{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.054442732408834,
  "eval_steps": 10,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001027221366204417,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7525,
      "step": 1
    },
    {
      "epoch": 0.002054442732408834,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7605,
      "step": 2
    },
    {
      "epoch": 0.0030816640986132513,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8008,
      "step": 3
    },
    {
      "epoch": 0.004108885464817668,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.809,
      "step": 4
    },
    {
      "epoch": 0.005136106831022085,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7841,
      "step": 5
    },
    {
      "epoch": 0.0061633281972265025,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7924,
      "step": 6
    },
    {
      "epoch": 0.007190549563430919,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7732,
      "step": 7
    },
    {
      "epoch": 0.008217770929635337,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7953,
      "step": 8
    },
    {
      "epoch": 0.009244992295839754,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7747,
      "step": 9
    },
    {
      "epoch": 0.01027221366204417,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7909,
      "step": 10
    },
    {
      "epoch": 0.01027221366204417,
      "eval_loss": 2.811410665512085,
      "eval_runtime": 37.9307,
      "eval_samples_per_second": 26.364,
      "eval_steps_per_second": 1.107,
      "step": 10
    },
    {
      "epoch": 0.011299435028248588,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7592,
      "step": 11
    },
    {
      "epoch": 0.012326656394453005,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7394,
      "step": 12
    },
    {
      "epoch": 0.01335387776065742,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8111,
      "step": 13
    },
    {
      "epoch": 0.014381099126861838,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7786,
      "step": 14
    },
    {
      "epoch": 0.015408320493066256,
      "grad_norm": 5.814355373382568,
      "learning_rate": 1.7123287671232876e-08,
      "loss": 2.7711,
      "step": 15
    },
    {
      "epoch": 0.016435541859270673,
      "grad_norm": 6.2984514236450195,
      "learning_rate": 3.424657534246575e-08,
      "loss": 2.7517,
      "step": 16
    },
    {
      "epoch": 0.01746276322547509,
      "grad_norm": 6.235473155975342,
      "learning_rate": 5.136986301369863e-08,
      "loss": 2.7191,
      "step": 17
    },
    {
      "epoch": 0.01848998459167951,
      "grad_norm": 6.235473155975342,
      "learning_rate": 5.136986301369863e-08,
      "loss": 2.806,
      "step": 18
    },
    {
      "epoch": 0.019517205957883924,
      "grad_norm": 6.621068000793457,
      "learning_rate": 6.84931506849315e-08,
      "loss": 2.7192,
      "step": 19
    },
    {
      "epoch": 0.02054442732408834,
      "grad_norm": 6.781496524810791,
      "learning_rate": 8.561643835616439e-08,
      "loss": 2.7862,
      "step": 20
    },
    {
      "epoch": 0.02054442732408834,
      "eval_loss": 2.8109023571014404,
      "eval_runtime": 37.527,
      "eval_samples_per_second": 26.647,
      "eval_steps_per_second": 1.119,
      "step": 20
    },
    {
      "epoch": 0.02157164869029276,
      "grad_norm": 6.336451530456543,
      "learning_rate": 1.0273972602739726e-07,
      "loss": 2.7997,
      "step": 21
    },
    {
      "epoch": 0.022598870056497175,
      "grad_norm": 6.184426307678223,
      "learning_rate": 1.1986301369863014e-07,
      "loss": 2.7938,
      "step": 22
    },
    {
      "epoch": 0.02362609142270159,
      "grad_norm": 6.443874359130859,
      "learning_rate": 1.36986301369863e-07,
      "loss": 2.7485,
      "step": 23
    },
    {
      "epoch": 0.02465331278890601,
      "grad_norm": 6.17667293548584,
      "learning_rate": 1.541095890410959e-07,
      "loss": 2.7779,
      "step": 24
    },
    {
      "epoch": 0.025680534155110426,
      "grad_norm": 5.850281715393066,
      "learning_rate": 1.7123287671232878e-07,
      "loss": 2.7598,
      "step": 25
    },
    {
      "epoch": 0.02670775552131484,
      "grad_norm": 5.788112640380859,
      "learning_rate": 1.8835616438356165e-07,
      "loss": 2.7837,
      "step": 26
    },
    {
      "epoch": 0.02773497688751926,
      "grad_norm": 5.17655086517334,
      "learning_rate": 2.0547945205479452e-07,
      "loss": 2.7513,
      "step": 27
    },
    {
      "epoch": 0.028762198253723677,
      "grad_norm": 4.698268890380859,
      "learning_rate": 2.226027397260274e-07,
      "loss": 2.7714,
      "step": 28
    },
    {
      "epoch": 0.029789419619928096,
      "grad_norm": 4.562674522399902,
      "learning_rate": 2.397260273972603e-07,
      "loss": 2.7759,
      "step": 29
    },
    {
      "epoch": 0.030816640986132512,
      "grad_norm": 4.138142108917236,
      "learning_rate": 2.568493150684932e-07,
      "loss": 2.7656,
      "step": 30
    },
    {
      "epoch": 0.030816640986132512,
      "eval_loss": 2.784187078475952,
      "eval_runtime": 36.9691,
      "eval_samples_per_second": 27.05,
      "eval_steps_per_second": 1.136,
      "step": 30
    },
    {
      "epoch": 0.03184386235233693,
      "grad_norm": 4.39779806137085,
      "learning_rate": 2.73972602739726e-07,
      "loss": 2.7377,
      "step": 31
    },
    {
      "epoch": 0.03287108371854135,
      "grad_norm": 3.9160122871398926,
      "learning_rate": 2.910958904109589e-07,
      "loss": 2.7753,
      "step": 32
    },
    {
      "epoch": 0.03389830508474576,
      "grad_norm": 4.828811168670654,
      "learning_rate": 3.082191780821918e-07,
      "loss": 2.7311,
      "step": 33
    },
    {
      "epoch": 0.03492552645095018,
      "grad_norm": 4.6300435066223145,
      "learning_rate": 3.2534246575342466e-07,
      "loss": 2.7287,
      "step": 34
    },
    {
      "epoch": 0.035952747817154594,
      "grad_norm": 4.47259521484375,
      "learning_rate": 3.4246575342465755e-07,
      "loss": 2.7117,
      "step": 35
    },
    {
      "epoch": 0.03697996918335902,
      "grad_norm": 4.557999610900879,
      "learning_rate": 3.595890410958904e-07,
      "loss": 2.6291,
      "step": 36
    },
    {
      "epoch": 0.03800719054956343,
      "grad_norm": 4.3974761962890625,
      "learning_rate": 3.767123287671233e-07,
      "loss": 2.7421,
      "step": 37
    },
    {
      "epoch": 0.03903441191576785,
      "grad_norm": 3.6151702404022217,
      "learning_rate": 3.938356164383562e-07,
      "loss": 2.6744,
      "step": 38
    },
    {
      "epoch": 0.040061633281972264,
      "grad_norm": 3.165523052215576,
      "learning_rate": 4.1095890410958903e-07,
      "loss": 2.6694,
      "step": 39
    },
    {
      "epoch": 0.04108885464817668,
      "grad_norm": 3.2827563285827637,
      "learning_rate": 4.2808219178082193e-07,
      "loss": 2.7177,
      "step": 40
    },
    {
      "epoch": 0.04108885464817668,
      "eval_loss": 2.705791711807251,
      "eval_runtime": 36.995,
      "eval_samples_per_second": 27.031,
      "eval_steps_per_second": 1.135,
      "step": 40
    },
    {
      "epoch": 0.042116076014381096,
      "grad_norm": 3.241729736328125,
      "learning_rate": 4.452054794520548e-07,
      "loss": 2.6703,
      "step": 41
    },
    {
      "epoch": 0.04314329738058552,
      "grad_norm": 3.401049852371216,
      "learning_rate": 4.6232876712328767e-07,
      "loss": 2.6036,
      "step": 42
    },
    {
      "epoch": 0.044170518746789934,
      "grad_norm": 3.2498552799224854,
      "learning_rate": 4.794520547945206e-07,
      "loss": 2.6666,
      "step": 43
    },
    {
      "epoch": 0.04519774011299435,
      "grad_norm": 3.1376283168792725,
      "learning_rate": 4.965753424657534e-07,
      "loss": 2.6331,
      "step": 44
    },
    {
      "epoch": 0.046224961479198766,
      "grad_norm": 2.885286569595337,
      "learning_rate": 5.136986301369864e-07,
      "loss": 2.6609,
      "step": 45
    },
    {
      "epoch": 0.04725218284540318,
      "grad_norm": 2.636420249938965,
      "learning_rate": 5.308219178082192e-07,
      "loss": 2.6462,
      "step": 46
    },
    {
      "epoch": 0.048279404211607604,
      "grad_norm": 2.5609490871429443,
      "learning_rate": 5.47945205479452e-07,
      "loss": 2.6821,
      "step": 47
    },
    {
      "epoch": 0.04930662557781202,
      "grad_norm": 2.7683653831481934,
      "learning_rate": 5.65068493150685e-07,
      "loss": 2.61,
      "step": 48
    },
    {
      "epoch": 0.050333846944016436,
      "grad_norm": 3.3999135494232178,
      "learning_rate": 5.821917808219178e-07,
      "loss": 2.6166,
      "step": 49
    },
    {
      "epoch": 0.05136106831022085,
      "grad_norm": 2.922145128250122,
      "learning_rate": 5.993150684931507e-07,
      "loss": 2.5657,
      "step": 50
    },
    {
      "epoch": 0.05136106831022085,
