360
    },
    {
      "epoch": 8.829808660624371,
      "grad_norm": 1.2848976850509644,
      "learning_rate": 1.9134760175020906e-05,
      "loss": 2.4119,
      "step": 1370
    },
    {
      "epoch": 8.894259818731118,
      "grad_norm": 1.195146083831787,
      "learning_rate": 1.9117313694125482e-05,
      "loss": 2.4047,
      "step": 1380
    },
    {
      "epoch": 8.958710976837866,
      "grad_norm": 1.5599223375320435,
      "learning_rate": 1.9099701187897927e-05,
      "loss": 2.4238,
      "step": 1390
    },
    {
      "epoch": 9.023162134944613,
      "grad_norm": 1.7016339302062988,
      "learning_rate": 1.9081922977060146e-05,
      "loss": 2.5924,
      "step": 1400
    },
    {
      "epoch": 9.023162134944613,
      "eval_loss": 2.425938844680786,
      "eval_runtime": 20.1925,
      "eval_samples_per_second": 79.485,
      "eval_steps_per_second": 5.002,
      "step": 1400
    },
    {
      "epoch": 9.08761329305136,
      "grad_norm": 1.5949656963348389,
      "learning_rate": 1.9063979385351512e-05,
      "loss": 2.4341,
      "step": 1410
    },
    {
      "epoch": 9.152064451158108,
      "grad_norm": 1.4340884685516357,
      "learning_rate": 1.9045870739522953e-05,
      "loss": 2.4071,
      "step": 1420
    },
    {
      "epoch": 9.216515609264855,
      "grad_norm": 1.2564432621002197,
      "learning_rate": 1.902759736933102e-05,
      "loss": 2.3866,
      "step": 1430
    },
    {
      "epoch": 9.280966767371602,
      "grad_norm": 1.338267207145691,
      "learning_rate": 1.9009159607531886e-05,
      "loss": 2.3687,
      "step": 1440
    },
    {
      "epoch": 9.345417925478348,
      "grad_norm": 1.4795355796813965,
      "learning_rate": 1.8990557789875265e-05,
      "loss": 2.3932,
      "step": 1450
    },
    {
      "epoch": 9.409869083585095,
      "grad_norm": 1.3495395183563232,
      "learning_rate": 1.8971792255098326e-05,
      "loss": 2.3738,
      "step": 1460
    },
    {
      "epoch": 9.474320241691842,
      "grad_norm": 1.5323916673660278,
      "learning_rate": 1.8952863344919495e-05,
      "loss": 2.3773,
      "step": 1470
    },
    {
      "epoch": 9.53877139979859,
      "grad_norm": 1.2470097541809082,
      "learning_rate": 1.893377140403225e-05,
      "loss": 2.364,
      "step": 1480
    },
    {
      "epoch": 9.603222557905337,
      "grad_norm": 1.2765355110168457,
      "learning_rate": 1.891451678009886e-05,
      "loss": 2.3414,
      "step": 1490
    },
    {
      "epoch": 9.667673716012084,
      "grad_norm": 1.2045294046401978,
      "learning_rate": 1.8895099823744005e-05,
      "loss": 2.3869,
      "step": 1500
    },
    {
      "epoch": 9.667673716012084,
      "eval_loss": 2.387402057647705,
      "eval_runtime": 20.2004,
      "eval_samples_per_second": 79.454,
      "eval_steps_per_second": 5.0,
      "step": 1500
    },
    {
      "epoch": 9.732124874118831,
      "grad_norm": 1.488773226737976,
      "learning_rate": 1.887552088854844e-05,
      "loss": 2.3653,
      "step": 1510
    },
    {
      "epoch": 9.796576032225579,
      "grad_norm": 1.442879319190979,
      "learning_rate": 1.8855780331042538e-05,
      "loss": 2.3483,
      "step": 1520
    },
    {
      "epoch": 9.861027190332326,
      "grad_norm": 1.2660679817199707,
      "learning_rate": 1.8835878510699793e-05,
      "loss": 2.3569,
      "step": 1530
    },
    {
      "epoch": 9.925478348439073,
      "grad_norm": 1.575183391571045,
      "learning_rate": 1.8815815789930277e-05,
      "loss": 2.3385,
      "step": 1540
    },
    {
      "epoch": 9.98992950654582,
      "grad_norm": 1.270186185836792,
      "learning_rate": 1.8795592534074045e-05,
      "loss": 2.3276,
      "step": 1550
    },
    {
      "epoch": 10.054380664652568,
      "grad_norm": 1.4595534801483154,
      "learning_rate": 1.877520911139448e-05,
      "loss": 2.5222,
      "step": 1560
    },
    {
      "epoch": 10.118831822759315,
      "grad_norm": 1.2067569494247437,
      "learning_rate": 1.8754665893071583e-05,
      "loss": 2.3175,
      "step": 1570
    },
    {
      "epoch": 10.183282980866062,
      "grad_norm": 1.1168224811553955,
      "learning_rate": 1.8733963253195217e-05,
      "loss": 2.3505,
      "step": 1580
    },
    {
      "epoch": 10.24773413897281,
      "grad_norm": 1.3407734632492065,
      "learning_rate": 1.8713101568758295e-05,
      "loss": 2.3406,
      "step": 1590
    },
    {
      "epoch": 10.312185297079557,
      "grad_norm": 1.2485662698745728,
      "learning_rate": 1.8692081219649926e-05,
      "loss": 2.3084,
      "step": 1600
    },
    {
      "epoch": 10.312185297079557,
      "eval_loss": 2.3460793495178223,
      "eval_runtime": 19.8125,
      "eval_samples_per_second": 81.009,
      "eval_steps_per_second": 5.098,
      "step": 1600
    },
    {
      "epoch": 10.376636455186304,
      "grad_norm": 1.335404872894287,
      "learning_rate": 1.8670902588648467e-05,
      "loss": 2.3151,
      "step": 1610
    },
    {
      "epoch": 10.441087613293051,
      "grad_norm": 1.5283225774765015,
      "learning_rate": 1.8649566061414583e-05,
      "loss": 2.3153,
      "step": 1620
    },
    {
      "epoch": 10.505538771399799,
      "grad_norm": 1.2574470043182373,
      "learning_rate": 1.8628072026484215e-05,
      "loss": 2.303,
      "step": 1630
    },
    {
      "epoch": 10.569989929506546,
      "grad_norm": 1.3589491844177246,
      "learning_rate": 1.8606420875261492e-05,
      "loss": 2.3191,
      "step": 1640
    },
    {
      "epoch": 10.634441087613293,
      "grad_norm": 1.3146897554397583,
      "learning_rate": 1.858461300201163e-05,
      "loss": 2.3128,
      "step": 1650
    },
    {
      "epoch": 10.69889224572004,
      "grad_norm": 1.363336205482483,
      "learning_rate": 1.856264880385372e-05,
      "loss": 2.2795,
      "step": 1660
    },
    {
      "epoch": 10.763343403826788,
      "grad_norm": 1.402933120727539,
      "learning_rate": 1.8540528680753525e-05,
      "loss": 2.2819,
      "step": 1670
    },
    {
      "epoch": 10.827794561933535,
      "grad_norm": 1.199621558189392,
      "learning_rate": 1.851825303551618e-05,
      "loss": 2.2937,
      "step": 1680
    },
    {
      "epoch": 10.892245720040282,
      "grad_norm": 1.4294910430908203,
      "learning_rate": 1.8495822273778867e-05,
      "loss": 2.2985,
      "step": 1690
    },
    {
      "epoch": 10.95669687814703,
      "grad_norm": 1.211239218711853,
      "learning_rate": 1.8473236804003412e-05,
      "loss": 2.2803,
      "step": 1700
    },
    {
      "epoch": 10.95669687814703,
      "eval_loss": 2.2990825176239014,
      "eval_runtime": 19.7801,
      "eval_samples_per_second": 81.142,
      "eval_steps_per_second": 5.106,
      "step": 1700
    },
    {
      "epoch": 11.021148036253777,
      "grad_norm": 1.610541582107544,
      "learning_rate": 1.8450497037468876e-05,
      "loss": 2.4507,
      "step": 1710
    },
    {
      "epoch": 11.085599194360524,
