{
  "best_metric": 5.585614204406738,
  "best_model_checkpoint": "./iiif_manuscripts_label_ge_50_convnext/checkpoint-12228",
  "epoch": 30.0,
  "global_step": 61140,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 0.00019869152764147858,
      "loss": 7.9686,
      "step": 400
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019738632646385348,
      "loss": 7.2475,
      "step": 800
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019607785410533205,
      "loss": 6.9462,
      "step": 1200
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001947693817468106,
      "loss": 6.7615,
      "step": 1600
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001934641805691855,
      "loss": 6.5753,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.0015526657039264577,
      "eval_loss": 6.412076950073242,
      "eval_runtime": 102.5706,
      "eval_samples_per_second": 224.353,
      "eval_steps_per_second": 3.51,
      "step": 2038
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00019215570821066406,
      "loss": 6.324,
      "step": 2400
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00019084723585214262,
      "loss": 6.2462,
      "step": 2800
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00018953876349362122,
      "loss": 6.1389,
      "step": 3200
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00018823029113509978,
      "loss": 6.0919,
      "step": 3600
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018692508995747466,
      "loss": 5.9865,
      "step": 4000
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.002120415247137495,
      "eval_loss": 5.946579456329346,
      "eval_runtime": 101.947,
      "eval_samples_per_second": 225.725,
      "eval_steps_per_second": 3.531,
      "step": 4076
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00018561661759895323,
      "loss": 5.7381,
      "step": 4400
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0001843114164213281,
      "loss": 5.6812,
      "step": 4800
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0001830029440628067,
      "loss": 5.6661,
      "step": 5200
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00018169447170428526,
      "loss": 5.6508,
      "step": 5600
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.00018038599934576383,
      "loss": 5.6521,
      "step": 6000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.002850390938715168,
      "eval_loss": 5.764512062072754,
      "eval_runtime": 99.5338,
      "eval_samples_per_second": 231.198,
      "eval_steps_per_second": 3.617,
      "step": 6114
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0001790775269872424,
      "loss": 5.3744,
      "step": 6400
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.00017776905462872097,
      "loss": 5.3159,
      "step": 6800
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.00017646058227019956,
      "loss": 5.3162,
      "step": 7200
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.00017515210991167813,
      "loss": 5.3139,
      "step": 7600
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.0001738436375531567,
      "loss": 5.3123,
      "step": 8000
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.0032535394402472113,
      "eval_loss": 5.688981056213379,
      "eval_runtime": 98.5752,
      "eval_samples_per_second": 233.446,
      "eval_steps_per_second": 3.652,
      "step": 8152
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.00017253516519463526,
      "loss": 5.0622,
      "step": 8400
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00017122996401701014,
      "loss": 4.9616,
      "step": 8800
    },
    {
      "epoch": 4.51,
      "learning_rate": 0.00016992476283938504,
      "loss": 5.0117,
