{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02247191011235955,
      "grad_norm": 3.921875,
      "learning_rate": 4.49438202247191e-06,
      "loss": 2.2098,
      "step": 20
    },
    {
      "epoch": 0.0449438202247191,
      "grad_norm": 2.03125,
      "learning_rate": 8.98876404494382e-06,
      "loss": 2.1052,
      "step": 40
    },
    {
      "epoch": 0.06741573033707865,
      "grad_norm": 1.21875,
      "learning_rate": 1.348314606741573e-05,
      "loss": 2.1605,
      "step": 60
    },
    {
      "epoch": 0.0898876404494382,
      "grad_norm": 1.0234375,
      "learning_rate": 1.797752808988764e-05,
      "loss": 1.8331,
      "step": 80
    },
    {
      "epoch": 0.11235955056179775,
      "grad_norm": 0.61328125,
      "learning_rate": 2.2471910112359552e-05,
      "loss": 1.8186,
      "step": 100
    },
    {
      "epoch": 0.1348314606741573,
      "grad_norm": 0.703125,
      "learning_rate": 2.696629213483146e-05,
      "loss": 1.5533,
      "step": 120
    },
    {
      "epoch": 0.15730337078651685,
      "grad_norm": 0.4921875,
      "learning_rate": 3.1460674157303374e-05,
      "loss": 1.5419,
      "step": 140
    },
    {
      "epoch": 0.1797752808988764,
      "grad_norm": 0.470703125,
      "learning_rate": 3.595505617977528e-05,
      "loss": 1.3374,
      "step": 160
    },
    {
      "epoch": 0.20224719101123595,
      "grad_norm": 0.52734375,
      "learning_rate": 4.044943820224719e-05,
      "loss": 1.3418,
      "step": 180
    },
    {
      "epoch": 0.2247191011235955,
      "grad_norm": 0.59375,
      "learning_rate": 4.4943820224719104e-05,
      "loss": 1.3921,
      "step": 200
    },
    {
      "epoch": 0.24719101123595505,
      "grad_norm": 0.67578125,
      "learning_rate": 4.943820224719101e-05,
      "loss": 1.1401,
      "step": 220
    },
    {
      "epoch": 0.2696629213483146,
      "grad_norm": 0.72265625,
      "learning_rate": 5.393258426966292e-05,
      "loss": 1.2139,
      "step": 240
    },
    {
      "epoch": 0.29213483146067415,
      "grad_norm": 0.49609375,
      "learning_rate": 5.8426966292134835e-05,
      "loss": 1.171,
      "step": 260
    },
    {
      "epoch": 0.3146067415730337,
      "grad_norm": 0.94921875,
      "learning_rate": 6.292134831460675e-05,
      "loss": 1.1424,
      "step": 280
    },
    {
      "epoch": 0.33707865168539325,
      "grad_norm": 0.8046875,
      "learning_rate": 6.741573033707866e-05,
      "loss": 1.2171,
      "step": 300
    },
    {
      "epoch": 0.3595505617977528,
      "grad_norm": 1.09375,
      "learning_rate": 7.191011235955056e-05,
      "loss": 1.1575,
      "step": 320
    },
    {
      "epoch": 0.38202247191011235,
      "grad_norm": 1.4453125,
      "learning_rate": 7.640449438202247e-05,
      "loss": 1.2041,
      "step": 340
    },
    {
      "epoch": 0.4044943820224719,
      "grad_norm": 1.0,
      "learning_rate": 8.089887640449438e-05,
      "loss": 1.106,
      "step": 360
    },
    {
      "epoch": 0.42696629213483145,
      "grad_norm": 1.0390625,
      "learning_rate": 8.53932584269663e-05,
      "loss": 1.0728,
      "step": 380
    },
    {
      "epoch": 0.449438202247191,
      "grad_norm": 1.0859375,
      "learning_rate": 8.988764044943821e-05,
      "loss": 0.9622,
      "step": 400
    },
    {
      "epoch": 0.47191011235955055,
      "grad_norm": 0.99609375,
      "learning_rate": 9.438202247191012e-05,
      "loss": 1.0835,
      "step": 420
    },
    {
      "epoch": 0.4943820224719101,
      "grad_norm": 0.8828125,
      "learning_rate": 9.887640449438202e-05,
      "loss": 1.0557,
      "step": 440
    },
    {
      "epoch": 0.5168539325842697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00010337078651685395,
      "loss": 1.1037,
      "step": 460
    },
    {
      "epoch": 0.5393258426966292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010786516853932584,
      "loss": 1.058,
      "step": 480
    },
    {
      "epoch": 0.5617977528089888,
      "grad_norm": 1.03125,
      "learning_rate": 0.00011235955056179777,
      "loss": 1.1904,
      "step": 500
    },
    {
      "epoch": 0.5617977528089888,
      "eval_loss": 1.0617437362670898,
      "eval_runtime": 206.3616,
      "eval_samples_per_second": 1.856,
      "eval_steps_per_second": 1.856,
      "step": 500
    },
    {
      "epoch": 0.5842696629213483,
      "grad_norm": 0.60546875,
      "learning_rate": 0.00011685393258426967,
      "loss": 1.1192,
      "step": 520
    },
    {
      "epoch": 0.6067415730337079,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00012134831460674158,
      "loss": 1.0356,
      "step": 540
    },
    {
      "epoch": 0.6292134831460674,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001258426966292135,