      "grad_norm": 1.265767216682434,
      "learning_rate": 1.8427603388264027e-05,
      "loss": 2.2681,
      "step": 1720
    },
    {
      "epoch": 11.150050352467272,
      "grad_norm": 1.1170353889465332,
      "learning_rate": 1.8404556273279835e-05,
      "loss": 2.2513,
      "step": 1730
    },
    {
      "epoch": 11.214501510574019,
      "grad_norm": 1.5260467529296875,
      "learning_rate": 1.8381356112201863e-05,
      "loss": 2.2696,
      "step": 1740
    },
    {
      "epoch": 11.278952668680766,
      "grad_norm": 1.3105249404907227,
      "learning_rate": 1.835800332750263e-05,
      "loss": 2.2703,
      "step": 1750
    },
    {
      "epoch": 11.343403826787512,
      "grad_norm": 1.692631483078003,
      "learning_rate": 1.8334498344433903e-05,
      "loss": 2.2637,
      "step": 1760
    },
    {
      "epoch": 11.407854984894259,
      "grad_norm": 1.5331857204437256,
      "learning_rate": 1.8310841591018977e-05,
      "loss": 2.2499,
      "step": 1770
    },
    {
      "epoch": 11.472306143001006,
      "grad_norm": 1.3038716316223145,
      "learning_rate": 1.828703349804487e-05,
      "loss": 2.2658,
      "step": 1780
    },
    {
      "epoch": 11.536757301107754,
      "grad_norm": 1.1918790340423584,
      "learning_rate": 1.826307449905447e-05,
      "loss": 2.2517,
      "step": 1790
    },
    {
      "epoch": 11.6012084592145,
      "grad_norm": 1.4929193258285522,
      "learning_rate": 1.823896503033865e-05,
      "loss": 2.2393,
      "step": 1800
    },
    {
      "epoch": 11.6012084592145,
      "eval_loss": 2.256037712097168,
      "eval_runtime": 21.0369,
      "eval_samples_per_second": 76.294,
      "eval_steps_per_second": 4.801,
      "step": 1800
    },
    {
      "epoch": 11.665659617321248,
      "grad_norm": 1.3841142654418945,
      "learning_rate": 1.8214705530928322e-05,
      "loss": 2.228,
      "step": 1810
    },
    {
      "epoch": 11.730110775427995,
      "grad_norm": 1.3609843254089355,
      "learning_rate": 1.819029644258645e-05,
      "loss": 2.2211,
      "step": 1820
    },
    {
      "epoch": 11.794561933534743,
      "grad_norm": 1.7460395097732544,
      "learning_rate": 1.816573820979998e-05,
      "loss": 2.2081,
      "step": 1830
    },
    {
      "epoch": 11.85901309164149,
      "grad_norm": 1.2562885284423828,
      "learning_rate": 1.8141031279771777e-05,
      "loss": 2.2201,
      "step": 1840
    },
    {
      "epoch": 11.923464249748237,
      "grad_norm": 1.5480189323425293,
      "learning_rate": 1.811617610241246e-05,
      "loss": 2.2099,
      "step": 1850
    },
    {
      "epoch": 11.987915407854985,
      "grad_norm": 1.1817470788955688,
      "learning_rate": 1.8091173130332214e-05,
      "loss": 2.1935,
      "step": 1860
    },
    {
      "epoch": 12.052366565961732,
      "grad_norm": 1.283387541770935,
      "learning_rate": 1.8066022818832564e-05,
      "loss": 2.3754,
      "step": 1870
    },
    {
      "epoch": 12.11681772406848,
      "grad_norm": 1.4642541408538818,
      "learning_rate": 1.804072562589805e-05,
      "loss": 2.1914,
      "step": 1880
    },
    {
      "epoch": 12.181268882175226,
      "grad_norm": 1.3333503007888794,
      "learning_rate": 1.8015282012187927e-05,
      "loss": 2.1918,
      "step": 1890
    },
    {
      "epoch": 12.245720040281974,
      "grad_norm": 1.4526790380477905,
      "learning_rate": 1.7989692441027744e-05,
      "loss": 2.1748,
      "step": 1900
    },
    {
      "epoch": 12.245720040281974,
      "eval_loss": 2.212247133255005,
      "eval_runtime": 20.2138,
      "eval_samples_per_second": 79.401,
      "eval_steps_per_second": 4.997,
      "step": 1900
    },
    {
      "epoch": 12.310171198388721,
      "grad_norm": 1.2756415605545044,
      "learning_rate": 1.796395737840093e-05,
      "loss": 2.2262,
      "step": 1910
    },
    {
      "epoch": 12.374622356495468,
      "grad_norm": 1.2323824167251587,
      "learning_rate": 1.7938077292940288e-05,
      "loss": 2.1796,
      "step": 1920
    },
    {
      "epoch": 12.439073514602216,
      "grad_norm": 1.2132095098495483,
      "learning_rate": 1.7912052655919478e-05,
      "loss": 2.1934,
      "step": 1930
    },
    {
      "epoch": 12.503524672708963,
      "grad_norm": 1.45913827419281,
      "learning_rate": 1.7885883941244432e-05,
      "loss": 2.169,
      "step": 1940
    },
    {
      "epoch": 12.56797583081571,
      "grad_norm": 1.398886799812317,
      "learning_rate": 1.7859571625444712e-05,
      "loss": 2.1845,
      "step": 1950
    },
    {
      "epoch": 12.632426988922457,
      "grad_norm": 1.386767029762268,
      "learning_rate": 1.7833116187664846e-05,
      "loss": 2.1563,
      "step": 1960
    },
    {
      "epoch": 12.696878147029205,
      "grad_norm": 1.4831831455230713,
      "learning_rate": 1.7806518109655604e-05,
      "loss": 2.1592,
      "step": 1970
    },
    {
      "epoch": 12.761329305135952,
      "grad_norm": 1.3067753314971924,
      "learning_rate": 1.777977787576521e-05,
      "loss": 2.1587,
      "step": 1980
    },
    {
      "epoch": 12.8257804632427,
      "grad_norm": 1.4871938228607178,
      "learning_rate": 1.7752895972930538e-05,
      "loss": 2.1575,
      "step": 1990
    },
    {
      "epoch": 12.890231621349447,
      "grad_norm": 1.3559268712997437,
      "learning_rate": 1.772587289066823e-05,
      "loss": 2.1365,
      "step": 2000
    },
    {
      "epoch": 12.890231621349447,
      "eval_loss": 2.167628526687622,
      "eval_runtime": 19.7767,
      "eval_samples_per_second": 81.156,
      "eval_steps_per_second": 5.107,
      "step": 2000
    },
    {
      "epoch": 12.954682779456194,
      "grad_norm": 1.4033281803131104,
      "learning_rate": 1.769870912106581e-05,
      "loss": 2.1484,
      "step": 2010
    },
    {
      "epoch": 13.019133937562941,
      "grad_norm": 1.3120031356811523,
      "learning_rate": 1.7671405158772686e-05,
      "loss": 2.3176,
      "step": 2020
    },
    {
      "epoch": 13.083585095669688,
      "grad_norm": 1.3632259368896484,
      "learning_rate": 1.764396150099116e-05,
      "loss": 2.1399,
      "step": 2030
    },
    {
      "epoch": 13.148036253776436,
      "grad_norm": 1.3641666173934937,
      "learning_rate": 1.7616378647467387e-05,
      "loss": 2.1302,
      "step": 2040
    },
    {
      "epoch": 13.212487411883183,
      "grad_norm": 1.2037588357925415,
      "learning_rate": 1.758865710048225e-05,
      "loss": 2.1169,
      "step": 2050
    },
    {
      "epoch": 13.27693856998993,
      "grad_norm": 1.3655686378479004,
      "learning_rate": 1.7560797364842235e-05,
      "loss": 2.1228,
      "step": 2060
    },
    {
      "epoch": 13.341389728096678,
      "grad_norm": 1.2204645872116089,
      "learning_rate": 1.7532799947870224e-05,
      "loss": 2.1027,
      "step": 2070
    },
    {
      "epoch": 13.405840886203425,
      "grad_norm": 1.3521157503128052,