      "step": 9200
    },
    {
      "epoch": 4.71,
      "learning_rate": 0.0001686162904808636,
      "loss": 5.0168,
      "step": 9600
    },
    {
      "epoch": 4.91,
      "learning_rate": 0.00016730781812234217,
      "loss": 5.0337,
      "step": 10000
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.003394306441476293,
      "eval_loss": 5.6691813468933105,
      "eval_runtime": 106.5631,
      "eval_samples_per_second": 215.947,
      "eval_steps_per_second": 3.378,
      "step": 10190
    },
    {
      "epoch": 5.1,
      "learning_rate": 0.00016599934576382074,
      "loss": 4.8182,
      "step": 10400
    },
    {
      "epoch": 5.3,
      "learning_rate": 0.0001646908734052993,
      "loss": 4.6457,
      "step": 10800
    },
    {
      "epoch": 5.5,
      "learning_rate": 0.0001633824010467779,
      "loss": 4.6918,
      "step": 11200
    },
    {
      "epoch": 5.69,
      "learning_rate": 0.00016207719986915278,
      "loss": 4.7441,
      "step": 11600
    },
    {
      "epoch": 5.89,
      "learning_rate": 0.00016076872751063134,
      "loss": 4.743,
      "step": 12000
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.003679096298311292,
      "eval_loss": 5.585614204406738,
      "eval_runtime": 103.5743,
      "eval_samples_per_second": 222.179,
      "eval_steps_per_second": 3.476,
      "step": 12228
    },
    {
      "epoch": 6.08,
      "learning_rate": 0.0001594602551521099,
      "loss": 4.5617,
      "step": 12400
    },
    {
      "epoch": 6.28,
      "learning_rate": 0.00015815178279358848,
      "loss": 4.3335,
      "step": 12800
    },
    {
      "epoch": 6.48,
      "learning_rate": 0.00015684331043506705,
      "loss": 4.3856,
      "step": 13200
    },
    {
      "epoch": 6.67,
      "learning_rate": 0.00015553810925744195,
      "loss": 4.3955,
      "step": 13600
    },
    {
      "epoch": 6.87,
      "learning_rate": 0.00015422963689892052,
      "loss": 4.4387,
      "step": 14000
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.004160028924400371,
      "eval_loss": 5.596897125244141,
      "eval_runtime": 104.1264,
      "eval_samples_per_second": 221.001,
      "eval_steps_per_second": 3.457,
      "step": 14266
    },
    {
      "epoch": 7.07,
      "learning_rate": 0.00015292116454039908,
      "loss": 4.3069,
      "step": 14400
    },
    {
      "epoch": 7.26,
      "learning_rate": 0.00015161269218187765,
      "loss": 3.9931,
      "step": 14800
    },
    {
      "epoch": 7.46,
      "learning_rate": 0.00015030421982335622,
      "loss": 4.0845,
      "step": 15200
    },
    {
      "epoch": 7.65,
      "learning_rate": 0.00014899901864573112,
      "loss": 4.1085,
      "step": 15600
    },
    {
      "epoch": 7.85,
      "learning_rate": 0.00014769054628720969,
      "loss": 4.1422,
      "step": 16000
    },
    {
      "epoch": 8.0,
      "eval_f1": 0.004266402669461609,
      "eval_loss": 5.671103477478027,
      "eval_runtime": 107.1495,
      "eval_samples_per_second": 214.765,
      "eval_steps_per_second": 3.36,
      "step": 16304
    },
    {
      "epoch": 8.05,
      "learning_rate": 0.00014638207392868825,
      "loss": 4.0586,
      "step": 16400
    },
    {
      "epoch": 8.24,
      "learning_rate": 0.00014507360157016682,
      "loss": 3.6644,
      "step": 16800
    },
    {
      "epoch": 8.44,
      "learning_rate": 0.00014376512921164541,
      "loss": 3.7197,
      "step": 17200
    },
    {
      "epoch": 8.64,
      "learning_rate": 0.0001424599280340203,
      "loss": 3.796,
      "step": 17600
    },
    {
      "epoch": 8.83,
      "learning_rate": 0.00014115145567549886,
      "loss": 3.8372,
      "step": 18000
    },
    {
      "epoch": 9.0,
      "eval_f1": 0.004420450860447925,