      "eval_loss": 2.6353814601898193,
      "eval_runtime": 37.0591,
      "eval_samples_per_second": 26.984,
      "eval_steps_per_second": 1.133,
      "step": 50
    },
    {
      "epoch": 0.05238828967642527,
      "grad_norm": 2.5821900367736816,
      "learning_rate": 6.164383561643836e-07,
      "loss": 2.5512,
      "step": 51
    },
    {
      "epoch": 0.05341551104262968,
      "grad_norm": 2.2299063205718994,
      "learning_rate": 6.335616438356165e-07,
      "loss": 2.6082,
      "step": 52
    },
    {
      "epoch": 0.054442732408834106,
      "grad_norm": 2.0938310623168945,
      "learning_rate": 6.506849315068493e-07,
      "loss": 2.5301,
      "step": 53
    },
    {
      "epoch": 0.05546995377503852,
      "grad_norm": 2.186629295349121,
      "learning_rate": 6.678082191780823e-07,
      "loss": 2.5755,
      "step": 54
    },
    {
      "epoch": 0.05649717514124294,
      "grad_norm": 2.1747493743896484,
      "learning_rate": 6.849315068493151e-07,
      "loss": 2.5451,
      "step": 55
    },
    {
      "epoch": 0.05752439650744735,
      "grad_norm": 2.5470354557037354,
      "learning_rate": 7.020547945205481e-07,
      "loss": 2.5159,
      "step": 56
    },
    {
      "epoch": 0.05855161787365177,
      "grad_norm": 2.5056023597717285,
      "learning_rate": 7.191780821917808e-07,
      "loss": 2.5431,
      "step": 57
    },
    {
      "epoch": 0.05957883923985619,
      "grad_norm": 2.6059749126434326,
      "learning_rate": 7.363013698630137e-07,
      "loss": 2.4684,
      "step": 58
    },
    {
      "epoch": 0.06060606060606061,
      "grad_norm": 1.87150239944458,
      "learning_rate": 7.534246575342466e-07,
      "loss": 2.5269,
      "step": 59
    },
    {
      "epoch": 0.061633281972265024,
      "grad_norm": 1.8777480125427246,
      "learning_rate": 7.705479452054795e-07,
      "loss": 2.5253,
      "step": 60
    },
    {
      "epoch": 0.061633281972265024,
      "eval_loss": 2.5705313682556152,
      "eval_runtime": 37.0929,
      "eval_samples_per_second": 26.959,
      "eval_steps_per_second": 1.132,
      "step": 60
    },
    {
      "epoch": 0.06266050333846944,
      "grad_norm": 1.8500003814697266,
      "learning_rate": 7.876712328767124e-07,
      "loss": 2.4987,
      "step": 61
    },
    {
      "epoch": 0.06368772470467386,
      "grad_norm": 1.9473347663879395,
      "learning_rate": 8.047945205479453e-07,
      "loss": 2.4866,
      "step": 62
    },
    {
      "epoch": 0.06471494607087827,
      "grad_norm": 2.2350873947143555,
      "learning_rate": 8.219178082191781e-07,
      "loss": 2.5343,
      "step": 63
    },
    {
      "epoch": 0.0657421674370827,
      "grad_norm": 2.1791350841522217,
      "learning_rate": 8.39041095890411e-07,
      "loss": 2.5185,
      "step": 64
    },
    {
      "epoch": 0.0667693888032871,
      "grad_norm": 1.8400903940200806,
      "learning_rate": 8.561643835616439e-07,
      "loss": 2.3897,
      "step": 65
    },
    {
      "epoch": 0.06779661016949153,
      "grad_norm": 1.7193644046783447,
      "learning_rate": 8.732876712328768e-07,
      "loss": 2.485,
      "step": 66
    },
    {
      "epoch": 0.06882383153569595,
      "grad_norm": 1.8478909730911255,
      "learning_rate": 8.904109589041097e-07,
      "loss": 2.4404,
      "step": 67
    },
    {
      "epoch": 0.06985105290190036,
      "grad_norm": 1.8803719282150269,
      "learning_rate": 9.075342465753426e-07,
      "loss": 2.4933,
      "step": 68
    },
    {
      "epoch": 0.07087827426810478,
      "grad_norm": 1.6866114139556885,
      "learning_rate": 9.246575342465753e-07,
      "loss": 2.4441,
      "step": 69
    },
    {
      "epoch": 0.07190549563430919,
      "grad_norm": 1.807783842086792,
      "learning_rate": 9.417808219178083e-07,
      "loss": 2.5477,
      "step": 70
    },
    {
      "epoch": 0.07190549563430919,
      "eval_loss": 2.5140035152435303,
      "eval_runtime": 37.1384,
      "eval_samples_per_second": 26.926,
      "eval_steps_per_second": 1.131,
      "step": 70
    },
    {
      "epoch": 0.07293271700051361,
      "grad_norm": 1.7754000425338745,
      "learning_rate": 9.589041095890411e-07,
      "loss": 2.44,
      "step": 71
    },
    {
      "epoch": 0.07395993836671803,
      "grad_norm": 1.7819652557373047,
      "learning_rate": 9.76027397260274e-07,
      "loss": 2.402,
      "step": 72
    },
    {
      "epoch": 0.07498715973292244,
      "grad_norm": 1.6666278839111328,
      "learning_rate": 9.931506849315068e-07,
      "loss": 2.4511,
      "step": 73
    },
    {
      "epoch": 0.07601438109912687,
      "grad_norm": 2.0625369548797607,
      "learning_rate": 1.0102739726027399e-06,
      "loss": 2.424,
      "step": 74
    },
    {
      "epoch": 0.07704160246533127,
      "grad_norm": 1.9408000707626343,
      "learning_rate": 1.0273972602739727e-06,
      "loss": 2.4286,
      "step": 75
    },
    {
      "epoch": 0.0780688238315357,
      "grad_norm": 1.7670842409133911,
      "learning_rate": 1.0445205479452056e-06,
      "loss": 2.407,
      "step": 76
    },
    {
      "epoch": 0.07909604519774012,
      "grad_norm": 3.045717716217041,
      "learning_rate": 1.0616438356164384e-06,
      "loss": 2.425,
      "step": 77
    },
    {
      "epoch": 0.08012326656394453,
      "grad_norm": 1.9278619289398193,
      "learning_rate": 1.0787671232876712e-06,
      "loss": 2.4306,
      "step": 78
    },
    {
      "epoch": 0.08115048793014895,
      "grad_norm": 2.405146837234497,
      "learning_rate": 1.095890410958904e-06,
      "loss": 2.3499,
      "step": 79
    },
    {
      "epoch": 0.08217770929635336,
      "grad_norm": 2.1083860397338867,
      "learning_rate": 1.1130136986301371e-06,
      "loss": 2.4115,
      "step": 80
    },
    {
      "epoch": 0.08217770929635336,
      "eval_loss": 2.4501075744628906,
      "eval_runtime": 37.1247,
      "eval_samples_per_second": 26.936,
      "eval_steps_per_second": 1.131,
      "step": 80
    },
    {
      "epoch": 0.08320493066255778,
      "grad_norm": 2.181162118911743,
      "learning_rate": 1.13013698630137e-06,
      "loss": 2.3517,
      "step": 81
    },
    {
      "epoch": 0.08423215202876219,
      "grad_norm": 2.1833999156951904,
      "learning_rate": 1.1472602739726028e-06,
      "loss": 2.38,
      "step": 82
    },
    {
      "epoch": 0.08525937339496661,
      "grad_norm": 2.670384645462036,
      "learning_rate": 1.1643835616438357e-06,
      "loss": 2.3833,
      "step": 83
    },
    {
      "epoch": 0.08628659476117104,
      "grad_norm": 2.438676118850708,
      "learning_rate": 1.1815068493150685e-06,
      "loss": 2.3399,
      "step": 84
    },
    {
      "epoch": 0.08731381612737545,
      "grad_norm": 2.7584919929504395,
      "learning_rate": 1.1986301369863014e-06,
      "loss": 2.31,
      "step": 85
    },
    {
      "epoch": 0.08834103749357987,
      "grad_norm": 2.4129064083099365,
      "learning_rate": 1.2157534246575344e-06,
      "loss": 2.3758,
      "step": 86
    },
    {
      "epoch": 0.08936825885978428,
      "grad_norm": 2.366154193878174,
      "learning_rate": 1.2328767123287673e-06,
      "loss": 2.3154,
      "step": 87
    },
    {
      "epoch": 0.0903954802259887,
      "grad_norm": 2.4916300773620605,
      "learning_rate": 1.25e-06,
      "loss": 2.3022,
      "step": 88
    },
    {
      "epoch": 0.09142270159219312,
      "grad_norm": 2.4606640338897705,
      "learning_rate": 1.267123287671233e-06,
      "loss": 2.3103,
      "step": 89
    },
    {
      "epoch": 0.09244992295839753,
      "grad_norm": 2.4134910106658936,
      "learning_rate": 1.284246575342466e-06,
      "loss": 2.2672,
      "step": 90
    },
    {
      "epoch": 0.09244992295839753,
      "eval_loss": 2.3742051124572754,
      "eval_runtime": 37.1258,
      "eval_samples_per_second": 26.935,
      "eval_steps_per_second": 1.131,
      "step": 90
    },
    {
      "epoch": 0.09347714432460195,
      "grad_norm": 2.302776336669922,
      "learning_rate": 1.3013698630136986e-06,
      "loss": 2.2798,
      "step": 91
    },
    {
      "epoch": 0.09450436569080636,
      "grad_norm": 3.397819757461548,
      "learning_rate": 1.3184931506849317e-06,
      "loss": 2.2947,
      "step": 92
    },
    {
      "epoch": 0.09553158705701079,
      "grad_norm": 2.6030948162078857,
      "learning_rate": 1.3356164383561645e-06,
      "loss": 2.2734,
      "step": 93
    },
    {
      "epoch": 0.09655880842321521,
      "grad_norm": 2.9380528926849365,
      "learning_rate": 1.3527397260273976e-06,
      "loss": 2.2635,
      "step": 94
    },
    {
      "epoch": 0.09758602978941962,
      "grad_norm": 2.9196345806121826,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 2.3161,
      "step": 95
    },
    {
      "epoch": 0.09861325115562404,
      "grad_norm": 2.2380659580230713,
      "learning_rate": 1.386986301369863e-06,
      "loss": 2.21,
      "step": 96
    },
    {