      "learning_rate": 1.7504665359396255e-05,
      "loss": 2.1038,
      "step": 2080
    },
    {
      "epoch": 13.47029204431017,
      "grad_norm": 1.3395577669143677,
      "learning_rate": 1.7476394111748262e-05,
      "loss": 2.0887,
      "step": 2090
    },
    {
      "epoch": 13.534743202416918,
      "grad_norm": 1.427236795425415,
      "learning_rate": 1.7447986719742708e-05,
      "loss": 2.1054,
      "step": 2100
    },
    {
      "epoch": 13.534743202416918,
      "eval_loss": 2.1201364994049072,
      "eval_runtime": 19.5957,
      "eval_samples_per_second": 81.906,
      "eval_steps_per_second": 5.154,
      "step": 2100
    },
    {
      "epoch": 13.599194360523665,
      "grad_norm": 1.7869354486465454,
      "learning_rate": 1.7419443700675248e-05,
      "loss": 2.0844,
      "step": 2110
    },
    {
      "epoch": 13.663645518630412,
      "grad_norm": 1.4408442974090576,
      "learning_rate": 1.7390765574311287e-05,
      "loss": 2.1039,
      "step": 2120
    },
    {
      "epoch": 13.72809667673716,
      "grad_norm": 1.5654326677322388,
      "learning_rate": 1.7361952862876505e-05,
      "loss": 2.0889,
      "step": 2130
    },
    {
      "epoch": 13.792547834843907,
      "grad_norm": 1.5477581024169922,
      "learning_rate": 1.7333006091047386e-05,
      "loss": 2.0645,
      "step": 2140
    },
    {
      "epoch": 13.856998992950654,
      "grad_norm": 1.3517593145370483,
      "learning_rate": 1.730392578594162e-05,
      "loss": 2.0899,
      "step": 2150
    },
    {
      "epoch": 13.921450151057401,
      "grad_norm": 1.4705018997192383,
      "learning_rate": 1.7274712477108538e-05,
      "loss": 2.0876,
      "step": 2160
    },
    {
      "epoch": 13.985901309164149,
      "grad_norm": 1.5158582925796509,
      "learning_rate": 1.7245366696519448e-05,
      "loss": 2.0757,
      "step": 2170
    },
    {
      "epoch": 14.050352467270896,
      "grad_norm": 1.5032602548599243,
      "learning_rate": 1.7215888978557953e-05,
      "loss": 2.2535,
      "step": 2180
    },
    {
      "epoch": 14.114803625377643,
      "grad_norm": 1.3072115182876587,
      "learning_rate": 1.7186279860010228e-05,
      "loss": 2.0597,
      "step": 2190
    },
    {
      "epoch": 14.17925478348439,
      "grad_norm": 1.522055745124817,
      "learning_rate": 1.7156539880055236e-05,
      "loss": 2.0326,
      "step": 2200
    },
    {
      "epoch": 14.17925478348439,
      "eval_loss": 2.0740458965301514,
      "eval_runtime": 19.8046,
      "eval_samples_per_second": 81.042,
      "eval_steps_per_second": 5.1,
      "step": 2200
    },
    {
      "epoch": 14.243705941591138,
      "grad_norm": 1.2756038904190063,
      "learning_rate": 1.7126669580254908e-05,
      "loss": 2.0255,
      "step": 2210
    },
    {
      "epoch": 14.308157099697885,
      "grad_norm": 1.6488431692123413,
      "learning_rate": 1.7096669504544293e-05,
      "loss": 2.0271,
      "step": 2220
    },
    {
      "epoch": 14.372608257804632,
      "grad_norm": 1.4515522718429565,
      "learning_rate": 1.706654019922164e-05,
      "loss": 2.055,
      "step": 2230
    },
    {
      "epoch": 14.43705941591138,
      "grad_norm": 1.530522108078003,
      "learning_rate": 1.7036282212938468e-05,
      "loss": 2.0461,
      "step": 2240
    },
    {
      "epoch": 14.501510574018127,
      "grad_norm": 1.4453781843185425,
      "learning_rate": 1.7005896096689544e-05,
      "loss": 2.0376,
      "step": 2250
    },
    {
      "epoch": 14.565961732124874,
      "grad_norm": 1.578723430633545,
      "learning_rate": 1.697538240380288e-05,
      "loss": 2.0439,
      "step": 2260
    },
    {
      "epoch": 14.630412890231622,
      "grad_norm": 1.3801120519638062,
      "learning_rate": 1.6944741689929646e-05,
      "loss": 2.0133,
      "step": 2270
    },
    {
      "epoch": 14.694864048338369,
      "grad_norm": 1.2451757192611694,
      "learning_rate": 1.6913974513034046e-05,
      "loss": 2.014,
      "step": 2280
    },
    {
      "epoch": 14.759315206445116,
      "grad_norm": 1.3197917938232422,
      "learning_rate": 1.6883081433383163e-05,
      "loss": 2.0191,
      "step": 2290
    },
    {
      "epoch": 14.823766364551863,
      "grad_norm": 1.5146923065185547,
      "learning_rate": 1.6852063013536765e-05,
      "loss": 1.992,
      "step": 2300
    },
    {
      "epoch": 14.823766364551863,
      "eval_loss": 2.0283043384552,
      "eval_runtime": 20.2251,
      "eval_samples_per_second": 79.357,
      "eval_steps_per_second": 4.994,
      "step": 2300
    },
    {
      "epoch": 14.88821752265861,
      "grad_norm": 1.3689446449279785,
      "learning_rate": 1.6820919818337035e-05,
      "loss": 1.9953,
      "step": 2310
    },
    {
      "epoch": 14.952668680765358,
      "grad_norm": 1.3700356483459473,
      "learning_rate": 1.6789652414898315e-05,
      "loss": 2.0154,
      "step": 2320
    },
    {
      "epoch": 15.017119838872105,
      "grad_norm": 1.4734395742416382,
      "learning_rate": 1.6758261372596768e-05,
      "loss": 2.1667,
      "step": 2330
    },
    {
      "epoch": 15.081570996978853,
      "grad_norm": 1.5056065320968628,
      "learning_rate": 1.6726747263059996e-05,
      "loss": 1.9786,
      "step": 2340
    },
    {
      "epoch": 15.1460221550856,
      "grad_norm": 1.51126229763031,
      "learning_rate": 1.6695110660156652e-05,
      "loss": 1.9809,
      "step": 2350
    },
    {
      "epoch": 15.210473313192347,
      "grad_norm": 1.5719977617263794,
      "learning_rate": 1.6663352139985977e-05,
      "loss": 1.9794,
      "step": 2360
    },
    {
      "epoch": 15.274924471299094,
      "grad_norm": 1.413379192352295,
      "learning_rate": 1.6631472280867314e-05,
      "loss": 1.9688,
      "step": 2370
    },
    {
      "epoch": 15.339375629405842,
      "grad_norm": 1.4571571350097656,
      "learning_rate": 1.6599471663329577e-05,
      "loss": 1.9489,
      "step": 2380
    },
    {
      "epoch": 15.403826787512589,
      "grad_norm": 1.4078450202941895,
      "learning_rate": 1.656735087010067e-05,
      "loss": 1.9597,
      "step": 2390
    },
    {
      "epoch": 15.468277945619334,
      "grad_norm": 1.5099396705627441,
      "learning_rate": 1.653511048609689e-05,
      "loss": 1.9692,
      "step": 2400
    },
    {
      "epoch": 15.468277945619334,
      "eval_loss": 1.986576795578003,
      "eval_runtime": 20.2391,
      "eval_samples_per_second": 79.302,
      "eval_steps_per_second": 4.99,
      "step": 2400
    },
    {
      "epoch": 15.532729103726084,
      "grad_norm": 1.317844271659851,
      "learning_rate": 1.6502751098412282e-05,
      "loss": 1.9565,
      "step": 2410
    },
    {
      "epoch": 15.597180261832829,
      "grad_norm": 1.5470753908157349,
      "learning_rate": 1.6470273296307907e-05,
      "loss": 1.9569,
      "step": 2420
    },
    {
      "epoch": 15.661631419939576,
      "grad_norm": 1.732040524482727,
      "learning_rate": 1.643767767120117e-05,