      "loss": 1.0108,
      "step": 560
    },
    {
      "epoch": 0.651685393258427,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001303370786516854,
      "loss": 0.8835,
      "step": 580
    },
    {
      "epoch": 0.6741573033707865,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00013483146067415732,
      "loss": 0.9925,
      "step": 600
    },
    {
      "epoch": 0.6966292134831461,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013932584269662923,
      "loss": 0.9548,
      "step": 620
    },
    {
      "epoch": 0.7191011235955056,
      "grad_norm": 0.9375,
      "learning_rate": 0.00014382022471910112,
      "loss": 1.113,
      "step": 640
    },
    {
      "epoch": 0.7415730337078652,
      "grad_norm": 0.8125,
      "learning_rate": 0.00014831460674157306,
      "loss": 0.9504,
      "step": 660
    },
    {
      "epoch": 0.7640449438202247,
      "grad_norm": 0.6953125,
      "learning_rate": 0.00015280898876404494,
      "loss": 1.1266,
      "step": 680
    },
    {
      "epoch": 0.7865168539325843,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015730337078651685,
      "loss": 1.0707,
      "step": 700
    },
    {
      "epoch": 0.8089887640449438,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00016179775280898877,
      "loss": 0.9222,
      "step": 720
    },
    {
      "epoch": 0.8314606741573034,
      "grad_norm": 0.79296875,
      "learning_rate": 0.00016629213483146068,
      "loss": 1.1053,
      "step": 740
    },
    {
      "epoch": 0.8539325842696629,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0001707865168539326,
      "loss": 1.25,
      "step": 760
    },
    {
      "epoch": 0.8764044943820225,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001752808988764045,
      "loss": 0.933,
      "step": 780
    },
    {
      "epoch": 0.898876404494382,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00017977528089887642,
      "loss": 0.885,
      "step": 800
    },
    {
      "epoch": 0.9213483146067416,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018426966292134833,
      "loss": 0.9786,
      "step": 820
    },
    {
      "epoch": 0.9438202247191011,
      "grad_norm": 0.96875,
      "learning_rate": 0.00018876404494382024,
      "loss": 0.9101,
      "step": 840
    },
    {
      "epoch": 0.9662921348314607,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00019325842696629215,
      "loss": 0.9005,
      "step": 860
    },
    {
      "epoch": 0.9887640449438202,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00019775280898876404,
      "loss": 0.9485,
      "step": 880
    },
    {
      "epoch": 1.0112359550561798,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00019999610626011892,
      "loss": 1.0071,
      "step": 900
    },
    {
      "epoch": 1.0337078651685394,
      "grad_norm": 0.84375,
      "learning_rate": 0.00019996495816039186,
      "loss": 0.9138,
      "step": 920
    },
    {
      "epoch": 1.0561797752808988,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00019990267166335664,
      "loss": 0.7752,
      "step": 940
    },
    {
      "epoch": 1.0786516853932584,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00019980926617082901,
      "loss": 0.9331,
      "step": 960
    },
    {
      "epoch": 1.101123595505618,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019968477077797781,
      "loss": 1.0037,
      "step": 980
    },
    {
      "epoch": 1.1235955056179776,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00019952922426426207,
      "loss": 0.765,
      "step": 1000
    },
    {
      "epoch": 1.1235955056179776,
      "eval_loss": 0.944207489490509,
      "eval_runtime": 206.3524,
      "eval_samples_per_second": 1.856,
      "eval_steps_per_second": 1.856,
      "step": 1000
    },
    {
      "epoch": 1.146067415730337,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019934267508135164,
      "loss": 0.861,
      "step": 1020
    },
    {
      "epoch": 1.1685393258426966,
      "grad_norm": 1.25,
      "learning_rate": 0.00019912518133803465,
      "loss": 0.8251,
      "step": 1040
    },
    {
      "epoch": 1.1910112359550562,
      "grad_norm": 0.609375,
      "learning_rate": 0.00019887681078211707,
      "loss": 0.9779,
      "step": 1060
    },
    {
      "epoch": 1.2134831460674158,
      "grad_norm": 0.765625,
      "learning_rate": 0.00019859764077931978,
      "loss": 0.8112,
      "step": 1080
    },
    {
      "epoch": 1.2359550561797752,
      "grad_norm": 0.890625,
      "learning_rate": 0.00019828775828917964,
      "loss": 0.9084,
      "step": 1100
    },
    {
      "epoch": 1.2584269662921348,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019794725983796218,
      "loss": 0.8429,
      "step": 1120