      "eval_loss": 5.676065921783447,
      "eval_runtime": 104.689,
      "eval_samples_per_second": 219.813,
      "eval_steps_per_second": 3.439,
      "step": 18342
    },
    {
      "epoch": 9.03,
      "learning_rate": 0.00013984298331697742,
      "loss": 3.7798,
      "step": 18400
    },
    {
      "epoch": 9.22,
      "learning_rate": 0.000138534510958456,
      "loss": 3.3063,
      "step": 18800
    },
    {
      "epoch": 9.42,
      "learning_rate": 0.00013722603859993459,
      "loss": 3.389,
      "step": 19200
    },
    {
      "epoch": 9.62,
      "learning_rate": 0.00013592083742230946,
      "loss": 3.4438,
      "step": 19600
    },
    {
      "epoch": 9.81,
      "learning_rate": 0.00013461236506378803,
      "loss": 3.5244,
      "step": 20000
    },
    {
      "epoch": 10.0,
      "eval_f1": 0.004167666888342888,
      "eval_loss": 5.8468732833862305,
      "eval_runtime": 105.8207,
      "eval_samples_per_second": 217.462,
      "eval_steps_per_second": 3.402,
      "step": 20380
    },
    {
      "epoch": 10.01,
      "learning_rate": 0.0001333038927052666,
      "loss": 3.542,
      "step": 20400
    },
    {
      "epoch": 10.21,
      "learning_rate": 0.0001319954203467452,
      "loss": 3.023,
      "step": 20800
    },
    {
      "epoch": 10.4,
      "learning_rate": 0.00013068694798822376,
      "loss": 3.0507,
      "step": 21200
    },
    {
      "epoch": 10.6,
      "learning_rate": 0.00012938174681059863,
      "loss": 3.132,
      "step": 21600
    },
    {
      "epoch": 10.79,
      "learning_rate": 0.0001280732744520772,
      "loss": 3.1857,
      "step": 22000
    },
    {
      "epoch": 10.99,
      "learning_rate": 0.0001267648020935558,
      "loss": 3.2321,
      "step": 22400
    },
    {
      "epoch": 11.0,
      "eval_f1": 0.004533528112732692,
      "eval_loss": 5.87736177444458,
      "eval_runtime": 105.6895,
      "eval_samples_per_second": 217.732,
      "eval_steps_per_second": 3.406,
      "step": 22418
    },
    {
      "epoch": 11.19,
      "learning_rate": 0.00012545632973503436,
      "loss": 2.705,
      "step": 22800
    },
    {
      "epoch": 11.38,
      "learning_rate": 0.00012414785737651293,
      "loss": 2.7688,
      "step": 23200
    },
    {
      "epoch": 11.58,
      "learning_rate": 0.0001228393850179915,
      "loss": 2.8223,
      "step": 23600
    },
    {
      "epoch": 11.78,
      "learning_rate": 0.00012153745502126268,
      "loss": 2.8647,
      "step": 24000
    },
    {
      "epoch": 11.97,
      "learning_rate": 0.00012022898266274125,
      "loss": 2.9004,
      "step": 24400
    },
    {
      "epoch": 12.0,
      "eval_f1": 0.004656155608992903,
      "eval_loss": 6.118570804595947,
      "eval_runtime": 103.8763,
      "eval_samples_per_second": 221.533,
      "eval_steps_per_second": 3.466,
      "step": 24456
    },
    {
      "epoch": 12.17,
      "learning_rate": 0.00011892051030421983,
      "loss": 2.4632,
      "step": 24800
    },
    {
      "epoch": 12.37,
      "learning_rate": 0.0001176120379456984,
      "loss": 2.4821,
      "step": 25200
    },
    {
      "epoch": 12.56,
      "learning_rate": 0.00011630356558717696,
      "loss": 2.5336,
      "step": 25600
    },
    {
      "epoch": 12.76,
      "learning_rate": 0.00011499509322865556,
      "loss": 2.5646,
      "step": 26000
    },
    {
      "epoch": 12.95,
      "learning_rate": 0.00011368662087013414,
      "loss": 2.5937,
      "step": 26400
    },
    {
      "epoch": 13.0,
      "eval_f1": 0.004601475826989957,
      "eval_loss": 6.239825248718262,
      "eval_runtime": 107.5857,
      "eval_samples_per_second": 213.895,
      "eval_steps_per_second": 3.346,
      "step": 26494
    },
    {
      "epoch": 13.15,