      "epoch": 0.09964047252182845,
      "grad_norm": 2.730910539627075,
      "learning_rate": 1.4041095890410961e-06,
      "loss": 2.2562,
      "step": 97
    },
    {
      "epoch": 0.10066769388803287,
      "grad_norm": 2.994748830795288,
      "learning_rate": 1.421232876712329e-06,
      "loss": 2.1597,
      "step": 98
    },
    {
      "epoch": 0.1016949152542373,
      "grad_norm": 2.2228171825408936,
      "learning_rate": 1.4383561643835616e-06,
      "loss": 2.1835,
      "step": 99
    },
    {
      "epoch": 0.1027221366204417,
      "grad_norm": 2.715527296066284,
      "learning_rate": 1.4554794520547946e-06,
      "loss": 2.2074,
      "step": 100
    },
    {
      "epoch": 0.1027221366204417,
      "eval_loss": 2.3061234951019287,
      "eval_runtime": 37.1393,
      "eval_samples_per_second": 26.926,
      "eval_steps_per_second": 1.131,
      "step": 100
    },
    {
      "epoch": 0.10374935798664613,
      "grad_norm": 2.7054452896118164,
      "learning_rate": 1.4726027397260275e-06,
      "loss": 2.2089,
      "step": 101
    },
    {
      "epoch": 0.10477657935285054,
      "grad_norm": 2.795694589614868,
      "learning_rate": 1.4897260273972605e-06,
      "loss": 2.1802,
      "step": 102
    },
    {
      "epoch": 0.10580380071905496,
      "grad_norm": 2.634822368621826,
      "learning_rate": 1.5068493150684932e-06,
      "loss": 2.1428,
      "step": 103
    },
    {
      "epoch": 0.10683102208525937,
      "grad_norm": 2.2418437004089355,
      "learning_rate": 1.5239726027397262e-06,
      "loss": 2.1706,
      "step": 104
    },
    {
      "epoch": 0.10785824345146379,
      "grad_norm": 2.471231698989868,
      "learning_rate": 1.541095890410959e-06,
      "loss": 2.2248,
      "step": 105
    },
    {
      "epoch": 0.10888546481766821,
      "grad_norm": 2.4154577255249023,
      "learning_rate": 1.5582191780821921e-06,
      "loss": 2.1873,
      "step": 106
    },
    {
      "epoch": 0.10991268618387262,
      "grad_norm": 2.2823996543884277,
      "learning_rate": 1.5753424657534248e-06,
      "loss": 2.2237,
      "step": 107
    },
    {
      "epoch": 0.11093990755007704,
      "grad_norm": 2.8695461750030518,
      "learning_rate": 1.5924657534246576e-06,
      "loss": 2.1966,
      "step": 108
    },
    {
      "epoch": 0.11196712891628145,
      "grad_norm": 3.319650650024414,
      "learning_rate": 1.6095890410958907e-06,
      "loss": 2.066,
      "step": 109
    },
    {
      "epoch": 0.11299435028248588,
      "grad_norm": 2.815196990966797,
      "learning_rate": 1.6267123287671235e-06,
      "loss": 2.1355,
      "step": 110
    },
    {
      "epoch": 0.11299435028248588,
      "eval_loss": 2.250187635421753,
      "eval_runtime": 37.1645,
      "eval_samples_per_second": 26.907,
      "eval_steps_per_second": 1.13,
      "step": 110
    },
    {
      "epoch": 0.1140215716486903,
      "grad_norm": 2.9195759296417236,
      "learning_rate": 1.6438356164383561e-06,
      "loss": 2.0713,
      "step": 111
    },
    {
      "epoch": 0.1150487930148947,
      "grad_norm": 2.5600695610046387,
      "learning_rate": 1.6609589041095892e-06,
      "loss": 2.1573,
      "step": 112
    },
    {
      "epoch": 0.11607601438109913,
      "grad_norm": 2.7566795349121094,
      "learning_rate": 1.678082191780822e-06,
      "loss": 2.0909,
      "step": 113
    },
    {
      "epoch": 0.11710323574730354,
      "grad_norm": 2.5593106746673584,
      "learning_rate": 1.695205479452055e-06,
      "loss": 2.144,
      "step": 114
    },
    {
      "epoch": 0.11813045711350796,
      "grad_norm": 2.594931125640869,
      "learning_rate": 1.7123287671232877e-06,
      "loss": 2.0606,
      "step": 115
    },
    {
      "epoch": 0.11915767847971238,
      "grad_norm": 2.769916534423828,
      "learning_rate": 1.7294520547945206e-06,
      "loss": 2.1737,
      "step": 116
    },
    {
      "epoch": 0.12018489984591679,
      "grad_norm": 2.523179054260254,
      "learning_rate": 1.7465753424657536e-06,
      "loss": 2.1247,
      "step": 117
    },
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 2.8153676986694336,
      "learning_rate": 1.7636986301369865e-06,
      "loss": 2.1257,
      "step": 118
    },
    {
      "epoch": 0.12223934257832562,
      "grad_norm": 2.5389721393585205,
      "learning_rate": 1.7808219178082193e-06,
      "loss": 2.1049,
      "step": 119
    },
    {
      "epoch": 0.12326656394453005,
      "grad_norm": 2.9753856658935547,
      "learning_rate": 1.7979452054794521e-06,
      "loss": 2.1871,
      "step": 120
    },
    {
      "epoch": 0.12326656394453005,
      "eval_loss": 2.2089297771453857,
      "eval_runtime": 37.897,
      "eval_samples_per_second": 26.387,
      "eval_steps_per_second": 1.108,
      "step": 120
    },
    {
      "epoch": 0.12429378531073447,
      "grad_norm": 2.737180471420288,
      "learning_rate": 1.8150684931506852e-06,
      "loss": 2.1166,
      "step": 121
    },
    {
      "epoch": 0.12532100667693888,
      "grad_norm": 2.783801794052124,
      "learning_rate": 1.832191780821918e-06,
      "loss": 2.0932,
      "step": 122
    },
    {
      "epoch": 0.1263482280431433,
      "grad_norm": 2.9615402221679688,
      "learning_rate": 1.8493150684931507e-06,
      "loss": 2.0327,
      "step": 123
    },
    {
      "epoch": 0.12737544940934772,
      "grad_norm": 3.053622007369995,
      "learning_rate": 1.8664383561643837e-06,
      "loss": 2.0665,
      "step": 124
    },
    {
      "epoch": 0.12840267077555212,
      "grad_norm": 3.265355348587036,
      "learning_rate": 1.8835616438356166e-06,
      "loss": 2.0243,
      "step": 125
    },
    {
      "epoch": 0.12942989214175654,
      "grad_norm": 2.4237143993377686,
      "learning_rate": 1.9006849315068496e-06,
      "loss": 2.0902,
      "step": 126
    },
    {
      "epoch": 0.13045711350796096,
      "grad_norm": 2.926950693130493,
      "learning_rate": 1.9178082191780823e-06,
      "loss": 2.0057,
      "step": 127
    },
    {
      "epoch": 0.1314843348741654,
      "grad_norm": 2.5317165851593018,
      "learning_rate": 1.9349315068493153e-06,
      "loss": 2.0719,
      "step": 128
    },
    {
      "epoch": 0.1325115562403698,
      "grad_norm": 2.4999701976776123,
      "learning_rate": 1.952054794520548e-06,
      "loss": 2.0969,
      "step": 129
    },
    {
      "epoch": 0.1335387776065742,
      "grad_norm": 2.7063920497894287,
      "learning_rate": 1.969178082191781e-06,
      "loss": 2.0902,
      "step": 130
    },
    {
      "epoch": 0.1335387776065742,
      "eval_loss": 2.175224542617798,
      "eval_runtime": 37.4519,
      "eval_samples_per_second": 26.701,
      "eval_steps_per_second": 1.121,
      "step": 130
    },
    {
      "epoch": 0.13456599897277863,
      "grad_norm": 2.650860071182251,
      "learning_rate": 1.9863013698630136e-06,
      "loss": 2.0191,
      "step": 131
    },
    {
      "epoch": 0.13559322033898305,
      "grad_norm": 2.79182505607605,
      "learning_rate": 2.0034246575342467e-06,
      "loss": 2.0856,
      "step": 132
    },
    {
      "epoch": 0.13662044170518747,
      "grad_norm": 2.453162431716919,
      "learning_rate": 2.0205479452054797e-06,
      "loss": 2.0679,
      "step": 133
    },
    {
      "epoch": 0.1376476630713919,
      "grad_norm": 2.9630799293518066,
      "learning_rate": 2.037671232876713e-06,
      "loss": 2.0038,
      "step": 134
    },
    {
      "epoch": 0.1386748844375963,
      "grad_norm": 3.0807266235351562,
      "learning_rate": 2.0547945205479454e-06,
      "loss": 2.0367,
      "step": 135
    },
    {
      "epoch": 0.1397021058038007,
      "grad_norm": 2.9712910652160645,
      "learning_rate": 2.071917808219178e-06,
      "loss": 2.0571,
      "step": 136
    },
    {
      "epoch": 0.14072932717000514,
      "grad_norm": 3.0804853439331055,
      "learning_rate": 2.089041095890411e-06,
      "loss": 1.9725,
      "step": 137
    },
    {
      "epoch": 0.14175654853620956,
      "grad_norm": 2.659608840942383,
      "learning_rate": 2.106164383561644e-06,
      "loss": 2.0572,
      "step": 138
    },
    {
      "epoch": 0.14278376990241398,
      "grad_norm": 2.8093109130859375,
      "learning_rate": 2.123287671232877e-06,
      "loss": 2.0178,
      "step": 139
    },
    {
      "epoch": 0.14381099126861838,
      "grad_norm": 2.8163881301879883,
      "learning_rate": 2.14041095890411e-06,
      "loss": 2.0085,
      "step": 140
    },
    {
      "epoch": 0.14381099126861838,
      "eval_loss": 2.147092342376709,
      "eval_runtime": 37.2453,
      "eval_samples_per_second": 26.849,
      "eval_steps_per_second": 1.128,
      "step": 140
    },
    {
      "epoch": 0.1448382126348228,
      "grad_norm": 3.795241594314575,
      "learning_rate": 2.1575342465753425e-06,
      "loss": 2.0149,
      "step": 141
    },
    {
      "epoch": 0.14586543400102722,
      "grad_norm": 3.3287174701690674,