    },
    {
      "epoch": 1.2808988764044944,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00019757625148859441,
      "loss": 0.8029,
      "step": 1140
    },
    {
      "epoch": 1.303370786516854,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00019717484880762685,
      "loss": 0.9478,
      "step": 1160
    },
    {
      "epoch": 1.3258426966292136,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00019674317682923532,
      "loss": 0.6985,
      "step": 1180
    },
    {
      "epoch": 1.348314606741573,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00019628137001627383,
      "loss": 0.8653,
      "step": 1200
    },
    {
      "epoch": 1.3707865168539326,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00019578957221839014,
      "loss": 0.891,
      "step": 1220
    },
    {
      "epoch": 1.3932584269662922,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00019526793662721768,
      "loss": 0.861,
      "step": 1240
    },
    {
      "epoch": 1.4157303370786516,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00019471662572865736,
      "loss": 0.7591,
      "step": 1260
    },
    {
      "epoch": 1.4382022471910112,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00019413581125226438,
      "loss": 0.7109,
      "step": 1280
    },
    {
      "epoch": 1.4606741573033708,
      "grad_norm": 0.6875,
      "learning_rate": 0.00019352567411775565,
      "loss": 0.8947,
      "step": 1300
    },
    {
      "epoch": 1.4831460674157304,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00019288640437865445,
      "loss": 0.8514,
      "step": 1320
    },
    {
      "epoch": 1.50561797752809,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0001922182011630902,
      "loss": 0.7379,
      "step": 1340
    },
    {
      "epoch": 1.5280898876404494,
      "grad_norm": 0.63671875,
      "learning_rate": 0.00019152127261177126,
      "loss": 0.6778,
      "step": 1360
    },
    {
      "epoch": 1.550561797752809,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00019079583581315076,
      "loss": 0.6591,
      "step": 1380
    },
    {
      "epoch": 1.5730337078651684,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0001900421167358048,
      "loss": 0.8635,
      "step": 1400
    },
    {
      "epoch": 1.595505617977528,
      "grad_norm": 0.8125,
      "learning_rate": 0.00018926035015804488,
      "loss": 0.924,
      "step": 1420
    },
    {
      "epoch": 1.6179775280898876,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018845077959478613,
      "loss": 0.8554,
      "step": 1440
    },
    {
      "epoch": 1.6404494382022472,
      "grad_norm": 0.6171875,
      "learning_rate": 0.00018761365722169403,
      "loss": 0.9471,
      "step": 1460
    },
    {
      "epoch": 1.6629213483146068,
      "grad_norm": 0.64453125,
      "learning_rate": 0.00018674924379663338,
      "loss": 0.9187,
      "step": 1480
    },
    {
      "epoch": 1.6853932584269664,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00018585780857844418,
      "loss": 0.782,
      "step": 1500
    },
    {
      "epoch": 1.6853932584269664,
      "eval_loss": 0.8689672350883484,
      "eval_runtime": 206.6921,
      "eval_samples_per_second": 1.853,
      "eval_steps_per_second": 1.853,
      "step": 1500
    },
    {
      "epoch": 1.7078651685393258,
      "grad_norm": 0.890625,
      "learning_rate": 0.00018493962924306912,
      "loss": 0.8983,
      "step": 1520
    },
    {
      "epoch": 1.7303370786516854,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0001839949917970596,
      "loss": 0.5218,
      "step": 1540
    },
    {
      "epoch": 1.7528089887640448,
      "grad_norm": 0.921875,
      "learning_rate": 0.00018302419048848667,
      "loss": 0.6711,
      "step": 1560
    },
    {
      "epoch": 1.7752808988764044,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0001820275277152846,
      "loss": 0.7932,
      "step": 1580
    },
    {
      "epoch": 1.797752808988764,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00018100531393105623,
      "loss": 0.7181,
      "step": 1600
    },
    {
      "epoch": 1.8202247191011236,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00017995786754836863,
      "loss": 0.8525,
      "step": 1620
    },
    {
      "epoch": 1.8426966292134832,
      "grad_norm": 0.6484375,
      "learning_rate": 0.00017888551483956987,
      "loss": 0.6968,
      "step": 1640
    },
    {
      "epoch": 1.8651685393258428,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00017778858983515743,
      "loss": 0.902,
      "step": 1660
    },
    {
      "epoch": 1.8876404494382022,
      "grad_norm": 0.875,
      "learning_rate": 0.00017666743421972987,
      "loss": 0.954,
      "step": 1680