      "loss": 1.9452,
      "step": 2430
    },
    {
      "epoch": 15.726082578046324,
      "grad_norm": 1.4074490070343018,
      "learning_rate": 1.6404964816654993e-05,
      "loss": 1.9543,
      "step": 2440
    },
    {
      "epoch": 15.79053373615307,
      "grad_norm": 1.6578551530838013,
      "learning_rate": 1.6372135328367058e-05,
      "loss": 1.9372,
      "step": 2450
    },
    {
      "epoch": 15.854984894259818,
      "grad_norm": 1.514443039894104,
      "learning_rate": 1.6339189804158922e-05,
      "loss": 1.9374,
      "step": 2460
    },
    {
      "epoch": 15.919436052366565,
      "grad_norm": 1.6120061874389648,
      "learning_rate": 1.630612884396515e-05,
      "loss": 1.962,
      "step": 2470
    },
    {
      "epoch": 15.983887210473313,
      "grad_norm": 1.9274513721466064,
      "learning_rate": 1.6272953049822376e-05,
      "loss": 1.9457,
      "step": 2480
    },
    {
      "epoch": 16.048338368580062,
      "grad_norm": 1.4026292562484741,
      "learning_rate": 1.6239663025858356e-05,
      "loss": 2.1124,
      "step": 2490
    },
    {
      "epoch": 16.112789526686807,
      "grad_norm": 1.543957233428955,
      "learning_rate": 1.6206259378280956e-05,
      "loss": 1.9235,
      "step": 2500
    },
    {
      "epoch": 16.112789526686807,
      "eval_loss": 1.9452571868896484,
      "eval_runtime": 20.1977,
      "eval_samples_per_second": 79.465,
      "eval_steps_per_second": 5.001,
      "step": 2500
    },
    {
      "epoch": 16.177240684793556,
      "grad_norm": 1.6722089052200317,
      "learning_rate": 1.6172742715367124e-05,
      "loss": 1.8982,
      "step": 2510
    },
    {
      "epoch": 16.241691842900302,
      "grad_norm": 1.649903416633606,
      "learning_rate": 1.613911364745179e-05,
      "loss": 1.9176,
      "step": 2520
    },
    {
      "epoch": 16.30614300100705,
      "grad_norm": 1.3199961185455322,
      "learning_rate": 1.6105372786916776e-05,
      "loss": 1.9226,
      "step": 2530
    },
    {
      "epoch": 16.370594159113796,
      "grad_norm": 1.513606309890747,
      "learning_rate": 1.607152074817964e-05,
      "loss": 1.9141,
      "step": 2540
    },
    {
      "epoch": 16.435045317220546,
      "grad_norm": 1.3551160097122192,
      "learning_rate": 1.6037558147682473e-05,
      "loss": 1.9051,
      "step": 2550
    },
    {
      "epoch": 16.49949647532729,
      "grad_norm": 1.5009561777114868,
      "learning_rate": 1.60034856038807e-05,
      "loss": 1.9184,
      "step": 2560
    },
    {
      "epoch": 16.56394763343404,
      "grad_norm": 1.4896663427352905,
      "learning_rate": 1.5969303737231786e-05,
      "loss": 1.9005,
      "step": 2570
    },
    {
      "epoch": 16.628398791540786,
      "grad_norm": 1.4002718925476074,
      "learning_rate": 1.593501317018396e-05,
      "loss": 1.9057,
      "step": 2580
    },
    {
      "epoch": 16.69284994964753,
      "grad_norm": 1.5425680875778198,
      "learning_rate": 1.5900614527164876e-05,
      "loss": 1.8948,
      "step": 2590
    },
    {
      "epoch": 16.75730110775428,
      "grad_norm": 1.4177629947662354,
      "learning_rate": 1.586610843457024e-05,
      "loss": 1.8748,
      "step": 2600
    },
    {
      "epoch": 16.75730110775428,
      "eval_loss": 1.913808822631836,
      "eval_runtime": 19.5797,
      "eval_samples_per_second": 81.973,
      "eval_steps_per_second": 5.158,
      "step": 2600
    },
    {
      "epoch": 16.821752265861026,
      "grad_norm": 1.6956419944763184,
      "learning_rate": 1.5831495520752395e-05,
      "loss": 1.8708,
      "step": 2610
    },
    {
      "epoch": 16.886203423967775,
      "grad_norm": 1.5349304676055908,
      "learning_rate": 1.5796776416008897e-05,
      "loss": 1.8814,
      "step": 2620
    },
    {
      "epoch": 16.95065458207452,
      "grad_norm": 1.4432286024093628,
      "learning_rate": 1.5761951752571032e-05,
      "loss": 1.8622,
      "step": 2630
    },
    {
      "epoch": 17.01510574018127,
      "grad_norm": 1.4395033121109009,
      "learning_rate": 1.5727022164592282e-05,
      "loss": 2.0162,
      "step": 2640
    },
    {
      "epoch": 17.079556898288015,
      "grad_norm": 1.451162338256836,
      "learning_rate": 1.569198828813681e-05,
      "loss": 1.8803,
      "step": 2650
    },
    {
      "epoch": 17.144008056394764,
      "grad_norm": 1.6109211444854736,
      "learning_rate": 1.5656850761167848e-05,
      "loss": 1.8511,
      "step": 2660
    },
    {
      "epoch": 17.20845921450151,
      "grad_norm": 1.504939079284668,
      "learning_rate": 1.562161022353611e-05,
      "loss": 1.8727,
      "step": 2670
    },
    {
      "epoch": 17.27291037260826,
      "grad_norm": 1.2910106182098389,
      "learning_rate": 1.55862673169681e-05,
      "loss": 1.859,
      "step": 2680
    },
    {
      "epoch": 17.337361530715004,
      "grad_norm": 1.4325050115585327,
      "learning_rate": 1.5550822685054475e-05,
      "loss": 1.874,
      "step": 2690
    },
    {
      "epoch": 17.401812688821753,
      "grad_norm": 1.5667049884796143,
      "learning_rate": 1.5515276973238286e-05,
      "loss": 1.8678,
      "step": 2700
    },
    {
      "epoch": 17.401812688821753,
      "eval_loss": 1.879951000213623,
      "eval_runtime": 21.041,
      "eval_samples_per_second": 76.28,
      "eval_steps_per_second": 4.8,
      "step": 2700
    },
    {
      "epoch": 17.4662638469285,
      "grad_norm": 1.5697641372680664,
      "learning_rate": 1.5479630828803235e-05,
      "loss": 1.8341,
      "step": 2710
    },
    {
      "epoch": 17.530715005035248,
      "grad_norm": 1.5967832803726196,
      "learning_rate": 1.5443884900861904e-05,
      "loss": 1.8426,
      "step": 2720
    },
    {
      "epoch": 17.595166163141993,
      "grad_norm": 1.2664175033569336,
      "learning_rate": 1.5408039840343903e-05,
      "loss": 1.8516,
      "step": 2730
    },
    {
      "epoch": 17.659617321248742,
      "grad_norm": 1.5191247463226318,
      "learning_rate": 1.5372096299984064e-05,
      "loss": 1.8435,
      "step": 2740
    },
    {
      "epoch": 17.724068479355488,
      "grad_norm": 1.5485187768936157,
      "learning_rate": 1.5336054934310502e-05,
      "loss": 1.8551,
      "step": 2750
    },
    {
      "epoch": 17.788519637462237,
      "grad_norm": 1.3325048685073853,
      "learning_rate": 1.5299916399632726e-05,
      "loss": 1.8329,
      "step": 2760
    },
    {
      "epoch": 17.852970795568982,
      "grad_norm": 1.4892091751098633,
      "learning_rate": 1.5263681354029694e-05,
      "loss": 1.8275,
      "step": 2770
    },
    {
      "epoch": 17.91742195367573,
      "grad_norm": 1.3611677885055542,
      "learning_rate": 1.5227350457337809e-05,
      "loss": 1.8343,
      "step": 2780
    },
    {
      "epoch": 17.981873111782477,
      "grad_norm": 1.3932119607925415,
      "learning_rate": 1.5190924371138908e-05,
      "loss": 1.8319,
      "step": 2790
    },
    {