      "learning_rate": 0.0001123781485116127,
      "loss": 2.231,
      "step": 26800
    },
    {
      "epoch": 13.35,
      "learning_rate": 0.00011106967615309127,
      "loss": 2.149,
      "step": 27200
    },
    {
      "epoch": 13.54,
      "learning_rate": 0.00010976120379456985,
      "loss": 2.2258,
      "step": 27600
    },
    {
      "epoch": 13.74,
      "learning_rate": 0.00010845600261694473,
      "loss": 2.2778,
      "step": 28000
    },
    {
      "epoch": 13.94,
      "learning_rate": 0.00010715080143931959,
      "loss": 2.2983,
      "step": 28400
    },
    {
      "epoch": 14.0,
      "eval_f1": 0.004861690958274386,
      "eval_loss": 6.3732194900512695,
      "eval_runtime": 106.2123,
      "eval_samples_per_second": 216.66,
      "eval_steps_per_second": 3.389,
      "step": 28532
    },
    {
      "epoch": 14.13,
      "learning_rate": 0.00010584232908079817,
      "loss": 2.0303,
      "step": 28800
    },
    {
      "epoch": 14.33,
      "learning_rate": 0.00010453385672227676,
      "loss": 1.9106,
      "step": 29200
    },
    {
      "epoch": 14.52,
      "learning_rate": 0.00010322538436375533,
      "loss": 1.9854,
      "step": 29600
    },
    {
      "epoch": 14.72,
      "learning_rate": 0.0001019169120052339,
      "loss": 2.0189,
      "step": 30000
    },
    {
      "epoch": 14.92,
      "learning_rate": 0.00010060843964671246,
      "loss": 2.0611,
      "step": 30400
    },
    {
      "epoch": 15.0,
      "eval_f1": 0.004546370201449235,
      "eval_loss": 6.502394199371338,
      "eval_runtime": 106.7026,
      "eval_samples_per_second": 215.665,
      "eval_steps_per_second": 3.374,
      "step": 30570
    },
    {
      "epoch": 15.11,
      "learning_rate": 9.929996728819105e-05,
      "loss": 1.8034,
      "step": 30800
    },
    {
      "epoch": 15.31,
      "learning_rate": 9.799476611056592e-05,
      "loss": 1.7043,
      "step": 31200
    },
    {
      "epoch": 15.51,
      "learning_rate": 9.668629375204449e-05,
      "loss": 1.731,
      "step": 31600
    },
    {
      "epoch": 15.7,
      "learning_rate": 9.537782139352307e-05,
      "loss": 1.7611,
      "step": 32000
    },
    {
      "epoch": 15.9,
      "learning_rate": 9.406934903500164e-05,
      "loss": 1.8153,
      "step": 32400
    },
    {
      "epoch": 16.0,
      "eval_f1": 0.004668311047328734,
      "eval_loss": 6.658481121063232,
      "eval_runtime": 106.8749,
      "eval_samples_per_second": 215.317,
      "eval_steps_per_second": 3.368,
      "step": 32608
    },
    {
      "epoch": 16.09,
      "learning_rate": 9.276087667648022e-05,
      "loss": 1.6449,
      "step": 32800
    },
    {
      "epoch": 16.29,
      "learning_rate": 9.145567549885509e-05,
      "loss": 1.4815,
      "step": 33200
    },
    {
      "epoch": 16.49,
      "learning_rate": 9.014720314033367e-05,
      "loss": 1.5311,
      "step": 33600
    },
    {
      "epoch": 16.68,
      "learning_rate": 8.883873078181224e-05,
      "loss": 1.5722,
      "step": 34000
    },
    {
      "epoch": 16.88,
      "learning_rate": 8.75302584232908e-05,
      "loss": 1.6075,
      "step": 34400
    },
    {
      "epoch": 17.0,
      "eval_f1": 0.004309847995944622,
      "eval_loss": 6.833281993865967,
      "eval_runtime": 104.6363,
      "eval_samples_per_second": 219.924,
      "eval_steps_per_second": 3.44,
      "step": 34646
    },
    {
      "epoch": 17.08,
      "learning_rate": 8.622178606476939e-05,
      "loss": 1.4882,
      "step": 34800
    },
    {
      "epoch": 17.27,
      "learning_rate": 8.491331370624795e-05,
      "loss": 1.3091,
      "step": 35200
    },
    {
      "epoch": 17.47,
      "learning_rate": 8.360811252862284e-05,
      "loss": 1.3476,
      "step": 35600
    },