    },
    {
      "epoch": 1.9101123595505618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001755223972255546,
      "loss": 0.791,
      "step": 1700
    },
    {
      "epoch": 1.9325842696629212,
      "grad_norm": 0.765625,
      "learning_rate": 0.00017435383552378428,
      "loss": 0.77,
      "step": 1720
    },
    {
      "epoch": 1.9550561797752808,
      "grad_norm": 1.25,
      "learning_rate": 0.0001731621131133564,
      "loss": 0.6294,
      "step": 1740
    },
    {
      "epoch": 1.9775280898876404,
      "grad_norm": 0.70703125,
      "learning_rate": 0.00017194760120760986,
      "loss": 0.7982,
      "step": 1760
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.7643,
      "step": 1780
    },
    {
      "epoch": 2.0224719101123596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001694517291395307,
      "loss": 0.5279,
      "step": 1800
    },
    {
      "epoch": 2.044943820224719,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016817114642419067,
      "loss": 0.6667,
      "step": 1820
    },
    {
      "epoch": 2.067415730337079,
      "grad_norm": 0.9375,
      "learning_rate": 0.00016686932886534781,
      "loss": 0.6427,
      "step": 1840
    },
    {
      "epoch": 2.0898876404494384,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00016554668197022295,
      "loss": 0.633,
      "step": 1860
    },
    {
      "epoch": 2.1123595505617976,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00016420361773423204,
      "loss": 0.5623,
      "step": 1880
    },
    {
      "epoch": 2.134831460674157,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016284055451265246,
      "loss": 0.6311,
      "step": 1900
    },
    {
      "epoch": 2.157303370786517,
      "grad_norm": 1.25,
      "learning_rate": 0.00016145791689030795,
      "loss": 0.7469,
      "step": 1920
    },
    {
      "epoch": 2.1797752808988764,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0001600561355493137,
      "loss": 0.7196,
      "step": 1940
    },
    {
      "epoch": 2.202247191011236,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0001586356471349215,
      "loss": 0.6328,
      "step": 1960
    },
    {
      "epoch": 2.2247191011235956,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00015719689411950808,
      "loss": 0.6349,
      "step": 1980
    },
    {
      "epoch": 2.247191011235955,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015574032466474775,
      "loss": 0.5591,
      "step": 2000
    },
    {
      "epoch": 2.247191011235955,
      "eval_loss": 0.8647096753120422,
      "eval_runtime": 206.6878,
      "eval_samples_per_second": 1.853,
      "eval_steps_per_second": 1.853,
      "step": 2000
    },
    {
      "epoch": 2.2696629213483144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015426639248201313,
      "loss": 0.5206,
      "step": 2020
    },
    {
      "epoch": 2.292134831460674,
      "grad_norm": 0.8125,
      "learning_rate": 0.0001527755566910474,
      "loss": 0.7186,
      "step": 2040
    },
    {
      "epoch": 2.3146067415730336,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00015126828167695146,
      "loss": 0.6533,
      "step": 2060
    },
    {
      "epoch": 2.337078651685393,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001497450369455312,
      "loss": 0.6324,
      "step": 2080
    },
    {
      "epoch": 2.359550561797753,
      "grad_norm": 0.875,
      "learning_rate": 0.00014820629697704965,
      "loss": 0.5276,
      "step": 2100
    },
    {
      "epoch": 2.3820224719101124,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00014665254107842964,
      "loss": 0.612,
      "step": 2120
    },
    {
      "epoch": 2.404494382022472,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00014508425323395317,
      "loss": 0.614,
      "step": 2140
    },
    {
      "epoch": 2.4269662921348316,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0001435019219545034,
      "loss": 0.4988,
      "step": 2160
    },
    {
      "epoch": 2.449438202247191,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00014190604012539684,
      "loss": 0.6777,
      "step": 2180
    },
    {
      "epoch": 2.4719101123595504,
      "grad_norm": 0.96875,
      "learning_rate": 0.00014029710485285324,
      "loss": 0.662,
      "step": 2200
    },
    {
      "epoch": 2.49438202247191,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00013867561730915016,
      "loss": 0.6087,
      "step": 2220
    },
    {
      "epoch": 2.5168539325842696,
      "grad_norm": 0.875,
      "learning_rate": 0.0001370420825765114,
      "loss": 0.56,
      "step": 2240
    },
    {
      "epoch": 2.539325842696629,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013539700948977717,