      "epoch": 18.046324269889226,
      "grad_norm": 1.4689136743545532,
      "learning_rate": 1.5154403758748228e-05,
      "loss": 1.9983,
      "step": 2800
    },
    {
      "epoch": 18.046324269889226,
      "eval_loss": 1.8525444269180298,
      "eval_runtime": 19.7876,
      "eval_samples_per_second": 81.111,
      "eval_steps_per_second": 5.104,
      "step": 2800
    },
    {
      "epoch": 18.11077542799597,
      "grad_norm": 1.7211060523986816,
      "learning_rate": 1.5117789285202313e-05,
      "loss": 1.8104,
      "step": 2810
    },
    {
      "epoch": 18.17522658610272,
      "grad_norm": 1.6318062543869019,
      "learning_rate": 1.5081081617246912e-05,
      "loss": 1.821,
      "step": 2820
    },
    {
      "epoch": 18.239677744209466,
      "grad_norm": 1.2402304410934448,
      "learning_rate": 1.5044281423324826e-05,
      "loss": 1.8104,
      "step": 2830
    },
    {
      "epoch": 18.304128902316215,
      "grad_norm": 1.438472032546997,
      "learning_rate": 1.500738937356376e-05,
      "loss": 1.8259,
      "step": 2840
    },
    {
      "epoch": 18.36858006042296,
      "grad_norm": 1.4083638191223145,
      "learning_rate": 1.4970406139764092e-05,
      "loss": 1.808,
      "step": 2850
    },
    {
      "epoch": 18.43303121852971,
      "grad_norm": 1.2466082572937012,
      "learning_rate": 1.4933332395386652e-05,
      "loss": 1.8232,
      "step": 2860
    },
    {
      "epoch": 18.497482376636455,
      "grad_norm": 1.2091064453125,
      "learning_rate": 1.4896168815540464e-05,
      "loss": 1.812,
      "step": 2870
    },
    {
      "epoch": 18.561933534743204,
      "grad_norm": 1.6740643978118896,
      "learning_rate": 1.4858916076970444e-05,
      "loss": 1.7973,
      "step": 2880
    },
    {
      "epoch": 18.62638469284995,
      "grad_norm": 1.2571892738342285,
      "learning_rate": 1.4821574858045073e-05,
      "loss": 1.8096,
      "step": 2890
    },
    {
      "epoch": 18.690835850956695,
      "grad_norm": 1.3580372333526611,
      "learning_rate": 1.4784145838744067e-05,
      "loss": 1.8117,
      "step": 2900
    },
    {
      "epoch": 18.690835850956695,
      "eval_loss": 1.8296641111373901,
      "eval_runtime": 22.6507,
      "eval_samples_per_second": 70.859,
      "eval_steps_per_second": 4.459,
      "step": 2900
    },
    {
      "epoch": 18.755287009063444,
      "grad_norm": 1.4074209928512573,
      "learning_rate": 1.4746629700645955e-05,
      "loss": 1.8154,
      "step": 2910
    },
    {
      "epoch": 18.81973816717019,
      "grad_norm": 1.4824714660644531,
      "learning_rate": 1.470902712691571e-05,
      "loss": 1.8,
      "step": 2920
    },
    {
      "epoch": 18.88418932527694,
      "grad_norm": 1.4426990747451782,
      "learning_rate": 1.4671338802292274e-05,
      "loss": 1.7956,
      "step": 2930
    },
    {
      "epoch": 18.948640483383684,
      "grad_norm": 1.4474570751190186,
      "learning_rate": 1.4633565413076114e-05,
      "loss": 1.7948,
      "step": 2940
    },
    {
      "epoch": 19.013091641490433,
      "grad_norm": 1.2493271827697754,
      "learning_rate": 1.4595707647116713e-05,
      "loss": 1.9644,
      "step": 2950
    },
    {
      "epoch": 19.07754279959718,
      "grad_norm": 1.4267578125,
      "learning_rate": 1.4557766193800036e-05,
      "loss": 1.781,
      "step": 2960
    },
    {
      "epoch": 19.141993957703928,
      "grad_norm": 1.355020523071289,
      "learning_rate": 1.4519741744036e-05,
      "loss": 1.7878,
      "step": 2970
    },
    {
      "epoch": 19.206445115810673,
      "grad_norm": 1.3975870609283447,
      "learning_rate": 1.4481634990245871e-05,
      "loss": 1.7899,
      "step": 2980
    },
    {
      "epoch": 19.270896273917423,
      "grad_norm": 1.426984190940857,
      "learning_rate": 1.4443446626349662e-05,
      "loss": 1.775,
      "step": 2990
    },
    {
      "epoch": 19.335347432024168,
      "grad_norm": 1.5936192274093628,
      "learning_rate": 1.4405177347753503e-05,
      "loss": 1.7697,
      "step": 3000
    },
    {
      "epoch": 19.335347432024168,
      "eval_loss": 1.8065377473831177,
      "eval_runtime": 20.2136,
      "eval_samples_per_second": 79.402,
      "eval_steps_per_second": 4.997,
      "step": 3000
    },
    {
      "epoch": 19.399798590130917,
      "grad_norm": 1.3064271211624146,
      "learning_rate": 1.4366827851336964e-05,
      "loss": 1.7844,
      "step": 3010
    },
    {
      "epoch": 19.464249748237663,
      "grad_norm": 1.2223880290985107,
      "learning_rate": 1.4328398835440381e-05,
      "loss": 1.7739,
      "step": 3020
    },
    {
      "epoch": 19.52870090634441,
      "grad_norm": 1.5116280317306519,
      "learning_rate": 1.4289890999852126e-05,
      "loss": 1.7707,
      "step": 3030
    },
    {
      "epoch": 19.593152064451157,
      "grad_norm": 1.3726651668548584,
      "learning_rate": 1.4251305045795874e-05,
      "loss": 1.7789,
      "step": 3040
    },
    {
      "epoch": 19.657603222557906,
      "grad_norm": 1.4748300313949585,
      "learning_rate": 1.4212641675917823e-05,
      "loss": 1.7715,
      "step": 3050
    },
    {
      "epoch": 19.72205438066465,
      "grad_norm": 1.4205374717712402,
      "learning_rate": 1.4173901594273917e-05,
      "loss": 1.7823,
      "step": 3060
    },
    {
      "epoch": 19.7865055387714,
      "grad_norm": 1.2562229633331299,
      "learning_rate": 1.4135085506316997e-05,
      "loss": 1.7643,
      "step": 3070
    },
    {
      "epoch": 19.850956696878146,
      "grad_norm": 1.3158072233200073,
      "learning_rate": 1.4096194118883982e-05,
      "loss": 1.7599,
      "step": 3080
    },
    {
      "epoch": 19.915407854984895,
      "grad_norm": 1.3303531408309937,
      "learning_rate": 1.4057228140182982e-05,
      "loss": 1.7701,
      "step": 3090
    },
    {
      "epoch": 19.97985901309164,
      "grad_norm": 1.4447650909423828,
      "learning_rate": 1.4018188279780412e-05,
      "loss": 1.7587,
      "step": 3100
    },
    {
      "epoch": 19.97985901309164,
      "eval_loss": 1.7869038581848145,
      "eval_runtime": 19.8118,
      "eval_samples_per_second": 81.012,
      "eval_steps_per_second": 5.098,
      "step": 3100
    },
    {
      "epoch": 20.04431017119839,
      "grad_norm": 1.4862005710601807,
      "learning_rate": 1.3979075248588054e-05,
      "loss": 1.9169,
      "step": 3110
    },
    {
      "epoch": 20.108761329305135,
      "grad_norm": 1.2900787591934204,
      "learning_rate": 1.3939889758850138e-05,
      "loss": 1.7603,
      "step": 3120
    },
    {
      "epoch": 20.173212487411885,
      "grad_norm": 1.1818557977676392,
      "learning_rate": 1.3900632524130343e-05,
      "loss": 1.7501,
      "step": 3130
    },
    {
      "epoch": 20.23766364551863,
      "grad_norm": 1.4612010717391968,
      "learning_rate": 1.3861304259298823e-05,
      "loss": 1.7412,
      "step": 3140
    },
    {
      "epoch": 20.30211480362538,