    {
      "epoch": 17.66,
      "learning_rate": 8.229964017010141e-05,
      "loss": 1.3937,
      "step": 36000
    },
    {
      "epoch": 17.86,
      "learning_rate": 8.099116781157998e-05,
      "loss": 1.4342,
      "step": 36400
    },
    {
      "epoch": 18.0,
      "eval_f1": 0.004381425028314271,
      "eval_loss": 6.952894687652588,
      "eval_runtime": 106.0924,
      "eval_samples_per_second": 216.905,
      "eval_steps_per_second": 3.393,
      "step": 36684
    },
    {
      "epoch": 18.06,
      "learning_rate": 7.968269545305856e-05,
      "loss": 1.3502,
      "step": 36800
    },
    {
      "epoch": 18.25,
      "learning_rate": 7.837422309453714e-05,
      "loss": 1.1797,
      "step": 37200
    },
    {
      "epoch": 18.45,
      "learning_rate": 7.70657507360157e-05,
      "loss": 1.234,
      "step": 37600
    },
    {
      "epoch": 18.65,
      "learning_rate": 7.575727837749427e-05,
      "loss": 1.2631,
      "step": 38000
    },
    {
      "epoch": 18.84,
      "learning_rate": 7.445207719986915e-05,
      "loss": 1.2614,
      "step": 38400
    },
    {
      "epoch": 19.0,
      "eval_f1": 0.0045730417031692346,
      "eval_loss": 7.112914085388184,
      "eval_runtime": 106.9736,
      "eval_samples_per_second": 215.118,
      "eval_steps_per_second": 3.365,
      "step": 38722
    },
    {
      "epoch": 19.04,
      "learning_rate": 7.314360484134773e-05,
      "loss": 1.2353,
      "step": 38800
    },
    {
      "epoch": 19.23,
      "learning_rate": 7.183513248282631e-05,
      "loss": 1.0694,
      "step": 39200
    },
    {
      "epoch": 19.43,
      "learning_rate": 7.052666012430488e-05,
      "loss": 1.0988,
      "step": 39600
    },
    {
      "epoch": 19.63,
      "learning_rate": 6.921818776578344e-05,
      "loss": 1.125,
      "step": 40000
    },
    {
      "epoch": 19.82,
      "learning_rate": 6.791298658815833e-05,
      "loss": 1.1463,
      "step": 40400
    },
    {
      "epoch": 20.0,
      "eval_f1": 0.003947589716911441,
      "eval_loss": 7.197678089141846,
      "eval_runtime": 105.623,
      "eval_samples_per_second": 217.869,
      "eval_steps_per_second": 3.408,
      "step": 40760
    },
    {
      "epoch": 20.02,
      "learning_rate": 6.66045142296369e-05,
      "loss": 1.1429,
      "step": 40800
    },
    {
      "epoch": 20.22,
      "learning_rate": 6.529604187111548e-05,
      "loss": 0.9563,
      "step": 41200
    },
    {
      "epoch": 20.41,
      "learning_rate": 6.398756951259405e-05,
      "loss": 1.0085,
      "step": 41600
    },
    {
      "epoch": 20.61,
      "learning_rate": 6.267909715407262e-05,
      "loss": 1.0361,
      "step": 42000
    },
    {
      "epoch": 20.8,
      "learning_rate": 6.13738959764475e-05,
      "loss": 1.0387,
      "step": 42400
    },
    {
      "epoch": 21.0,
      "eval_f1": 0.004382334218618053,
      "eval_loss": 7.270018100738525,
      "eval_runtime": 104.2085,
      "eval_samples_per_second": 220.827,
      "eval_steps_per_second": 3.455,
      "step": 42798
    },
    {
      "epoch": 21.0,
      "learning_rate": 6.006542361792608e-05,
      "loss": 1.0315,
      "step": 42800
    },
    {
      "epoch": 21.2,
      "learning_rate": 5.8756951259404646e-05,
      "loss": 0.8991,
      "step": 43200
    },
    {
      "epoch": 21.39,
      "learning_rate": 5.744847890088322e-05,
      "loss": 0.907,
      "step": 43600
    },
    {
      "epoch": 21.59,
      "learning_rate": 5.61400065423618e-05,
      "loss": 0.9289,
      "step": 44000
    },
    {
      "epoch": 21.79,
      "learning_rate": 5.4831534183840375e-05,
      "loss": 0.9542,
      "step": 44400
    },
    {
      "epoch": 21.98,
      "learning_rate": 5.352306182531894e-05,
      "loss": 0.9635,