      "learning_rate": 2.1746575342465755e-06,
      "loss": 1.9658,
      "step": 142
    },
    {
      "epoch": 0.14689265536723164,
      "grad_norm": 2.803213596343994,
      "learning_rate": 2.191780821917808e-06,
      "loss": 2.0983,
      "step": 143
    },
    {
      "epoch": 0.14791987673343607,
      "grad_norm": 3.2640199661254883,
      "learning_rate": 2.2089041095890412e-06,
      "loss": 1.9865,
      "step": 144
    },
    {
      "epoch": 0.14894709809964046,
      "grad_norm": 2.95979380607605,
      "learning_rate": 2.2260273972602743e-06,
      "loss": 1.9669,
      "step": 145
    },
    {
      "epoch": 0.14997431946584489,
      "grad_norm": 3.737175226211548,
      "learning_rate": 2.243150684931507e-06,
      "loss": 1.9691,
      "step": 146
    },
    {
      "epoch": 0.1510015408320493,
      "grad_norm": 2.7248430252075195,
      "learning_rate": 2.26027397260274e-06,
      "loss": 1.9764,
      "step": 147
    },
    {
      "epoch": 0.15202876219825373,
      "grad_norm": 2.929856061935425,
      "learning_rate": 2.2773972602739726e-06,
      "loss": 1.9967,
      "step": 148
    },
    {
      "epoch": 0.15305598356445815,
      "grad_norm": 2.8998641967773438,
      "learning_rate": 2.2945205479452057e-06,
      "loss": 2.0842,
      "step": 149
    },
    {
      "epoch": 0.15408320493066255,
      "grad_norm": 2.9672694206237793,
      "learning_rate": 2.3116438356164387e-06,
      "loss": 1.9915,
      "step": 150
    },
    {
      "epoch": 0.15408320493066255,
      "eval_loss": 2.122715473175049,
      "eval_runtime": 37.1938,
      "eval_samples_per_second": 26.886,
      "eval_steps_per_second": 1.129,
      "step": 150
    },
    {
      "epoch": 0.15511042629686697,
      "grad_norm": 2.8529887199401855,
      "learning_rate": 2.3287671232876713e-06,
      "loss": 1.9022,
      "step": 151
    },
    {
      "epoch": 0.1561376476630714,
      "grad_norm": 2.914121389389038,
      "learning_rate": 2.3458904109589044e-06,
      "loss": 1.956,
      "step": 152
    },
    {
      "epoch": 0.15716486902927582,
      "grad_norm": 3.6286494731903076,
      "learning_rate": 2.363013698630137e-06,
      "loss": 2.0118,
      "step": 153
    },
    {
      "epoch": 0.15819209039548024,
      "grad_norm": 2.789562702178955,
      "learning_rate": 2.38013698630137e-06,
      "loss": 1.9935,
      "step": 154
    },
    {
      "epoch": 0.15921931176168463,
      "grad_norm": 3.367119550704956,
      "learning_rate": 2.3972602739726027e-06,
      "loss": 1.9519,
      "step": 155
    },
    {
      "epoch": 0.16024653312788906,
      "grad_norm": 2.9345803260803223,
      "learning_rate": 2.4143835616438358e-06,
      "loss": 1.9017,
      "step": 156
    },
    {
      "epoch": 0.16127375449409348,
      "grad_norm": 2.8674590587615967,
      "learning_rate": 2.431506849315069e-06,
      "loss": 1.9731,
      "step": 157
    },
    {
      "epoch": 0.1623009758602979,
      "grad_norm": 3.144766092300415,
      "learning_rate": 2.4486301369863015e-06,
      "loss": 1.9811,
      "step": 158
    },
    {
      "epoch": 0.1633281972265023,
      "grad_norm": 3.845395088195801,
      "learning_rate": 2.4657534246575345e-06,
      "loss": 1.9523,
      "step": 159
    },
    {
      "epoch": 0.16435541859270672,
      "grad_norm": 2.776700973510742,
      "learning_rate": 2.482876712328767e-06,
      "loss": 1.9687,
      "step": 160
    },
    {
      "epoch": 0.16435541859270672,
      "eval_loss": 2.0929579734802246,
      "eval_runtime": 37.1917,
      "eval_samples_per_second": 26.888,
      "eval_steps_per_second": 1.129,
      "step": 160
    },
    {
      "epoch": 0.16538263995891114,
      "grad_norm": 3.4260904788970947,
      "learning_rate": 2.5e-06,
      "loss": 1.9665,
      "step": 161
    },
    {
      "epoch": 0.16640986132511557,
      "grad_norm": 3.1217620372772217,
      "learning_rate": 2.5171232876712333e-06,
      "loss": 1.9361,
      "step": 162
    },
    {
      "epoch": 0.16743708269132,
      "grad_norm": 3.342068910598755,
      "learning_rate": 2.534246575342466e-06,
      "loss": 1.9084,
      "step": 163
    },
    {
      "epoch": 0.16846430405752438,
      "grad_norm": 3.6485838890075684,
      "learning_rate": 2.551369863013699e-06,
      "loss": 1.9542,
      "step": 164
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 3.1247780323028564,
      "learning_rate": 2.568493150684932e-06,
      "loss": 1.9717,
      "step": 165
    },
    {
      "epoch": 0.17051874678993323,
      "grad_norm": 2.9856860637664795,
      "learning_rate": 2.585616438356164e-06,
      "loss": 1.9574,
      "step": 166
    },
    {
      "epoch": 0.17154596815613765,
      "grad_norm": 3.337902069091797,
      "learning_rate": 2.6027397260273973e-06,
      "loss": 1.8884,
      "step": 167
    },
    {
      "epoch": 0.17257318952234207,
      "grad_norm": 2.844534158706665,
      "learning_rate": 2.6198630136986303e-06,
      "loss": 1.9545,
      "step": 168
    },
    {
      "epoch": 0.17360041088854647,
      "grad_norm": 3.75907564163208,
      "learning_rate": 2.6369863013698634e-06,
      "loss": 1.9907,
      "step": 169
    },
    {
      "epoch": 0.1746276322547509,
      "grad_norm": 2.9321961402893066,
      "learning_rate": 2.654109589041096e-06,
      "loss": 1.9663,
      "step": 170
    },
    {
      "epoch": 0.1746276322547509,
      "eval_loss": 2.0753843784332275,
      "eval_runtime": 37.2389,
      "eval_samples_per_second": 26.854,
      "eval_steps_per_second": 1.128,
      "step": 170
    },
    {
      "epoch": 0.17565485362095531,
      "grad_norm": 2.952523946762085,
      "learning_rate": 2.671232876712329e-06,
      "loss": 1.9065,
      "step": 171
    },
    {
      "epoch": 0.17668207498715974,
      "grad_norm": 2.952463150024414,
      "learning_rate": 2.688356164383562e-06,
      "loss": 1.8994,
      "step": 172
    },
    {
      "epoch": 0.17770929635336416,
      "grad_norm": 3.1352880001068115,
      "learning_rate": 2.705479452054795e-06,
      "loss": 1.8902,
      "step": 173
    },
    {
      "epoch": 0.17873651771956856,
      "grad_norm": 3.5727322101593018,
      "learning_rate": 2.7226027397260274e-06,
      "loss": 1.8454,
      "step": 174
    },
    {
      "epoch": 0.17976373908577298,
      "grad_norm": 3.0562493801116943,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 1.984,
      "step": 175
    },
    {
      "epoch": 0.1807909604519774,
      "grad_norm": 3.367924690246582,
      "learning_rate": 2.7568493150684935e-06,
      "loss": 1.9398,
      "step": 176
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 3.1982340812683105,
      "learning_rate": 2.773972602739726e-06,
      "loss": 1.8566,
      "step": 177
    },
    {
      "epoch": 0.18284540318438625,
      "grad_norm": 3.8962080478668213,
      "learning_rate": 2.791095890410959e-06,
      "loss": 1.9459,
      "step": 178
    },
    {
      "epoch": 0.18387262455059064,
      "grad_norm": 3.3503036499023438,
      "learning_rate": 2.8082191780821922e-06,
      "loss": 1.8348,
      "step": 179
    },
    {
      "epoch": 0.18489984591679506,
      "grad_norm": 3.276139497756958,
      "learning_rate": 2.8253424657534253e-06,
      "loss": 1.9849,
      "step": 180
    },
    {
      "epoch": 0.18489984591679506,
      "eval_loss": 2.0605413913726807,
      "eval_runtime": 37.2276,
      "eval_samples_per_second": 26.862,
      "eval_steps_per_second": 1.128,
      "step": 180
    },
    {
      "epoch": 0.1859270672829995,
      "grad_norm": 3.3787975311279297,
      "learning_rate": 2.842465753424658e-06,
      "loss": 1.9517,
      "step": 181
    },
    {
      "epoch": 0.1869542886492039,
      "grad_norm": 3.054290294647217,
      "learning_rate": 2.8595890410958905e-06,
      "loss": 1.9572,
      "step": 182
    },
    {
      "epoch": 0.18798151001540833,
      "grad_norm": 3.0737054347991943,
      "learning_rate": 2.876712328767123e-06,
      "loss": 1.9674,
      "step": 183
    },
    {
      "epoch": 0.18900873138161273,
      "grad_norm": 3.2833855152130127,
      "learning_rate": 2.8938356164383562e-06,
      "loss": 1.8185,
      "step": 184
    },
    {
      "epoch": 0.19003595274781715,
      "grad_norm": 2.70452880859375,
      "learning_rate": 2.9109589041095893e-06,
      "loss": 1.8552,
      "step": 185
    },
    {
      "epoch": 0.19106317411402157,
      "grad_norm": 3.2723655700683594,
      "learning_rate": 2.9280821917808223e-06,
      "loss": 1.8433,
      "step": 186
    },
    {
      "epoch": 0.192090395480226,
      "grad_norm": 2.844536066055298,
      "learning_rate": 2.945205479452055e-06,
      "loss": 1.8488,
      "step": 187
    },
    {
      "epoch": 0.19311761684643042,
      "grad_norm": 2.944545269012451,
      "learning_rate": 2.962328767123288e-06,
      "loss": 1.9131,
      "step": 188
    },
    {
      "epoch": 0.1941448382126348,
      "grad_norm": 2.812204122543335,