      "loss": 0.572,
      "step": 2260
    },
    {
      "epoch": 2.561797752808989,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013374091047790585,
      "loss": 0.7334,
      "step": 2280
    },
    {
      "epoch": 2.5842696629213484,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013207430140435556,
      "loss": 0.5377,
      "step": 2300
    },
    {
      "epoch": 2.606741573033708,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00013039770140639654,
      "loss": 0.6306,
      "step": 2320
    },
    {
      "epoch": 2.629213483146067,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012871163273340307,
      "loss": 0.582,
      "step": 2340
    },
    {
      "epoch": 2.6516853932584272,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00012701662058417688,
      "loss": 0.6326,
      "step": 2360
    },
    {
      "epoch": 2.6741573033707864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00012531319294335086,
      "loss": 0.6907,
      "step": 2380
    },
    {
      "epoch": 2.696629213483146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012360188041692582,
      "loss": 0.656,
      "step": 2400
    },
    {
      "epoch": 2.7191011235955056,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012188321606699016,
      "loss": 0.5817,
      "step": 2420
    },
    {
      "epoch": 2.741573033707865,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00012015773524567479,
      "loss": 0.5046,
      "step": 2440
    },
    {
      "epoch": 2.764044943820225,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011842597542839462,
      "loss": 0.6293,
      "step": 2460
    },
    {
      "epoch": 2.7865168539325844,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011668847604642861,
      "loss": 0.6067,
      "step": 2480
    },
    {
      "epoch": 2.808988764044944,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011494577831889067,
      "loss": 0.5669,
      "step": 2500
    },
    {
      "epoch": 2.808988764044944,
      "eval_loss": 0.8295581340789795,
      "eval_runtime": 206.704,
      "eval_samples_per_second": 1.853,
      "eval_steps_per_second": 1.853,
      "step": 2500
    },
    {
      "epoch": 2.831460674157303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011319842508414365,
      "loss": 0.5429,
      "step": 2520
    },
    {
      "epoch": 2.853932584269663,
      "grad_norm": 1.03125,
      "learning_rate": 0.00011144696063070883,
      "loss": 0.5481,
      "step": 2540
    },
    {
      "epoch": 2.8764044943820224,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00010969193052772396,
      "loss": 0.5308,
      "step": 2560
    },
    {
      "epoch": 2.898876404494382,
      "grad_norm": 0.703125,
      "learning_rate": 0.00010793388145500198,
      "loss": 0.4527,
      "step": 2580
    },
    {
      "epoch": 2.9213483146067416,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010617336103274424,
      "loss": 0.5333,
      "step": 2600
    },
    {
      "epoch": 2.943820224719101,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00010441091765096047,
      "loss": 0.5886,
      "step": 2620
    },
    {
      "epoch": 2.966292134831461,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001026471002986491,
      "loss": 0.626,
      "step": 2640
    },
    {
      "epoch": 2.98876404494382,
      "grad_norm": 0.953125,
      "learning_rate": 0.00010088245839279082,
      "loss": 0.6703,
      "step": 2660
    },
    {
      "epoch": 3.0112359550561796,
      "grad_norm": 0.64453125,
      "learning_rate": 9.911754160720923e-05,
      "loss": 0.4819,
      "step": 2680
    },
    {
      "epoch": 3.033707865168539,
      "grad_norm": 0.71875,
      "learning_rate": 9.735289970135095e-05,
      "loss": 0.4379,
      "step": 2700
    },
    {
      "epoch": 3.056179775280899,
      "grad_norm": 1.7734375,
      "learning_rate": 9.558908234903954e-05,
      "loss": 0.3811,
      "step": 2720
    },
    {
      "epoch": 3.0786516853932584,
      "grad_norm": 0.6953125,
      "learning_rate": 9.382663896725578e-05,
      "loss": 0.3855,
      "step": 2740
    },
    {
      "epoch": 3.101123595505618,
      "grad_norm": 1.4140625,
      "learning_rate": 9.206611854499805e-05,
      "loss": 0.4749,
      "step": 2760
    },
    {
      "epoch": 3.1235955056179776,
      "grad_norm": 0.609375,
      "learning_rate": 9.030806947227607e-05,
      "loss": 0.501,
      "step": 2780
    },
    {
      "epoch": 3.146067415730337,
      "grad_norm": 1.1015625,
      "learning_rate": 8.855303936929117e-05,
      "loss": 0.4239,
      "step": 2800
    },
    {
      "epoch": 3.168539325842697,
      "grad_norm": 0.84375,
      "learning_rate": 8.680157491585636e-05,
      "loss": 0.5388,