      "step": 44800
    },
    {
      "epoch": 22.0,
      "eval_f1": 0.0039901118782942465,
      "eval_loss": 7.337534427642822,
      "eval_runtime": 105.9291,
      "eval_samples_per_second": 217.24,
      "eval_steps_per_second": 3.399,
      "step": 44836
    },
    {
      "epoch": 22.18,
      "learning_rate": 5.2214589466797516e-05,
      "loss": 0.8345,
      "step": 45200
    },
    {
      "epoch": 22.37,
      "learning_rate": 5.090938828917239e-05,
      "loss": 0.8363,
      "step": 45600
    },
    {
      "epoch": 22.57,
      "learning_rate": 4.9600915930650965e-05,
      "loss": 0.8663,
      "step": 46000
    },
    {
      "epoch": 22.77,
      "learning_rate": 4.829244357212954e-05,
      "loss": 0.8618,
      "step": 46400
    },
    {
      "epoch": 22.96,
      "learning_rate": 4.698397121360811e-05,
      "loss": 0.8872,
      "step": 46800
    },
    {
      "epoch": 23.0,
      "eval_f1": 0.003935013281715019,
      "eval_loss": 7.400303363800049,
      "eval_runtime": 115.223,
      "eval_samples_per_second": 199.717,
      "eval_steps_per_second": 3.124,
      "step": 46874
    },
    {
      "epoch": 23.16,
      "learning_rate": 4.5675498855086694e-05,
      "loss": 0.7942,
      "step": 47200
    },
    {
      "epoch": 23.36,
      "learning_rate": 4.437029767746156e-05,
      "loss": 0.803,
      "step": 47600
    },
    {
      "epoch": 23.55,
      "learning_rate": 4.306182531894014e-05,
      "loss": 0.7919,
      "step": 48000
    },
    {
      "epoch": 23.75,
      "learning_rate": 4.175335296041871e-05,
      "loss": 0.8018,
      "step": 48400
    },
    {
      "epoch": 23.95,
      "learning_rate": 4.044488060189729e-05,
      "loss": 0.8156,
      "step": 48800
    },
    {
      "epoch": 24.0,
      "eval_f1": 0.003947052519025861,
      "eval_loss": 7.488423824310303,
      "eval_runtime": 106.3546,
      "eval_samples_per_second": 216.371,
      "eval_steps_per_second": 3.385,
      "step": 48912
    },
    {
      "epoch": 24.14,
      "learning_rate": 3.913640824337586e-05,
      "loss": 0.7462,
      "step": 49200
    },
    {
      "epoch": 24.34,
      "learning_rate": 3.782793588485443e-05,
      "loss": 0.7192,
      "step": 49600
    },
    {
      "epoch": 24.53,
      "learning_rate": 3.6519463526333006e-05,
      "loss": 0.7476,
      "step": 50000
    },
    {
      "epoch": 24.73,
      "learning_rate": 3.521099116781158e-05,
      "loss": 0.7602,
      "step": 50400
    },
    {
      "epoch": 24.93,
      "learning_rate": 3.3902518809290154e-05,
      "loss": 0.7544,
      "step": 50800
    },
    {
      "epoch": 25.0,
      "eval_f1": 0.003877386201510839,
      "eval_loss": 7.476434707641602,
      "eval_runtime": 105.8268,
      "eval_samples_per_second": 217.45,
      "eval_steps_per_second": 3.402,
      "step": 50950
    },
    {
      "epoch": 25.12,
      "learning_rate": 3.259404645076873e-05,
      "loss": 0.7114,
      "step": 51200
    },
    {
      "epoch": 25.32,
      "learning_rate": 3.12855740922473e-05,
      "loss": 0.683,
      "step": 51600
    },
    {
      "epoch": 25.52,
      "learning_rate": 2.998037291462218e-05,
      "loss": 0.6972,
      "step": 52000
    },
    {
      "epoch": 25.71,
      "learning_rate": 2.8671900556100754e-05,
      "loss": 0.7087,
      "step": 52400
    },
    {
      "epoch": 25.91,
      "learning_rate": 2.736669937847563e-05,
      "loss": 0.6893,
      "step": 52800
    },
    {
      "epoch": 26.0,
      "eval_f1": 0.004237152566679073,
      "eval_loss": 7.515334606170654,
      "eval_runtime": 104.8855,
      "eval_samples_per_second": 219.401,
      "eval_steps_per_second": 3.432,
      "step": 52988
    },
    {
      "epoch": 26.1,
      "learning_rate": 2.6058227019954207e-05,