      "learning_rate": 2.979452054794521e-06,
      "loss": 1.8977,
      "step": 189
    },
    {
      "epoch": 0.19517205957883924,
      "grad_norm": 2.8999950885772705,
      "learning_rate": 2.9965753424657533e-06,
      "loss": 1.8477,
      "step": 190
    },
    {
      "epoch": 0.19517205957883924,
      "eval_loss": 2.051161527633667,
      "eval_runtime": 37.2363,
      "eval_samples_per_second": 26.856,
      "eval_steps_per_second": 1.128,
      "step": 190
    },
    {
      "epoch": 0.19619928094504366,
      "grad_norm": 2.7814741134643555,
      "learning_rate": 3.0136986301369864e-06,
      "loss": 1.8989,
      "step": 191
    },
    {
      "epoch": 0.19722650231124808,
      "grad_norm": 2.93346905708313,
      "learning_rate": 3.0308219178082194e-06,
      "loss": 1.8748,
      "step": 192
    },
    {
      "epoch": 0.1982537236774525,
      "grad_norm": 3.00400447845459,
      "learning_rate": 3.0479452054794525e-06,
      "loss": 1.838,
      "step": 193
    },
    {
      "epoch": 0.1992809450436569,
      "grad_norm": 2.949695110321045,
      "learning_rate": 3.065068493150685e-06,
      "loss": 1.8817,
      "step": 194
    },
    {
      "epoch": 0.20030816640986132,
      "grad_norm": 2.8858301639556885,
      "learning_rate": 3.082191780821918e-06,
      "loss": 1.8333,
      "step": 195
    },
    {
      "epoch": 0.20133538777606574,
      "grad_norm": 3.123706102371216,
      "learning_rate": 3.099315068493151e-06,
      "loss": 1.8533,
      "step": 196
    },
    {
      "epoch": 0.20236260914227017,
      "grad_norm": 3.0130505561828613,
      "learning_rate": 3.1164383561643843e-06,
      "loss": 1.842,
      "step": 197
    },
    {
      "epoch": 0.2033898305084746,
      "grad_norm": 2.8823280334472656,
      "learning_rate": 3.1335616438356165e-06,
      "loss": 1.8025,
      "step": 198
    },
    {
      "epoch": 0.20441705187467898,
      "grad_norm": 2.920884132385254,
      "learning_rate": 3.1506849315068495e-06,
      "loss": 1.9488,
      "step": 199
    },
    {
      "epoch": 0.2054442732408834,
      "grad_norm": 2.9152207374572754,
      "learning_rate": 3.167808219178082e-06,
      "loss": 1.8163,
      "step": 200
    },
    {
      "epoch": 0.2054442732408834,
      "eval_loss": 2.033583641052246,
      "eval_runtime": 37.127,
      "eval_samples_per_second": 26.935,
      "eval_steps_per_second": 1.131,
      "step": 200
    },
    {
      "epoch": 0.20647149460708783,
      "grad_norm": 2.904230833053589,
      "learning_rate": 3.184931506849315e-06,
      "loss": 1.8207,
      "step": 201
    },
    {
      "epoch": 0.20749871597329225,
      "grad_norm": 3.128448486328125,
      "learning_rate": 3.2020547945205483e-06,
      "loss": 1.8051,
      "step": 202
    },
    {
      "epoch": 0.20852593733949668,
      "grad_norm": 3.0049452781677246,
      "learning_rate": 3.2191780821917813e-06,
      "loss": 1.7823,
      "step": 203
    },
    {
      "epoch": 0.20955315870570107,
      "grad_norm": 2.7612497806549072,
      "learning_rate": 3.236301369863014e-06,
      "loss": 1.7703,
      "step": 204
    },
    {
      "epoch": 0.2105803800719055,
      "grad_norm": 2.9178879261016846,
      "learning_rate": 3.253424657534247e-06,
      "loss": 1.7427,
      "step": 205
    },
    {
      "epoch": 0.21160760143810992,
      "grad_norm": 2.836967945098877,
      "learning_rate": 3.2705479452054796e-06,
      "loss": 1.8347,
      "step": 206
    },
    {
      "epoch": 0.21263482280431434,
      "grad_norm": 2.8437204360961914,
      "learning_rate": 3.2876712328767123e-06,
      "loss": 1.8147,
      "step": 207
    },
    {
      "epoch": 0.21366204417051873,
      "grad_norm": 3.0728225708007812,
      "learning_rate": 3.3047945205479453e-06,
      "loss": 1.7823,
      "step": 208
    },
    {
      "epoch": 0.21468926553672316,
      "grad_norm": 3.357631206512451,
      "learning_rate": 3.3219178082191784e-06,
      "loss": 1.8013,
      "step": 209
    },
    {
      "epoch": 0.21571648690292758,
      "grad_norm": 2.7996842861175537,
      "learning_rate": 3.3390410958904114e-06,
      "loss": 1.8279,
      "step": 210
    },
    {
      "epoch": 0.21571648690292758,
      "eval_loss": 2.024500846862793,
      "eval_runtime": 37.0122,
      "eval_samples_per_second": 27.018,
      "eval_steps_per_second": 1.135,
      "step": 210
    },
    {
      "epoch": 0.216743708269132,
      "grad_norm": 3.3635380268096924,
      "learning_rate": 3.356164383561644e-06,
      "loss": 1.8509,
      "step": 211
    },
    {
      "epoch": 0.21777092963533642,
      "grad_norm": 2.703932523727417,
      "learning_rate": 3.373287671232877e-06,
      "loss": 1.8984,
      "step": 212
    },
    {
      "epoch": 0.21879815100154082,
      "grad_norm": 2.9341695308685303,
      "learning_rate": 3.39041095890411e-06,
      "loss": 1.8381,
      "step": 213
    },
    {
      "epoch": 0.21982537236774524,
      "grad_norm": 3.072448253631592,
      "learning_rate": 3.4075342465753424e-06,
      "loss": 1.8779,
      "step": 214
    },
    {
      "epoch": 0.22085259373394966,
      "grad_norm": 2.850555419921875,
      "learning_rate": 3.4246575342465754e-06,
      "loss": 1.7656,
      "step": 215
    },
    {
      "epoch": 0.2218798151001541,
      "grad_norm": 3.0218873023986816,
      "learning_rate": 3.4417808219178085e-06,
      "loss": 1.829,
      "step": 216
    },
    {
      "epoch": 0.2229070364663585,
      "grad_norm": 2.8475091457366943,
      "learning_rate": 3.458904109589041e-06,
      "loss": 1.8121,
      "step": 217
    },
    {
      "epoch": 0.2239342578325629,
      "grad_norm": 2.8760111331939697,
      "learning_rate": 3.476027397260274e-06,
      "loss": 1.8281,
      "step": 218
    },
    {
      "epoch": 0.22496147919876733,
      "grad_norm": 2.976670265197754,
      "learning_rate": 3.4931506849315072e-06,
      "loss": 1.8274,
      "step": 219
    },
    {
      "epoch": 0.22598870056497175,
      "grad_norm": 2.912505865097046,
      "learning_rate": 3.5102739726027403e-06,
      "loss": 1.8316,
      "step": 220
    },
    {
      "epoch": 0.22598870056497175,
      "eval_loss": 2.0203635692596436,
      "eval_runtime": 37.0433,
      "eval_samples_per_second": 26.995,
      "eval_steps_per_second": 1.134,
      "step": 220
    },
    {
      "epoch": 0.22701592193117617,
      "grad_norm": 3.1133334636688232,
      "learning_rate": 3.527397260273973e-06,
      "loss": 1.8277,
      "step": 221
    },
    {
      "epoch": 0.2280431432973806,
      "grad_norm": 3.289062261581421,
      "learning_rate": 3.5445205479452056e-06,
      "loss": 1.7893,
      "step": 222
    },
    {
      "epoch": 0.229070364663585,
      "grad_norm": 2.814558267593384,
      "learning_rate": 3.5616438356164386e-06,
      "loss": 1.8095,
      "step": 223
    },
    {
      "epoch": 0.2300975860297894,
      "grad_norm": 3.48117733001709,
      "learning_rate": 3.5787671232876712e-06,
      "loss": 1.8701,
      "step": 224
    },
    {
      "epoch": 0.23112480739599384,
      "grad_norm": 2.922144651412964,
      "learning_rate": 3.5958904109589043e-06,
      "loss": 1.8479,
      "step": 225
    },
    {
      "epoch": 0.23215202876219826,
      "grad_norm": 3.552428960800171,
      "learning_rate": 3.6130136986301373e-06,
      "loss": 1.7822,
      "step": 226
    },
    {
      "epoch": 0.23317925012840268,
      "grad_norm": 3.058274269104004,
      "learning_rate": 3.6301369863013704e-06,
      "loss": 1.8228,
      "step": 227
    },
    {
      "epoch": 0.23420647149460708,
      "grad_norm": 3.365823268890381,
      "learning_rate": 3.647260273972603e-06,
      "loss": 1.8415,
      "step": 228
    },
    {
      "epoch": 0.2352336928608115,
      "grad_norm": 2.87333607673645,
      "learning_rate": 3.664383561643836e-06,
      "loss": 1.8474,
      "step": 229
    },
    {
      "epoch": 0.23626091422701592,
      "grad_norm": 3.185807228088379,
      "learning_rate": 3.6815068493150687e-06,
      "loss": 1.7854,
      "step": 230
    },
    {
      "epoch": 0.23626091422701592,
      "eval_loss": 2.0132367610931396,
      "eval_runtime": 37.3246,
      "eval_samples_per_second": 26.792,
      "eval_steps_per_second": 1.125,
      "step": 230
    },
    {
      "epoch": 0.23728813559322035,
      "grad_norm": 3.02864933013916,
      "learning_rate": 3.6986301369863014e-06,
      "loss": 1.8081,
      "step": 231
    },
    {
      "epoch": 0.23831535695942477,
      "grad_norm": 2.892451286315918,
      "learning_rate": 3.7157534246575344e-06,
      "loss": 1.7619,
      "step": 232
    },
    {
      "epoch": 0.23934257832562916,
      "grad_norm": 3.24947452545166,
      "learning_rate": 3.7328767123287675e-06,
      "loss": 1.8729,
      "step": 233
    },
    {
      "epoch": 0.24036979969183359,
      "grad_norm": 3.13212513923645,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 1.7849,
      "step": 234
    },
    {