      "grad_norm": 1.2869311571121216,
      "learning_rate": 1.3821905680519181e-05,
      "loss": 1.7442,
      "step": 3150
    },
    {
      "epoch": 20.366565961732125,
      "grad_norm": 1.4396709203720093,
      "learning_rate": 1.378243750523543e-05,
      "loss": 1.7634,
      "step": 3160
    },
    {
      "epoch": 20.431017119838874,
      "grad_norm": 1.2522846460342407,
      "learning_rate": 1.3742900452158932e-05,
      "loss": 1.7422,
      "step": 3170
    },
    {
      "epoch": 20.49546827794562,
      "grad_norm": 1.4543647766113281,
      "learning_rate": 1.3703295241255296e-05,
      "loss": 1.75,
      "step": 3180
    },
    {
      "epoch": 20.55991943605237,
      "grad_norm": 1.191945195198059,
      "learning_rate": 1.3663622593731294e-05,
      "loss": 1.7302,
      "step": 3190
    },
    {
      "epoch": 20.624370594159114,
      "grad_norm": 1.4461244344711304,
      "learning_rate": 1.3623883232021693e-05,
      "loss": 1.7381,
      "step": 3200
    },
    {
      "epoch": 20.624370594159114,
      "eval_loss": 1.77000892162323,
      "eval_runtime": 19.6254,
      "eval_samples_per_second": 81.782,
      "eval_steps_per_second": 5.146,
      "step": 3200
    },
    {
      "epoch": 20.68882175226586,
      "grad_norm": 1.2070411443710327,
      "learning_rate": 1.3584077879776132e-05,
      "loss": 1.739,
      "step": 3210
    },
    {
      "epoch": 20.75327291037261,
      "grad_norm": 1.3720213174819946,
      "learning_rate": 1.3544207261845928e-05,
      "loss": 1.7366,
      "step": 3220
    },
    {
      "epoch": 20.817724068479354,
      "grad_norm": 1.3589030504226685,
      "learning_rate": 1.3504272104270876e-05,
      "loss": 1.7427,
      "step": 3230
    },
    {
      "epoch": 20.882175226586103,
      "grad_norm": 1.1105351448059082,
      "learning_rate": 1.3464273134266037e-05,
      "loss": 1.7487,
      "step": 3240
    },
    {
      "epoch": 20.94662638469285,
      "grad_norm": 1.2116628885269165,
      "learning_rate": 1.3424211080208478e-05,
      "loss": 1.7388,
      "step": 3250
    },
    {
      "epoch": 21.011077542799597,
      "grad_norm": 1.281503438949585,
      "learning_rate": 1.338408667162404e-05,
      "loss": 1.8821,
      "step": 3260
    },
    {
      "epoch": 21.075528700906343,
      "grad_norm": 1.3089734315872192,
      "learning_rate": 1.3343900639174007e-05,
      "loss": 1.7235,
      "step": 3270
    },
    {
      "epoch": 21.139979859013092,
      "grad_norm": 1.1901700496673584,
      "learning_rate": 1.3303653714641853e-05,
      "loss": 1.7281,
      "step": 3280
    },
    {
      "epoch": 21.204431017119838,
      "grad_norm": 1.307697057723999,
      "learning_rate": 1.3263346630919875e-05,
      "loss": 1.7273,
      "step": 3290
    },
    {
      "epoch": 21.268882175226587,
      "grad_norm": 1.1000251770019531,
      "learning_rate": 1.3222980121995867e-05,
      "loss": 1.7264,
      "step": 3300
    },
    {
      "epoch": 21.268882175226587,
      "eval_loss": 1.7537308931350708,
      "eval_runtime": 20.2292,
      "eval_samples_per_second": 79.341,
      "eval_steps_per_second": 4.993,
      "step": 3300
    },
    {
      "epoch": 21.333333333333332,
      "grad_norm": 1.1835920810699463,
      "learning_rate": 1.3182554922939748e-05,
      "loss": 1.7096,
      "step": 3310
    },
    {
      "epoch": 21.39778449144008,
      "grad_norm": 1.3662201166152954,
      "learning_rate": 1.3142071769890182e-05,
      "loss": 1.716,
      "step": 3320
    },
    {
      "epoch": 21.462235649546827,
      "grad_norm": 1.1962809562683105,
      "learning_rate": 1.3101531400041163e-05,
      "loss": 1.7298,
      "step": 3330
    },
    {
      "epoch": 21.526686807653576,
      "grad_norm": 1.1254611015319824,
      "learning_rate": 1.3060934551628603e-05,
      "loss": 1.7316,
      "step": 3340
    },
    {
      "epoch": 21.59113796576032,
      "grad_norm": 1.4266749620437622,
      "learning_rate": 1.3020281963916883e-05,
      "loss": 1.7149,
      "step": 3350
    },
    {
      "epoch": 21.65558912386707,
      "grad_norm": 1.2904185056686401,
      "learning_rate": 1.2979574377185385e-05,
      "loss": 1.7324,
      "step": 3360
    },
    {
      "epoch": 21.720040281973816,
      "grad_norm": 1.357013463973999,
      "learning_rate": 1.293881253271502e-05,
      "loss": 1.7155,
      "step": 3370
    },
    {
      "epoch": 21.784491440080565,
      "grad_norm": 1.285480260848999,
      "learning_rate": 1.289799717277473e-05,
      "loss": 1.724,
      "step": 3380
    },
    {
      "epoch": 21.84894259818731,
      "grad_norm": 1.2539184093475342,
      "learning_rate": 1.2857129040607963e-05,
      "loss": 1.7297,
      "step": 3390
    },
    {
      "epoch": 21.91339375629406,
      "grad_norm": 1.2335858345031738,
      "learning_rate": 1.281620888041915e-05,
      "loss": 1.7112,
      "step": 3400
    },
    {
      "epoch": 21.91339375629406,
      "eval_loss": 1.738411784172058,
      "eval_runtime": 19.8069,
      "eval_samples_per_second": 81.032,
      "eval_steps_per_second": 5.099,
      "step": 3400
    },
    {
      "epoch": 21.977844914400805,
      "grad_norm": 1.2766244411468506,
      "learning_rate": 1.2775237437360137e-05,
      "loss": 1.6879,
      "step": 3410
    },
    {
      "epoch": 22.042296072507554,
      "grad_norm": 1.1971969604492188,
      "learning_rate": 1.2734215457516639e-05,
      "loss": 1.8451,
      "step": 3420
    },
    {
      "epoch": 22.1067472306143,
      "grad_norm": 1.2040929794311523,
      "learning_rate": 1.269314368789463e-05,
      "loss": 1.7007,
      "step": 3430
    },
    {
      "epoch": 22.17119838872105,
      "grad_norm": 1.2112030982971191,
      "learning_rate": 1.2652022876406756e-05,
      "loss": 1.7094,
      "step": 3440
    },
    {
      "epoch": 22.235649546827794,
      "grad_norm": 1.2944767475128174,
      "learning_rate": 1.2610853771858702e-05,
      "loss": 1.69,
      "step": 3450
    },
    {
      "epoch": 22.300100704934543,
      "grad_norm": 1.3381768465042114,
      "learning_rate": 1.2569637123935581e-05,
      "loss": 1.7046,
      "step": 3460
    },
    {
      "epoch": 22.36455186304129,
      "grad_norm": 1.205153465270996,
      "learning_rate": 1.2528373683188247e-05,
      "loss": 1.7066,
      "step": 3470
    },
    {
      "epoch": 22.429003021148038,
      "grad_norm": 1.2156826257705688,
      "learning_rate": 1.248706420101966e-05,
      "loss": 1.7052,
      "step": 3480
    },
    {
      "epoch": 22.493454179254783,
      "grad_norm": 1.2174049615859985,
      "learning_rate": 1.2445709429671184e-05,
      "loss": 1.688,
      "step": 3490
    },
    {
      "epoch": 22.557905337361532,
      "grad_norm": 1.1301804780960083,
      "learning_rate": 1.2404310122208895e-05,
      "loss": 1.7036,
      "step": 3500
    },
    {
      "epoch": 22.557905337361532,
      "eval_loss": 1.7260056734085083,
      "eval_runtime": 20.2022,