      "step": 2820
    },
    {
      "epoch": 3.191011235955056,
      "grad_norm": 1.2578125,
      "learning_rate": 8.505422168110934e-05,
      "loss": 0.3715,
      "step": 2840
    },
    {
      "epoch": 3.2134831460674156,
      "grad_norm": 0.94921875,
      "learning_rate": 8.331152395357141e-05,
      "loss": 0.4274,
      "step": 2860
    },
    {
      "epoch": 3.235955056179775,
      "grad_norm": 1.046875,
      "learning_rate": 8.157402457160539e-05,
      "loss": 0.4368,
      "step": 2880
    },
    {
      "epoch": 3.258426966292135,
      "grad_norm": 1.2421875,
      "learning_rate": 7.984226475432522e-05,
      "loss": 0.4026,
      "step": 2900
    },
    {
      "epoch": 3.2808988764044944,
      "grad_norm": 0.8359375,
      "learning_rate": 7.811678393300987e-05,
      "loss": 0.3971,
      "step": 2920
    },
    {
      "epoch": 3.303370786516854,
      "grad_norm": 0.94921875,
      "learning_rate": 7.63981195830742e-05,
      "loss": 0.395,
      "step": 2940
    },
    {
      "epoch": 3.3258426966292136,
      "grad_norm": 1.328125,
      "learning_rate": 7.468680705664914e-05,
      "loss": 0.4165,
      "step": 2960
    },
    {
      "epoch": 3.348314606741573,
      "grad_norm": 0.88671875,
      "learning_rate": 7.298337941582314e-05,
      "loss": 0.4071,
      "step": 2980
    },
    {
      "epoch": 3.370786516853933,
      "grad_norm": 1.4296875,
      "learning_rate": 7.128836726659696e-05,
      "loss": 0.4205,
      "step": 3000
    },
    {
      "epoch": 3.370786516853933,
      "eval_loss": 0.8820343613624573,
      "eval_runtime": 206.8369,
      "eval_samples_per_second": 1.852,
      "eval_steps_per_second": 1.852,
      "step": 3000
    },
    {
      "epoch": 3.393258426966292,
      "grad_norm": 1.46875,
      "learning_rate": 6.960229859360353e-05,
      "loss": 0.3759,
      "step": 3020
    },
    {
      "epoch": 3.4157303370786516,
      "grad_norm": 1.2265625,
      "learning_rate": 6.792569859564445e-05,
      "loss": 0.4457,
      "step": 3040
    },
    {
      "epoch": 3.438202247191011,
      "grad_norm": 1.5390625,
      "learning_rate": 6.625908952209418e-05,
      "loss": 0.4088,
      "step": 3060
    },
    {
      "epoch": 3.460674157303371,
      "grad_norm": 0.92578125,
      "learning_rate": 6.460299051022285e-05,
      "loss": 0.4221,
      "step": 3080
    },
    {
      "epoch": 3.4831460674157304,
      "grad_norm": 1.34375,
      "learning_rate": 6.295791742348865e-05,
      "loss": 0.4304,
      "step": 3100
    },
    {
      "epoch": 3.50561797752809,
      "grad_norm": 1.0,
      "learning_rate": 6.132438269084985e-05,
      "loss": 0.3612,
      "step": 3120
    },
    {
      "epoch": 3.5280898876404496,
      "grad_norm": 1.4609375,
      "learning_rate": 5.970289514714677e-05,
      "loss": 0.4692,
      "step": 3140
    },
    {
      "epoch": 3.550561797752809,
      "grad_norm": 1.578125,
      "learning_rate": 5.8093959874603176e-05,
      "loss": 0.4579,
      "step": 3160
    },
    {
      "epoch": 3.5730337078651684,
      "grad_norm": 0.66796875,
      "learning_rate": 5.649807804549663e-05,
      "loss": 0.3754,
      "step": 3180
    },
    {
      "epoch": 3.595505617977528,
      "grad_norm": 1.1328125,
      "learning_rate": 5.491574676604682e-05,
      "loss": 0.3685,
      "step": 3200
    },
    {
      "epoch": 3.6179775280898876,
      "grad_norm": 1.09375,
      "learning_rate": 5.334745892157035e-05,
      "loss": 0.3809,
      "step": 3220
    },
    {
      "epoch": 3.640449438202247,
      "grad_norm": 1.4453125,
      "learning_rate": 5.179370302295037e-05,
      "loss": 0.4809,
      "step": 3240
    },
    {
      "epoch": 3.662921348314607,
      "grad_norm": 1.015625,
      "learning_rate": 5.02549630544688e-05,
      "loss": 0.3798,
      "step": 3260
    },
    {
      "epoch": 3.6853932584269664,
      "grad_norm": 1.1796875,
      "learning_rate": 4.8731718323048516e-05,
      "loss": 0.4153,
      "step": 3280
    },
    {
      "epoch": 3.7078651685393256,
      "grad_norm": 0.98046875,
      "learning_rate": 4.722444330895256e-05,
      "loss": 0.4612,
      "step": 3300
    },
    {
      "epoch": 3.7303370786516856,
      "grad_norm": 1.4921875,
      "learning_rate": 4.573360751798689e-05,
      "loss": 0.469,
      "step": 3320
    },
    {
      "epoch": 3.752808988764045,
      "grad_norm": 1.8203125,
      "learning_rate": 4.425967533525229e-05,
      "loss": 0.4523,
      "step": 3340
    },
    {
      "epoch": 3.7752808988764044,
      "grad_norm": 1.359375,
      "learning_rate": 4.2803105880491925e-05,
      "loss": 0.4214,
      "step": 3360
    },
    {
      "epoch": 3.797752808988764,
      "grad_norm": 1.734375,
      "learning_rate": 4.136435286507849e-05,