      "epoch": 0.241397021058038,
      "grad_norm": 3.004814863204956,
      "learning_rate": 3.767123287671233e-06,
      "loss": 1.8217,
      "step": 235
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 2.845555067062378,
      "learning_rate": 3.784246575342466e-06,
      "loss": 1.8438,
      "step": 236
    },
    {
      "epoch": 0.24345146379044685,
      "grad_norm": 3.1796751022338867,
      "learning_rate": 3.8013698630136993e-06,
      "loss": 1.7807,
      "step": 237
    },
    {
      "epoch": 0.24447868515665125,
      "grad_norm": 2.7663002014160156,
      "learning_rate": 3.818493150684932e-06,
      "loss": 1.7981,
      "step": 238
    },
    {
      "epoch": 0.24550590652285567,
      "grad_norm": 3.0547749996185303,
      "learning_rate": 3.8356164383561645e-06,
      "loss": 1.7543,
      "step": 239
    },
    {
      "epoch": 0.2465331278890601,
      "grad_norm": 2.6994550228118896,
      "learning_rate": 3.852739726027397e-06,
      "loss": 1.8612,
      "step": 240
    },
    {
      "epoch": 0.2465331278890601,
      "eval_loss": 2.0076475143432617,
      "eval_runtime": 37.3223,
      "eval_samples_per_second": 26.794,
      "eval_steps_per_second": 1.125,
      "step": 240
    },
    {
      "epoch": 0.24756034925526452,
      "grad_norm": 2.769717216491699,
      "learning_rate": 3.869863013698631e-06,
      "loss": 1.8352,
      "step": 241
    },
    {
      "epoch": 0.24858757062146894,
      "grad_norm": 2.7526769638061523,
      "learning_rate": 3.886986301369863e-06,
      "loss": 1.7775,
      "step": 242
    },
    {
      "epoch": 0.24961479198767333,
      "grad_norm": 3.0213067531585693,
      "learning_rate": 3.904109589041096e-06,
      "loss": 1.7901,
      "step": 243
    },
    {
      "epoch": 0.25064201335387776,
      "grad_norm": 2.8562498092651367,
      "learning_rate": 3.921232876712329e-06,
      "loss": 1.7483,
      "step": 244
    },
    {
      "epoch": 0.2516692347200822,
      "grad_norm": 2.812774896621704,
      "learning_rate": 3.938356164383562e-06,
      "loss": 1.7643,
      "step": 245
    },
    {
      "epoch": 0.2526964560862866,
      "grad_norm": 3.2509777545928955,
      "learning_rate": 3.9554794520547955e-06,
      "loss": 1.729,
      "step": 246
    },
    {
      "epoch": 0.253723677452491,
      "grad_norm": 3.6394171714782715,
      "learning_rate": 3.972602739726027e-06,
      "loss": 1.8253,
      "step": 247
    },
    {
      "epoch": 0.25475089881869545,
      "grad_norm": 3.1234781742095947,
      "learning_rate": 3.989726027397261e-06,
      "loss": 1.7457,
      "step": 248
    },
    {
      "epoch": 0.25577812018489987,
      "grad_norm": 2.8587522506713867,
      "learning_rate": 4.006849315068493e-06,
      "loss": 1.804,
      "step": 249
    },
    {
      "epoch": 0.25680534155110424,
      "grad_norm": 2.9718286991119385,
      "learning_rate": 4.023972602739726e-06,
      "loss": 1.8051,
      "step": 250
    },
    {
      "epoch": 0.25680534155110424,
      "eval_loss": 2.0024874210357666,
      "eval_runtime": 37.2884,
      "eval_samples_per_second": 26.818,
      "eval_steps_per_second": 1.126,
      "step": 250
    },
    {
      "epoch": 0.25783256291730866,
      "grad_norm": 3.007124423980713,
      "learning_rate": 4.0410958904109595e-06,
      "loss": 1.7946,
      "step": 251
    },
    {
      "epoch": 0.2588597842835131,
      "grad_norm": 3.1715426445007324,
      "learning_rate": 4.058219178082192e-06,
      "loss": 1.7811,
      "step": 252
    },
    {
      "epoch": 0.2598870056497175,
      "grad_norm": 2.814269542694092,
      "learning_rate": 4.075342465753426e-06,
      "loss": 1.8024,
      "step": 253
    },
    {
      "epoch": 0.26091422701592193,
      "grad_norm": 2.841212034225464,
      "learning_rate": 4.092465753424658e-06,
      "loss": 1.7713,
      "step": 254
    },
    {
      "epoch": 0.26194144838212635,
      "grad_norm": 3.0878522396087646,
      "learning_rate": 4.109589041095891e-06,
      "loss": 1.7579,
      "step": 255
    },
    {
      "epoch": 0.2629686697483308,
      "grad_norm": 2.946709394454956,
      "learning_rate": 4.1267123287671235e-06,
      "loss": 1.823,
      "step": 256
    },
    {
      "epoch": 0.2639958911145352,
      "grad_norm": 2.8136606216430664,
      "learning_rate": 4.143835616438356e-06,
      "loss": 1.7731,
      "step": 257
    },
    {
      "epoch": 0.2650231124807396,
      "grad_norm": 2.745762825012207,
      "learning_rate": 4.16095890410959e-06,
      "loss": 1.8445,
      "step": 258
    },
    {
      "epoch": 0.26605033384694404,
      "grad_norm": 3.0986413955688477,
      "learning_rate": 4.178082191780822e-06,
      "loss": 1.8231,
      "step": 259
    },
    {
      "epoch": 0.2670775552131484,
      "grad_norm": 2.6990394592285156,
      "learning_rate": 4.195205479452055e-06,
      "loss": 1.7635,
      "step": 260
    },
    {
      "epoch": 0.2670775552131484,
      "eval_loss": 1.9839800596237183,
      "eval_runtime": 37.1612,
      "eval_samples_per_second": 26.91,
      "eval_steps_per_second": 1.13,
      "step": 260
    },
    {
      "epoch": 0.26810477657935283,
      "grad_norm": 3.281421184539795,
      "learning_rate": 4.212328767123288e-06,
      "loss": 1.7587,
      "step": 261
    },
    {
      "epoch": 0.26913199794555726,
      "grad_norm": 3.1233623027801514,
      "learning_rate": 4.229452054794521e-06,
      "loss": 1.7673,
      "step": 262
    },
    {
      "epoch": 0.2701592193117617,
      "grad_norm": 3.3131625652313232,
      "learning_rate": 4.246575342465754e-06,
      "loss": 1.7755,
      "step": 263
    },
    {
      "epoch": 0.2711864406779661,
      "grad_norm": 3.3002350330352783,
      "learning_rate": 4.263698630136986e-06,
      "loss": 1.7575,
      "step": 264
    },
    {
      "epoch": 0.2722136620441705,
      "grad_norm": 2.965397357940674,
      "learning_rate": 4.28082191780822e-06,
      "loss": 1.7162,
      "step": 265
    },
    {
      "epoch": 0.27324088341037495,
      "grad_norm": 2.9904608726501465,
      "learning_rate": 4.297945205479452e-06,
      "loss": 1.8794,
      "step": 266
    },
    {
      "epoch": 0.27426810477657937,
      "grad_norm": 2.707996129989624,
      "learning_rate": 4.315068493150685e-06,
      "loss": 1.7646,
      "step": 267
    },
    {
      "epoch": 0.2752953261427838,
      "grad_norm": 2.8878190517425537,
      "learning_rate": 4.3321917808219185e-06,
      "loss": 1.7559,
      "step": 268
    },
    {
      "epoch": 0.2763225475089882,
      "grad_norm": 3.1676602363586426,
      "learning_rate": 4.349315068493151e-06,
      "loss": 1.7806,
      "step": 269
    },
    {
      "epoch": 0.2773497688751926,
      "grad_norm": 2.9179935455322266,
      "learning_rate": 4.3664383561643846e-06,
      "loss": 1.7493,
      "step": 270
    },
    {
      "epoch": 0.2773497688751926,
      "eval_loss": 1.9736579656600952,
      "eval_runtime": 37.0983,
      "eval_samples_per_second": 26.955,
      "eval_steps_per_second": 1.132,
      "step": 270
    },
    {
      "epoch": 0.278376990241397,
      "grad_norm": 3.0081794261932373,
      "learning_rate": 4.383561643835616e-06,
      "loss": 1.7405,
      "step": 271
    },
    {
      "epoch": 0.2794042116076014,
      "grad_norm": 3.240462064743042,
      "learning_rate": 4.40068493150685e-06,
      "loss": 1.8326,
      "step": 272
    },
    {
      "epoch": 0.28043143297380585,
      "grad_norm": 3.324059247970581,
      "learning_rate": 4.4178082191780825e-06,
      "loss": 1.7819,
      "step": 273
    },
    {
      "epoch": 0.2814586543400103,
      "grad_norm": 2.8438684940338135,
      "learning_rate": 4.434931506849315e-06,
      "loss": 1.7435,
      "step": 274
    },
    {
      "epoch": 0.2824858757062147,
      "grad_norm": 2.9525997638702393,
      "learning_rate": 4.4520547945205486e-06,
      "loss": 1.7123,
      "step": 275
    },
    {
      "epoch": 0.2835130970724191,
      "grad_norm": 2.972714900970459,
      "learning_rate": 4.469178082191781e-06,
      "loss": 1.7693,
      "step": 276
    },
    {
      "epoch": 0.28454031843862354,
      "grad_norm": 3.4248580932617188,
      "learning_rate": 4.486301369863014e-06,
      "loss": 1.7502,
      "step": 277
    },
    {
      "epoch": 0.28556753980482796,
      "grad_norm": 2.9928481578826904,
      "learning_rate": 4.503424657534247e-06,
      "loss": 1.8064,
      "step": 278
    },
    {
      "epoch": 0.28659476117103233,
      "grad_norm": 2.944093942642212,
      "learning_rate": 4.52054794520548e-06,
      "loss": 1.7075,
      "step": 279
    },
    {
      "epoch": 0.28762198253723675,
      "grad_norm": 3.2028305530548096,
      "learning_rate": 4.537671232876713e-06,
      "loss": 1.7366,
      "step": 280
    },
    {
      "epoch": 0.28762198253723675,
      "eval_loss": 1.9768972396850586,
      "eval_runtime": 37.1078,
      "eval_samples_per_second": 26.948,