      "eval_samples_per_second": 79.447,
      "eval_steps_per_second": 4.999,
      "step": 3500
    },
    {
      "epoch": 22.622356495468278,
      "grad_norm": 1.2046083211898804,
      "learning_rate": 1.2362867032509871e-05,
      "loss": 1.7096,
      "step": 3510
    },
    {
      "epoch": 22.686807653575023,
      "grad_norm": 1.3079477548599243,
      "learning_rate": 1.2321380915248446e-05,
      "loss": 1.7013,
      "step": 3520
    },
    {
      "epoch": 22.751258811681772,
      "grad_norm": 1.0908950567245483,
      "learning_rate": 1.2279852525882504e-05,
      "loss": 1.6883,
      "step": 3530
    },
    {
      "epoch": 22.815709969788518,
      "grad_norm": 1.3112512826919556,
      "learning_rate": 1.2238282620639677e-05,
      "loss": 1.7084,
      "step": 3540
    },
    {
      "epoch": 22.880161127895267,
      "grad_norm": 1.1353329420089722,
      "learning_rate": 1.2196671956503611e-05,
      "loss": 1.6871,
      "step": 3550
    },
    {
      "epoch": 22.944612286002013,
      "grad_norm": 1.1687755584716797,
      "learning_rate": 1.2155021291200161e-05,
      "loss": 1.6921,
      "step": 3560
    },
    {
      "epoch": 23.00906344410876,
      "grad_norm": 1.0935124158859253,
      "learning_rate": 1.2113331383183607e-05,
      "loss": 1.8299,
      "step": 3570
    },
    {
      "epoch": 23.073514602215507,
      "grad_norm": 1.1443811655044556,
      "learning_rate": 1.2071602991622822e-05,
      "loss": 1.695,
      "step": 3580
    },
    {
      "epoch": 23.137965760322256,
      "grad_norm": 1.0561773777008057,
      "learning_rate": 1.202983687638747e-05,
      "loss": 1.6879,
      "step": 3590
    },
    {
      "epoch": 23.202416918429,
      "grad_norm": 1.1481385231018066,
      "learning_rate": 1.198803379803416e-05,
      "loss": 1.6766,
      "step": 3600
    },
    {
      "epoch": 23.202416918429,
      "eval_loss": 1.7131911516189575,
      "eval_runtime": 20.2073,
      "eval_samples_per_second": 79.427,
      "eval_steps_per_second": 4.998,
      "step": 3600
    },
    {
      "epoch": 23.26686807653575,
      "grad_norm": 1.1637824773788452,
      "learning_rate": 1.1946194517792584e-05,
      "loss": 1.6887,
      "step": 3610
    },
    {
      "epoch": 23.331319234642496,
      "grad_norm": 1.168934941291809,
      "learning_rate": 1.190431979755168e-05,
      "loss": 1.6692,
      "step": 3620
    },
    {
      "epoch": 23.395770392749245,
      "grad_norm": 1.1618574857711792,
      "learning_rate": 1.1862410399845739e-05,
      "loss": 1.6696,
      "step": 3630
    },
    {
      "epoch": 23.46022155085599,
      "grad_norm": 1.118910312652588,
      "learning_rate": 1.1820467087840526e-05,
      "loss": 1.6804,
      "step": 3640
    },
    {
      "epoch": 23.52467270896274,
      "grad_norm": 1.1120346784591675,
      "learning_rate": 1.1778490625319376e-05,
      "loss": 1.6863,
      "step": 3650
    },
    {
      "epoch": 23.589123867069485,
      "grad_norm": 1.1935482025146484,
      "learning_rate": 1.1736481776669307e-05,
      "loss": 1.6881,
      "step": 3660
    },
    {
      "epoch": 23.653575025176234,
      "grad_norm": 1.1648095846176147,
      "learning_rate": 1.1694441306867062e-05,
      "loss": 1.6813,
      "step": 3670
    },
    {
      "epoch": 23.71802618328298,
      "grad_norm": 1.0669279098510742,
      "learning_rate": 1.1652369981465218e-05,
      "loss": 1.6737,
      "step": 3680
    },
    {
      "epoch": 23.78247734138973,
      "grad_norm": 1.2142657041549683,
      "learning_rate": 1.1610268566578233e-05,
      "loss": 1.6825,
      "step": 3690
    },
    {
      "epoch": 23.846928499496475,
      "grad_norm": 1.05051851272583,
      "learning_rate": 1.1568137828868478e-05,
      "loss": 1.6606,
      "step": 3700
    },
    {
      "epoch": 23.846928499496475,
      "eval_loss": 1.7030967473983765,
      "eval_runtime": 20.2178,
      "eval_samples_per_second": 79.385,
      "eval_steps_per_second": 4.996,
      "step": 3700
    },
    {
      "epoch": 23.911379657603224,
      "grad_norm": 1.1508402824401855,
      "learning_rate": 1.15259785355323e-05,
      "loss": 1.6664,
      "step": 3710
    },
    {
      "epoch": 23.97583081570997,
      "grad_norm": 1.260407567024231,
      "learning_rate": 1.1483791454286027e-05,
      "loss": 1.6875,
      "step": 3720
    },
    {
      "epoch": 24.040281973816718,
      "grad_norm": 1.032926082611084,
      "learning_rate": 1.1441577353352023e-05,
      "loss": 1.7966,
      "step": 3730
    },
    {
      "epoch": 24.104733131923464,
      "grad_norm": 1.0912493467330933,
      "learning_rate": 1.1399337001444658e-05,
      "loss": 1.6737,
      "step": 3740
    },
    {
      "epoch": 24.169184290030213,
      "grad_norm": 1.026868462562561,
      "learning_rate": 1.1357071167756341e-05,
      "loss": 1.657,
      "step": 3750
    },
    {
      "epoch": 24.23363544813696,
      "grad_norm": 1.212310552597046,
      "learning_rate": 1.13147806219435e-05,
      "loss": 1.673,
      "step": 3760
    },
    {
      "epoch": 24.298086606243707,
      "grad_norm": 1.0902752876281738,
      "learning_rate": 1.1272466134112562e-05,
      "loss": 1.6793,
      "step": 3770
    },
    {
      "epoch": 24.362537764350453,
      "grad_norm": 1.1671861410140991,
      "learning_rate": 1.1230128474805948e-05,
      "loss": 1.664,
      "step": 3780
    },
    {
      "epoch": 24.426988922457202,
      "grad_norm": 1.0966593027114868,
      "learning_rate": 1.1187768414988015e-05,
      "loss": 1.6649,
      "step": 3790
    },
    {
      "epoch": 24.491440080563947,
      "grad_norm": 1.0915180444717407,
      "learning_rate": 1.114538672603104e-05,
      "loss": 1.6749,
      "step": 3800
    },
    {
      "epoch": 24.491440080563947,
      "eval_loss": 1.6930159330368042,
      "eval_runtime": 19.486,
      "eval_samples_per_second": 82.367,
      "eval_steps_per_second": 5.183,
      "step": 3800
    },
    {
      "epoch": 24.555891238670696,
      "grad_norm": 1.0787924528121948,
      "learning_rate": 1.1102984179701157e-05,
      "loss": 1.659,
      "step": 3810
    },
    {
      "epoch": 24.620342396777442,
      "grad_norm": 1.0481791496276855,
      "learning_rate": 1.1060561548144321e-05,
      "loss": 1.6558,
      "step": 3820
    },
    {
      "epoch": 24.68479355488419,
      "grad_norm": 1.06913161277771,
      "learning_rate": 1.1018119603872228e-05,
      "loss": 1.6551,
      "step": 3830
    },
    {
      "epoch": 24.749244712990937,
      "grad_norm": 0.9699168801307678,
      "learning_rate": 1.0975659119748265e-05,
      "loss": 1.6579,
      "step": 3840
    },
    {
      "epoch": 24.813695871097686,
      "grad_norm": 1.0122530460357666,
      "learning_rate": 1.0933180868973414e-05,
      "loss": 1.6517,
      "step": 3850
    },
    {
      "epoch": 24.87814702920443,
      "grad_norm": 0.935750424861908,
      "learning_rate": 1.08906856250722e-05,
      "loss": 1.6567,