      "loss": 0.4981,
      "step": 3380
    },
    {
      "epoch": 3.8202247191011236,
      "grad_norm": 1.3515625,
      "learning_rate": 3.994386445068632e-05,
      "loss": 0.4029,
      "step": 3400
    },
    {
      "epoch": 3.842696629213483,
      "grad_norm": 1.4609375,
      "learning_rate": 3.854208310969204e-05,
      "loss": 0.3747,
      "step": 3420
    },
    {
      "epoch": 3.865168539325843,
      "grad_norm": 1.1015625,
      "learning_rate": 3.715944548734755e-05,
      "loss": 0.4113,
      "step": 3440
    },
    {
      "epoch": 3.8876404494382024,
      "grad_norm": 0.52734375,
      "learning_rate": 3.5796382265767937e-05,
      "loss": 0.3896,
      "step": 3460
    },
    {
      "epoch": 3.9101123595505616,
      "grad_norm": 1.0625,
      "learning_rate": 3.445331802977709e-05,
      "loss": 0.4709,
      "step": 3480
    },
    {
      "epoch": 3.932584269662921,
      "grad_norm": 1.4296875,
      "learning_rate": 3.313067113465222e-05,
      "loss": 0.3812,
      "step": 3500
    },
    {
      "epoch": 3.932584269662921,
      "eval_loss": 0.8859002590179443,
      "eval_runtime": 206.8365,
      "eval_samples_per_second": 1.852,
      "eval_steps_per_second": 1.852,
      "step": 3500
    },
    {
      "epoch": 3.955056179775281,
      "grad_norm": 1.09375,
      "learning_rate": 3.182885357580934e-05,
      "loss": 0.3906,
      "step": 3520
    },
    {
      "epoch": 3.9775280898876404,
      "grad_norm": 1.0390625,
      "learning_rate": 3.054827086046931e-05,
      "loss": 0.3987,
      "step": 3540
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.671875,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.3499,
      "step": 3560
    },
    {
      "epoch": 4.022471910112359,
      "grad_norm": 0.79296875,
      "learning_rate": 2.8052398792390154e-05,
      "loss": 0.3292,
      "step": 3580
    },
    {
      "epoch": 4.044943820224719,
      "grad_norm": 0.8125,
      "learning_rate": 2.6837886886643614e-05,
      "loss": 0.3343,
      "step": 3600
    },
    {
      "epoch": 4.067415730337078,
      "grad_norm": 0.66015625,
      "learning_rate": 2.5646164476215716e-05,
      "loss": 0.3236,
      "step": 3620
    },
    {
      "epoch": 4.089887640449438,
      "grad_norm": 1.25,
      "learning_rate": 2.447760277444543e-05,
      "loss": 0.2892,
      "step": 3640
    },
    {
      "epoch": 4.112359550561798,
      "grad_norm": 1.3515625,
      "learning_rate": 2.3332565780270165e-05,
      "loss": 0.2801,
      "step": 3660
    },
    {
      "epoch": 4.134831460674158,
      "grad_norm": 1.3671875,
      "learning_rate": 2.2211410164842606e-05,
      "loss": 0.3082,
      "step": 3680
    },
    {
      "epoch": 4.157303370786517,
      "grad_norm": 1.46875,
      "learning_rate": 2.1114485160430132e-05,
      "loss": 0.3128,
      "step": 3700
    },
    {
      "epoch": 4.179775280898877,
      "grad_norm": 0.8515625,
      "learning_rate": 2.0042132451631378e-05,
      "loss": 0.3846,
      "step": 3720
    },
    {
      "epoch": 4.202247191011236,
      "grad_norm": 1.1328125,
      "learning_rate": 1.899468606894379e-05,
      "loss": 0.2718,
      "step": 3740
    },
    {
      "epoch": 4.224719101123595,
      "grad_norm": 0.80859375,
      "learning_rate": 1.7972472284715415e-05,
      "loss": 0.302,
      "step": 3760
    },
    {
      "epoch": 4.247191011235955,
      "grad_norm": 1.6953125,
      "learning_rate": 1.6975809511513353e-05,
      "loss": 0.3785,
      "step": 3780
    },
    {
      "epoch": 4.269662921348314,
      "grad_norm": 1.0078125,
      "learning_rate": 1.600500820294041e-05,
      "loss": 0.3845,
      "step": 3800
    },
    {
      "epoch": 4.292134831460674,
      "grad_norm": 1.6796875,
      "learning_rate": 1.5060370756930919e-05,
      "loss": 0.327,
      "step": 3820
    },
    {
      "epoch": 4.314606741573034,
      "grad_norm": 1.359375,
      "learning_rate": 1.414219142155585e-05,
      "loss": 0.3589,
      "step": 3840
    },
    {
      "epoch": 4.337078651685394,
      "grad_norm": 0.75,
      "learning_rate": 1.3250756203366632e-05,
      "loss": 0.4057,
      "step": 3860
    },
    {
      "epoch": 4.359550561797753,
      "grad_norm": 0.96484375,
      "learning_rate": 1.2386342778305993e-05,
      "loss": 0.3862,
      "step": 3880
    },
    {
      "epoch": 4.382022471910112,
      "grad_norm": 1.4921875,
      "learning_rate": 1.1549220405213878e-05,
      "loss": 0.3319,
      "step": 3900
    },
    {
      "epoch": 4.404494382022472,
      "grad_norm": 1.2890625,
      "learning_rate": 1.0739649841955136e-05,
      "loss": 0.2832,
      "step": 3920
    },
    {
      "epoch": 4.426966292134831,
      "grad_norm": 1.234375,
      "learning_rate": 9.957883264195223e-06,