      "loss": 0.6589,
      "step": 53200
    },
    {
      "epoch": 26.3,
      "learning_rate": 2.4749754661432777e-05,
      "loss": 0.6609,
      "step": 53600
    },
    {
      "epoch": 26.5,
      "learning_rate": 2.344128230291135e-05,
      "loss": 0.6485,
      "step": 54000
    },
    {
      "epoch": 26.69,
      "learning_rate": 2.2132809944389925e-05,
      "loss": 0.667,
      "step": 54400
    },
    {
      "epoch": 26.89,
      "learning_rate": 2.08243375858685e-05,
      "loss": 0.6767,
      "step": 54800
    },
    {
      "epoch": 27.0,
      "eval_f1": 0.004296555657055152,
      "eval_loss": 7.542654514312744,
      "eval_runtime": 104.6361,
      "eval_samples_per_second": 219.924,
      "eval_steps_per_second": 3.44,
      "step": 55026
    },
    {
      "epoch": 27.09,
      "learning_rate": 1.9515865227347073e-05,
      "loss": 0.639,
      "step": 55200
    },
    {
      "epoch": 27.28,
      "learning_rate": 1.8207392868825644e-05,
      "loss": 0.6216,
      "step": 55600
    },
    {
      "epoch": 27.48,
      "learning_rate": 1.689892051030422e-05,
      "loss": 0.627,
      "step": 56000
    },
    {
      "epoch": 27.67,
      "learning_rate": 1.5590448151782795e-05,
      "loss": 0.613,
      "step": 56400
    },
    {
      "epoch": 27.87,
      "learning_rate": 1.428524697415767e-05,
      "loss": 0.6098,
      "step": 56800
    },
    {
      "epoch": 28.0,
      "eval_f1": 0.004164219476454643,
      "eval_loss": 7.554748058319092,
      "eval_runtime": 103.9745,
      "eval_samples_per_second": 221.324,
      "eval_steps_per_second": 3.462,
      "step": 57064
    },
    {
      "epoch": 28.07,
      "learning_rate": 1.2976774615636244e-05,
      "loss": 0.6228,
      "step": 57200
    },
    {
      "epoch": 28.26,
      "learning_rate": 1.1668302257114818e-05,
      "loss": 0.5921,
      "step": 57600
    },
    {
      "epoch": 28.46,
      "learning_rate": 1.0359829898593392e-05,
      "loss": 0.6023,
      "step": 58000
    },
    {
      "epoch": 28.66,
      "learning_rate": 9.051357540071966e-06,
      "loss": 0.5876,
      "step": 58400
    },
    {
      "epoch": 28.85,
      "learning_rate": 7.746156362446843e-06,
      "loss": 0.5871,
      "step": 58800
    },
    {
      "epoch": 29.0,
      "eval_f1": 0.00413622990961682,
      "eval_loss": 7.553284645080566,
      "eval_runtime": 104.0213,
      "eval_samples_per_second": 221.224,
      "eval_steps_per_second": 3.461,
      "step": 59102
    },
    {
      "epoch": 29.05,
      "learning_rate": 6.437684003925418e-06,
      "loss": 0.5787,
      "step": 59200
    },
    {
      "epoch": 29.24,
      "learning_rate": 5.129211645403991e-06,
      "loss": 0.5671,
      "step": 59600
    },
    {
      "epoch": 29.44,
      "learning_rate": 3.820739286882564e-06,
      "loss": 0.5596,
      "step": 60000
    },
    {
      "epoch": 29.64,
      "learning_rate": 2.5122669283611384e-06,
      "loss": 0.5595,
      "step": 60400
    },
    {
      "epoch": 29.83,
      "learning_rate": 1.2070657507360158e-06,
      "loss": 0.5696,
      "step": 60800
    },
    {
      "epoch": 30.0,
      "eval_f1": 0.004123942255601062,
      "eval_loss": 7.559465408325195,
      "eval_runtime": 98.4712,
      "eval_samples_per_second": 233.693,
      "eval_steps_per_second": 3.656,
      "step": 61140
    },
    {
      "epoch": 30.0,
      "step": 61140,
      "total_flos": 3.687496208074506e+20,
      "train_loss": 2.558345020757144,
      "train_runtime": 41575.9952,
      "train_samples_per_second": 94.093,
      "train_steps_per_second": 1.471
    }
  ],
  "max_steps": 61140,
  "num_train_epochs": 30,
  "total_flos": 3.687496208074506e+20,
  "trial_name": null,
  "trial_params": null
}