      "step": 3860
    },
    {
      "epoch": 24.942598187311177,
      "grad_norm": 0.929023027420044,
      "learning_rate": 1.0848174161878584e-05,
      "loss": 1.6501,
      "step": 3870
    },
    {
      "epoch": 25.007049345417926,
      "grad_norm": 1.1005076169967651,
      "learning_rate": 1.080564725352188e-05,
      "loss": 1.8029,
      "step": 3880
    },
    {
      "epoch": 25.07150050352467,
      "grad_norm": 0.9921239018440247,
      "learning_rate": 1.076310567441266e-05,
      "loss": 1.6426,
      "step": 3890
    },
    {
      "epoch": 25.13595166163142,
      "grad_norm": 1.0867966413497925,
      "learning_rate": 1.072055019922864e-05,
      "loss": 1.6545,
      "step": 3900
    },
    {
      "epoch": 25.13595166163142,
      "eval_loss": 1.6841835975646973,
      "eval_runtime": 19.9451,
      "eval_samples_per_second": 80.471,
      "eval_steps_per_second": 5.064,
      "step": 3900
    },
    {
      "epoch": 25.200402819738166,
      "grad_norm": 0.9666146636009216,
      "learning_rate": 1.067798160290059e-05,
      "loss": 1.6428,
      "step": 3910
    },
    {
      "epoch": 25.264853977844915,
      "grad_norm": 1.1212939023971558,
      "learning_rate": 1.0635400660598214e-05,
      "loss": 1.6361,
      "step": 3920
    },
    {
      "epoch": 25.32930513595166,
      "grad_norm": 1.075994610786438,
      "learning_rate": 1.0592808147716032e-05,
      "loss": 1.6567,
      "step": 3930
    },
    {
      "epoch": 25.39375629405841,
      "grad_norm": 0.9849887490272522,
      "learning_rate": 1.0550204839859265e-05,
      "loss": 1.6587,
      "step": 3940
    },
    {
      "epoch": 25.458207452165155,
      "grad_norm": 0.932995617389679,
      "learning_rate": 1.0507591512829707e-05,
      "loss": 1.6471,
      "step": 3950
    },
    {
      "epoch": 25.522658610271904,
      "grad_norm": 1.0005803108215332,
      "learning_rate": 1.0464968942611608e-05,
      "loss": 1.6508,
      "step": 3960
    },
    {
      "epoch": 25.58710976837865,
      "grad_norm": 1.2026358842849731,
      "learning_rate": 1.0422337905357523e-05,
      "loss": 1.6506,
      "step": 3970
    },
    {
      "epoch": 25.6515609264854,
      "grad_norm": 1.075043797492981,
      "learning_rate": 1.0379699177374199e-05,
      "loss": 1.6372,
      "step": 3980
    },
    {
      "epoch": 25.716012084592144,
      "grad_norm": 0.983676016330719,
      "learning_rate": 1.0337053535108427e-05,
      "loss": 1.6494,
      "step": 3990
    },
    {
      "epoch": 25.780463242698893,
      "grad_norm": 0.9707674384117126,
      "learning_rate": 1.0294401755132912e-05,
      "loss": 1.6509,
      "step": 4000
    },
    {
      "epoch": 25.780463242698893,
      "eval_loss": 1.6760313510894775,
      "eval_runtime": 19.7916,
      "eval_samples_per_second": 81.095,
      "eval_steps_per_second": 5.103,
      "step": 4000
    },
    {
      "epoch": 25.84491440080564,
      "grad_norm": 1.0340772867202759,
      "learning_rate": 1.0251744614132117e-05,
      "loss": 1.6614,
      "step": 4010
    },
    {
      "epoch": 25.909365558912388,
      "grad_norm": 0.9816955924034119,
      "learning_rate": 1.0209082888888143e-05,
      "loss": 1.6327,
      "step": 4020
    },
    {
      "epoch": 25.973816717019133,
      "grad_norm": 0.9904563426971436,
      "learning_rate": 1.0166417356266546e-05,
      "loss": 1.6408,
      "step": 4030
    },
    {
      "epoch": 26.038267875125882,
      "grad_norm": 0.9868249893188477,
      "learning_rate": 1.0123748793202242e-05,
      "loss": 1.7511,
      "step": 4040
    },
    {
      "epoch": 26.102719033232628,
      "grad_norm": 0.9166064262390137,
      "learning_rate": 1.0081077976685307e-05,
      "loss": 1.6361,
      "step": 4050
    },
    {
      "epoch": 26.167170191339377,
      "grad_norm": 0.9603624939918518,
      "learning_rate": 1.0038405683746868e-05,
      "loss": 1.6374,
      "step": 4060
    },
    {
      "epoch": 26.231621349446122,
      "grad_norm": 0.9980469346046448,
      "learning_rate": 9.995732691444932e-06,
      "loss": 1.6464,
      "step": 4070
    },
    {
      "epoch": 26.29607250755287,
      "grad_norm": 1.0243116617202759,
      "learning_rate": 9.953059776850238e-06,
      "loss": 1.6364,
      "step": 4080
    },
    {
      "epoch": 26.360523665659617,
      "grad_norm": 0.9727787971496582,
      "learning_rate": 9.910387717032115e-06,
      "loss": 1.6366,
      "step": 4090
    },
    {
      "epoch": 26.424974823766366,
      "grad_norm": 0.9883456826210022,
      "learning_rate": 9.86771728904433e-06,
      "loss": 1.6423,
      "step": 4100
    },
    {
      "epoch": 26.424974823766366,
      "eval_loss": 1.6685707569122314,
      "eval_runtime": 21.0152,
      "eval_samples_per_second": 76.373,
      "eval_steps_per_second": 4.806,
      "step": 4100
    },
    {
      "epoch": 26.48942598187311,
      "grad_norm": 0.9553434252738953,
      "learning_rate": 9.82504926991092e-06,
      "loss": 1.6439,
      "step": 4110
    },
    {
      "epoch": 26.55387713997986,
      "grad_norm": 1.038282036781311,
      "learning_rate": 9.782384436612072e-06,
      "loss": 1.641,
      "step": 4120
    },
    {
      "epoch": 26.618328298086606,
      "grad_norm": 1.0809876918792725,
      "learning_rate": 9.73972356606995e-06,
      "loss": 1.6335,
      "step": 4130
    },
    {
      "epoch": 26.682779456193355,
      "grad_norm": 1.0616182088851929,
      "learning_rate": 9.697067435134564e-06,
      "loss": 1.6168,
      "step": 4140
    },
    {
      "epoch": 26.7472306143001,
      "grad_norm": 0.8756535649299622,
      "learning_rate": 9.654416820569618e-06,
      "loss": 1.6203,
      "step": 4150
    },
    {
      "epoch": 26.81168177240685,
      "grad_norm": 1.1267576217651367,
      "learning_rate": 9.611772499038345e-06,
      "loss": 1.6279,
      "step": 4160
    },
    {
      "epoch": 26.876132930513595,
      "grad_norm": 0.9867852926254272,
      "learning_rate": 9.569135247089401e-06,
      "loss": 1.6323,
      "step": 4170
    },
    {
      "epoch": 26.94058408862034,
      "grad_norm": 0.9736083149909973,
      "learning_rate": 9.526505841142702e-06,
      "loss": 1.6328,
      "step": 4180
    },
    {
      "epoch": 27.00503524672709,
      "grad_norm": 1.8718211650848389,
      "learning_rate": 9.48388505747529e-06,
      "loss": 1.7769,
      "step": 4190
    },
    {
      "epoch": 27.069486404833835,
      "grad_norm": 0.9595508575439453,
      "learning_rate": 9.441273672207187e-06,
      "loss": 1.6209,
      "step": 4200
    },
    {
      "epoch": 27.069486404833835,
      "eval_loss": 1.6613610982894897,
      "eval_runtime": 20.2063,
      "eval_samples_per_second": 79.431,
      "eval_steps_per_second": 4.998,
      "step": 4200
    }
  ],
  "logging_steps": 10,
  "max_steps": 7750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3849836110034764e+20,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}