      "eval_steps_per_second": 1.132,
      "step": 280
    },
    {
      "epoch": 0.2886492039034412,
      "grad_norm": 2.8642148971557617,
      "learning_rate": 4.554794520547945e-06,
      "loss": 1.7909,
      "step": 281
    },
    {
      "epoch": 0.2896764252696456,
      "grad_norm": 3.2716004848480225,
      "learning_rate": 4.571917808219179e-06,
      "loss": 1.7906,
      "step": 282
    },
    {
      "epoch": 0.29070364663585,
      "grad_norm": 2.9601588249206543,
      "learning_rate": 4.589041095890411e-06,
      "loss": 1.7731,
      "step": 283
    },
    {
      "epoch": 0.29173086800205444,
      "grad_norm": 3.1850085258483887,
      "learning_rate": 4.606164383561644e-06,
      "loss": 1.7335,
      "step": 284
    },
    {
      "epoch": 0.29275808936825887,
      "grad_norm": 3.128908634185791,
      "learning_rate": 4.6232876712328774e-06,
      "loss": 1.748,
      "step": 285
    },
    {
      "epoch": 0.2937853107344633,
      "grad_norm": 3.063014507293701,
      "learning_rate": 4.64041095890411e-06,
      "loss": 1.7632,
      "step": 286
    },
    {
      "epoch": 0.2948125321006677,
      "grad_norm": 3.467817783355713,
      "learning_rate": 4.657534246575343e-06,
      "loss": 1.7295,
      "step": 287
    },
    {
      "epoch": 0.29583975346687214,
      "grad_norm": 2.7431066036224365,
      "learning_rate": 4.674657534246575e-06,
      "loss": 1.7728,
      "step": 288
    },
    {
      "epoch": 0.2968669748330765,
      "grad_norm": 3.271421432495117,
      "learning_rate": 4.691780821917809e-06,
      "loss": 1.7403,
      "step": 289
    },
    {
      "epoch": 0.2978941961992809,
      "grad_norm": 2.908351421356201,
      "learning_rate": 4.7089041095890414e-06,
      "loss": 1.7983,
      "step": 290
    },
    {
      "epoch": 0.2978941961992809,
      "eval_loss": 1.955810308456421,
      "eval_runtime": 37.0914,
      "eval_samples_per_second": 26.96,
      "eval_steps_per_second": 1.132,
      "step": 290
    },
    {
      "epoch": 0.29892141756548535,
      "grad_norm": 3.046114683151245,
      "learning_rate": 4.726027397260274e-06,
      "loss": 1.7612,
      "step": 291
    },
    {
      "epoch": 0.29994863893168977,
      "grad_norm": 3.275486707687378,
      "learning_rate": 4.7431506849315075e-06,
      "loss": 1.7548,
      "step": 292
    },
    {
      "epoch": 0.3009758602978942,
      "grad_norm": 2.819552183151245,
      "learning_rate": 4.76027397260274e-06,
      "loss": 1.7645,
      "step": 293
    },
    {
      "epoch": 0.3020030816640986,
      "grad_norm": 3.234227418899536,
      "learning_rate": 4.777397260273973e-06,
      "loss": 1.715,
      "step": 294
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 2.988713502883911,
      "learning_rate": 4.7945205479452054e-06,
      "loss": 1.7343,
      "step": 295
    },
    {
      "epoch": 0.30405752439650746,
      "grad_norm": 3.195387601852417,
      "learning_rate": 4.811643835616439e-06,
      "loss": 1.7421,
      "step": 296
    },
    {
      "epoch": 0.3050847457627119,
      "grad_norm": 3.617765426635742,
      "learning_rate": 4.8287671232876716e-06,
      "loss": 1.6661,
      "step": 297
    },
    {
      "epoch": 0.3061119671289163,
      "grad_norm": 2.741631269454956,
      "learning_rate": 4.845890410958904e-06,
      "loss": 1.7188,
      "step": 298
    },
    {
      "epoch": 0.3071391884951207,
      "grad_norm": 3.3230361938476562,
      "learning_rate": 4.863013698630138e-06,
      "loss": 1.721,
      "step": 299
    },
    {
      "epoch": 0.3081664098613251,
      "grad_norm": 2.696352005004883,
      "learning_rate": 4.88013698630137e-06,
      "loss": 1.7137,
      "step": 300
    },
    {
      "epoch": 0.3081664098613251,
      "eval_loss": 1.9677459001541138,
      "eval_runtime": 37.1101,
      "eval_samples_per_second": 26.947,
      "eval_steps_per_second": 1.132,
      "step": 300
    },
    {
      "epoch": 0.3091936312275295,
      "grad_norm": 3.162966251373291,
      "learning_rate": 4.897260273972603e-06,
      "loss": 1.6956,
      "step": 301
    },
    {
      "epoch": 0.31022085259373394,
      "grad_norm": 2.6055824756622314,
      "learning_rate": 4.914383561643836e-06,
      "loss": 1.8174,
      "step": 302
    },
    {
      "epoch": 0.31124807395993837,
      "grad_norm": 2.816882610321045,
      "learning_rate": 4.931506849315069e-06,
      "loss": 1.7444,
      "step": 303
    },
    {
      "epoch": 0.3122752953261428,
      "grad_norm": 2.9913320541381836,
      "learning_rate": 4.948630136986302e-06,
      "loss": 1.7744,
      "step": 304
    },
    {
      "epoch": 0.3133025166923472,
      "grad_norm": 2.992614507675171,
      "learning_rate": 4.965753424657534e-06,
      "loss": 1.8155,
      "step": 305
    },
    {
      "epoch": 0.31432973805855163,
      "grad_norm": 3.2482688426971436,
      "learning_rate": 4.982876712328768e-06,
      "loss": 1.6903,
      "step": 306
    },
    {
      "epoch": 0.31535695942475606,
      "grad_norm": 2.7701141834259033,
      "learning_rate": 5e-06,
      "loss": 1.7228,
      "step": 307
    },
    {
      "epoch": 0.3163841807909605,
      "grad_norm": 3.179245948791504,
      "learning_rate": 4.998096688237534e-06,
      "loss": 1.6644,
      "step": 308
    },
    {
      "epoch": 0.31741140215716485,
      "grad_norm": 2.845163106918335,
      "learning_rate": 4.996193376475067e-06,
      "loss": 1.772,
      "step": 309
    },
    {
      "epoch": 0.31843862352336927,
      "grad_norm": 3.1509721279144287,
      "learning_rate": 4.9942900647126005e-06,
      "loss": 1.6662,
      "step": 310
    },
    {
      "epoch": 0.31843862352336927,
      "eval_loss": 1.9684854745864868,
      "eval_runtime": 37.0507,
      "eval_samples_per_second": 26.99,
      "eval_steps_per_second": 1.134,
      "step": 310
    },
    {
      "epoch": 0.3194658448895737,
      "grad_norm": 2.8594536781311035,
      "learning_rate": 4.992386752950134e-06,
      "loss": 1.6587,
      "step": 311
    },
    {
      "epoch": 0.3204930662557781,
      "grad_norm": 3.358464002609253,
      "learning_rate": 4.990483441187667e-06,
      "loss": 1.7021,
      "step": 312
    },
    {
      "epoch": 0.32152028762198254,
      "grad_norm": 2.6954872608184814,
      "learning_rate": 4.9885801294252e-06,
      "loss": 1.7301,
      "step": 313
    },
    {
      "epoch": 0.32254750898818696,
      "grad_norm": 2.9256083965301514,
      "learning_rate": 4.986676817662733e-06,
      "loss": 1.7316,
      "step": 314
    },
    {
      "epoch": 0.3235747303543914,
      "grad_norm": 2.840461254119873,
      "learning_rate": 4.9847735059002665e-06,
      "loss": 1.6848,
      "step": 315
    },
    {
      "epoch": 0.3246019517205958,
      "grad_norm": 2.9138107299804688,
      "learning_rate": 4.9828701941378e-06,
      "loss": 1.765,
      "step": 316
    },
    {
      "epoch": 0.3256291730868002,
      "grad_norm": 2.858368396759033,
      "learning_rate": 4.980966882375333e-06,
      "loss": 1.7078,
      "step": 317
    },
    {
      "epoch": 0.3266563944530046,
      "grad_norm": 2.905238628387451,
      "learning_rate": 4.979063570612867e-06,
      "loss": 1.6765,
      "step": 318
    },
    {
      "epoch": 0.327683615819209,
      "grad_norm": 3.117845296859741,
      "learning_rate": 4.9771602588504e-06,
      "loss": 1.6419,
      "step": 319
    },
    {
      "epoch": 0.32871083718541344,
      "grad_norm": 2.5665884017944336,
      "learning_rate": 4.975256947087933e-06,
      "loss": 1.7424,
      "step": 320
    },
    {
      "epoch": 0.32871083718541344,
      "eval_loss": 1.9528932571411133,
      "eval_runtime": 37.2179,
      "eval_samples_per_second": 26.869,
      "eval_steps_per_second": 1.128,
      "step": 320
    },
    {
      "epoch": 0.32973805855161786,
      "grad_norm": 2.499392032623291,
      "learning_rate": 4.973353635325467e-06,
      "loss": 1.6544,
      "step": 321
    },
    {
      "epoch": 0.3307652799178223,
      "grad_norm": 2.8056774139404297,
      "learning_rate": 4.971450323563e-06,
      "loss": 1.7085,
      "step": 322
    },
    {
      "epoch": 0.3317925012840267,
      "grad_norm": 2.850602388381958,
      "learning_rate": 4.969547011800533e-06,
      "loss": 1.6489,
      "step": 323
    },
    {
      "epoch": 0.33281972265023113,
      "grad_norm": 2.783855438232422,
      "learning_rate": 4.967643700038067e-06,
      "loss": 1.7153,
      "step": 324
    },
    {
      "epoch": 0.33384694401643555,
      "grad_norm": 2.7178030014038086,
      "learning_rate": 4.9657403882756e-06,
      "loss": 1.7104,
      "step": 325
    },
    {
      "epoch": 0.33487416538264,
      "grad_norm": 2.7497682571411133,
      "learning_rate": 4.9638370765131335e-06,
      "loss": 1.6778,
      "step": 326
    },
    {
      "epoch": 0.3359013867488444,
      "grad_norm": 2.688120126724243,
      "learning_rate": 4.961933764750667e-06,
      "loss": 1.6673,
      "step": 327