      "loss": 0.2732,
      "step": 3940
    },
    {
      "epoch": 4.449438202247191,
      "grad_norm": 0.828125,
      "learning_rate": 9.20416418684924e-06,
      "loss": 0.2587,
      "step": 3960
    },
    {
      "epoch": 4.47191011235955,
      "grad_norm": 1.5234375,
      "learning_rate": 8.478727388228735e-06,
      "loss": 0.3469,
      "step": 3980
    },
    {
      "epoch": 4.49438202247191,
      "grad_norm": 0.90234375,
      "learning_rate": 7.781798836909826e-06,
      "loss": 0.3323,
      "step": 4000
    },
    {
      "epoch": 4.49438202247191,
      "eval_loss": 0.9360187649726868,
      "eval_runtime": 206.9644,
      "eval_samples_per_second": 1.851,
      "eval_steps_per_second": 1.851,
      "step": 4000
    },
    {
      "epoch": 4.51685393258427,
      "grad_norm": 2.046875,
      "learning_rate": 7.11359562134557e-06,
      "loss": 0.3441,
      "step": 4020
    },
    {
      "epoch": 4.539325842696629,
      "grad_norm": 1.296875,
      "learning_rate": 6.4743258822443695e-06,
      "loss": 0.3196,
      "step": 4040
    },
    {
      "epoch": 4.561797752808989,
      "grad_norm": 1.515625,
      "learning_rate": 5.8641887477356215e-06,
      "loss": 0.3226,
      "step": 4060
    },
    {
      "epoch": 4.584269662921348,
      "grad_norm": 1.6875,
      "learning_rate": 5.283374271342645e-06,
      "loss": 0.2859,
      "step": 4080
    },
    {
      "epoch": 4.606741573033708,
      "grad_norm": 0.87109375,
      "learning_rate": 4.732063372782336e-06,
      "loss": 0.3164,
      "step": 4100
    },
    {
      "epoch": 4.629213483146067,
      "grad_norm": 1.2734375,
      "learning_rate": 4.210427781609861e-06,
      "loss": 0.3275,
      "step": 4120
    },
    {
      "epoch": 4.651685393258427,
      "grad_norm": 1.25,
      "learning_rate": 3.718629983726185e-06,
      "loss": 0.367,
      "step": 4140
    },
    {
      "epoch": 4.674157303370786,
      "grad_norm": 1.6796875,
      "learning_rate": 3.256823170764689e-06,
      "loss": 0.3445,
      "step": 4160
    },
    {
      "epoch": 4.696629213483146,
      "grad_norm": 0.9453125,
      "learning_rate": 2.8251511923731655e-06,
      "loss": 0.4628,
      "step": 4180
    },
    {
      "epoch": 4.719101123595506,
      "grad_norm": 0.8515625,
      "learning_rate": 2.423748511405577e-06,
      "loss": 0.3252,
      "step": 4200
    },
    {
      "epoch": 4.741573033707866,
      "grad_norm": 0.90234375,
      "learning_rate": 2.052740162037814e-06,
      "loss": 0.3783,
      "step": 4220
    },
    {
      "epoch": 4.764044943820225,
      "grad_norm": 0.7421875,
      "learning_rate": 1.7122417108203726e-06,
      "loss": 0.294,
      "step": 4240
    },
    {
      "epoch": 4.786516853932584,
      "grad_norm": 0.75390625,
      "learning_rate": 1.4023592206802382e-06,
      "loss": 0.3194,
      "step": 4260
    },
    {
      "epoch": 4.808988764044944,
      "grad_norm": 1.375,
      "learning_rate": 1.1231892178829472e-06,
      "loss": 0.3145,
      "step": 4280
    },
    {
      "epoch": 4.831460674157303,
      "grad_norm": 1.4140625,
      "learning_rate": 8.74818661965382e-07,
      "loss": 0.3372,
      "step": 4300
    },
    {
      "epoch": 4.853932584269663,
      "grad_norm": 1.6640625,
      "learning_rate": 6.573249186483721e-07,
      "loss": 0.2791,
      "step": 4320
    },
    {
      "epoch": 4.876404494382022,
      "grad_norm": 0.8359375,
      "learning_rate": 4.707757357379383e-07,
      "loss": 0.2428,
      "step": 4340
    },
    {
      "epoch": 4.898876404494382,
      "grad_norm": 1.3671875,
      "learning_rate": 3.152292220222064e-07,
      "loss": 0.3225,
      "step": 4360
    },
    {
      "epoch": 4.921348314606742,
      "grad_norm": 1.0625,
      "learning_rate": 1.9073382917097483e-07,
      "loss": 0.3164,
      "step": 4380
    },
    {
      "epoch": 4.943820224719101,
      "grad_norm": 1.0546875,
      "learning_rate": 9.732833664334307e-08,
      "loss": 0.3571,
      "step": 4400
    },
    {
      "epoch": 4.966292134831461,
      "grad_norm": 0.87890625,
      "learning_rate": 3.5041839608151996e-08,
      "loss": 0.3002,
      "step": 4420
    },
    {
      "epoch": 4.98876404494382,
      "grad_norm": 0.6640625,
      "learning_rate": 3.893739881088987e-09,
      "loss": 0.366,
      "step": 4440
    },
    {
      "epoch": 5.0,
      "step": 4450,
      "total_flos": 1.024663401529344e+17,
      "train_loss": 0.6764815047617708,
      "train_runtime": 9686.8536,
      "train_samples_per_second": 0.459,
      "train_steps_per_second": 0.459
    }
  ],
  "logging_steps": 20,
  "max_steps": 4450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 1.024663401529344e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}