    },
    {
      "epoch": 0.33692860811504877,
      "grad_norm": 2.661578416824341,
      "learning_rate": 4.9600304529882e-06,
      "loss": 1.7353,
      "step": 328
    },
    {
      "epoch": 0.3379558294812532,
      "grad_norm": 2.759965419769287,
      "learning_rate": 4.9581271412257335e-06,
      "loss": 1.7506,
      "step": 329
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 2.972900390625,
      "learning_rate": 4.956223829463266e-06,
      "loss": 1.7114,
      "step": 330
    },
    {
      "epoch": 0.3389830508474576,
      "eval_loss": 1.9369672536849976,
      "eval_runtime": 37.2876,
      "eval_samples_per_second": 26.819,
      "eval_steps_per_second": 1.126,
      "step": 330
    },
    {
      "epoch": 0.34001027221366203,
      "grad_norm": 2.7194814682006836,
      "learning_rate": 4.954320517700799e-06,
      "loss": 1.621,
      "step": 331
    },
    {
      "epoch": 0.34103749357986646,
      "grad_norm": 2.822052001953125,
      "learning_rate": 4.952417205938333e-06,
      "loss": 1.7215,
      "step": 332
    },
    {
      "epoch": 0.3420647149460709,
      "grad_norm": 2.778348207473755,
      "learning_rate": 4.950513894175866e-06,
      "loss": 1.7049,
      "step": 333
    },
    {
      "epoch": 0.3430919363122753,
      "grad_norm": 2.8513247966766357,
      "learning_rate": 4.9486105824133995e-06,
      "loss": 1.684,
      "step": 334
    },
    {
      "epoch": 0.3441191576784797,
      "grad_norm": 2.5596683025360107,
      "learning_rate": 4.946707270650933e-06,
      "loss": 1.6992,
      "step": 335
    },
    {
      "epoch": 0.34514637904468415,
      "grad_norm": 2.850728750228882,
      "learning_rate": 4.944803958888466e-06,
      "loss": 1.7095,
      "step": 336
    },
    {
      "epoch": 0.34617360041088857,
      "grad_norm": 2.8497488498687744,
      "learning_rate": 4.942900647126e-06,
      "loss": 1.7361,
      "step": 337
    },
    {
      "epoch": 0.34720082177709294,
      "grad_norm": 2.9133079051971436,
      "learning_rate": 4.940997335363533e-06,
      "loss": 1.6783,
      "step": 338
    },
    {
      "epoch": 0.34822804314329736,
      "grad_norm": 2.7611589431762695,
      "learning_rate": 4.939094023601066e-06,
      "loss": 1.6407,
      "step": 339
    },
    {
      "epoch": 0.3492552645095018,
      "grad_norm": 2.7190754413604736,
      "learning_rate": 4.9371907118386e-06,
      "loss": 1.6886,
      "step": 340
    },
    {
      "epoch": 0.3492552645095018,
      "eval_loss": 1.9296809434890747,
      "eval_runtime": 37.4339,
      "eval_samples_per_second": 26.714,
      "eval_steps_per_second": 1.122,
      "step": 340
    },
    {
      "epoch": 0.3502824858757062,
      "grad_norm": 2.863070487976074,
      "learning_rate": 4.935287400076133e-06,
      "loss": 1.6912,
      "step": 341
    },
    {
      "epoch": 0.35130970724191063,
      "grad_norm": 2.5978336334228516,
      "learning_rate": 4.933384088313666e-06,
      "loss": 1.7107,
      "step": 342
    },
    {
      "epoch": 0.35233692860811505,
      "grad_norm": 2.8323776721954346,
      "learning_rate": 4.9314807765512e-06,
      "loss": 1.6723,
      "step": 343
    },
    {
      "epoch": 0.3533641499743195,
      "grad_norm": 2.7128350734710693,
      "learning_rate": 4.929577464788733e-06,
      "loss": 1.5993,
      "step": 344
    },
    {
      "epoch": 0.3543913713405239,
      "grad_norm": 2.618936061859131,
      "learning_rate": 4.9276741530262665e-06,
      "loss": 1.6769,
      "step": 345
    },
    {
      "epoch": 0.3554185927067283,
      "grad_norm": 2.9575355052948,
      "learning_rate": 4.9257708412638e-06,
      "loss": 1.6743,
      "step": 346
    },
    {
      "epoch": 0.35644581407293274,
      "grad_norm": 2.756464719772339,
      "learning_rate": 4.923867529501332e-06,
      "loss": 1.6858,
      "step": 347
    },
    {
      "epoch": 0.3574730354391371,
      "grad_norm": 2.6581053733825684,
      "learning_rate": 4.921964217738866e-06,
      "loss": 1.6515,
      "step": 348
    },
    {
      "epoch": 0.35850025680534153,
      "grad_norm": 2.890632390975952,
      "learning_rate": 4.920060905976399e-06,
      "loss": 1.7378,
      "step": 349
    },
    {
      "epoch": 0.35952747817154596,
      "grad_norm": 2.7895255088806152,
      "learning_rate": 4.9181575942139325e-06,
      "loss": 1.7131,
      "step": 350
    },
    {
      "epoch": 0.35952747817154596,
      "eval_loss": 1.9226924180984497,
      "eval_runtime": 37.5186,
      "eval_samples_per_second": 26.653,
      "eval_steps_per_second": 1.119,
      "step": 350
    },
    {
      "epoch": 0.3605546995377504,
      "grad_norm": 2.7698557376861572,
      "learning_rate": 4.916254282451466e-06,
      "loss": 1.71,
      "step": 351
    },
    {
      "epoch": 0.3615819209039548,
      "grad_norm": 2.956592321395874,
      "learning_rate": 4.914350970688999e-06,
      "loss": 1.7487,
      "step": 352
    },
    {
      "epoch": 0.3626091422701592,
      "grad_norm": 2.6062943935394287,
      "learning_rate": 4.9124476589265325e-06,
      "loss": 1.6486,
      "step": 353
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.588418483734131,
      "learning_rate": 4.910544347164066e-06,
      "loss": 1.6468,
      "step": 354
    },
    {
      "epoch": 0.36466358500256807,
      "grad_norm": 2.4665889739990234,
      "learning_rate": 4.908641035401599e-06,
      "loss": 1.7264,
      "step": 355
    },
    {
      "epoch": 0.3656908063687725,
      "grad_norm": 2.701957941055298,
      "learning_rate": 4.906737723639133e-06,
      "loss": 1.7914,
      "step": 356
    },
    {
      "epoch": 0.3667180277349769,
      "grad_norm": 2.68169903755188,
      "learning_rate": 4.904834411876665e-06,
      "loss": 1.6609,
      "step": 357
    },
    {
      "epoch": 0.3677452491011813,
      "grad_norm": 2.633120536804199,
      "learning_rate": 4.902931100114199e-06,
      "loss": 1.6998,
      "step": 358
    },
    {
      "epoch": 0.3687724704673857,
      "grad_norm": 2.549065113067627,
      "learning_rate": 4.901027788351733e-06,
      "loss": 1.6227,
      "step": 359
    },
    {
      "epoch": 0.3697996918335901,
      "grad_norm": 2.6537365913391113,
      "learning_rate": 4.899124476589266e-06,
      "loss": 1.6678,
      "step": 360
    },
    {
      "epoch": 0.3697996918335901,
      "eval_loss": 1.9240167140960693,
      "eval_runtime": 37.3212,
      "eval_samples_per_second": 26.794,
      "eval_steps_per_second": 1.125,
      "step": 360
    },
    {
      "epoch": 0.37082691319979455,
      "grad_norm": 2.8710854053497314,
      "learning_rate": 4.8972211648267994e-06,
      "loss": 1.6377,
      "step": 361
    },
    {
      "epoch": 0.371854134565999,
      "grad_norm": 2.8168065547943115,
      "learning_rate": 4.895317853064333e-06,
      "loss": 1.6888,
      "step": 362
    },
    {
      "epoch": 0.3728813559322034,
      "grad_norm": 2.7674975395202637,
      "learning_rate": 4.893414541301866e-06,
      "loss": 1.6532,
      "step": 363
    },
    {
      "epoch": 0.3739085772984078,
      "grad_norm": 2.5604195594787598,
      "learning_rate": 4.891511229539399e-06,
      "loss": 1.6513,
      "step": 364
    },
    {
      "epoch": 0.37493579866461224,
      "grad_norm": 2.5669636726379395,
      "learning_rate": 4.889607917776932e-06,
      "loss": 1.6832,
      "step": 365
    },
    {
      "epoch": 0.37596302003081666,
      "grad_norm": 2.650104522705078,
      "learning_rate": 4.887704606014465e-06,
      "loss": 1.6836,
      "step": 366
    },
    {
      "epoch": 0.37699024139702103,
      "grad_norm": 2.702894926071167,
      "learning_rate": 4.885801294251999e-06,
      "loss": 1.6916,
      "step": 367
    },
    {
      "epoch": 0.37801746276322545,
      "grad_norm": 2.991715431213379,
      "learning_rate": 4.883897982489532e-06,
      "loss": 1.6146,
      "step": 368
    },
    {
      "epoch": 0.3790446841294299,
      "grad_norm": 2.725093126296997,
      "learning_rate": 4.8819946707270655e-06,
      "loss": 1.7124,
      "step": 369
    },
    {
      "epoch": 0.3800719054956343,
      "grad_norm": 2.8419744968414307,
      "learning_rate": 4.880091358964599e-06,
      "loss": 1.7002,
      "step": 370
    },
    {
      "epoch": 0.3800719054956343,
      "eval_loss": 1.9024126529693604,
      "eval_runtime": 37.231,
      "eval_samples_per_second": 26.859,
      "eval_steps_per_second": 1.128,
      "step": 370
    },
    {
      "epoch": 0.3810991268618387,
      "grad_norm": 2.3915929794311523,
      "learning_rate": 4.878188047202132e-06,
      "loss": 1.6096,
      "step": 371
    },
    {
      "epoch": 0.38212634822804314,
      "grad_norm": 2.577662944793701,
      "learning_rate": 4.876284735439666e-06,
      "loss": 1.772,
      "step": 372
    },
    {
      "epoch": 0.38315356959424757,
      "grad_norm": 2.6686151027679443,
      "learning_rate": 4.874381423677199e-06,
      "loss": 1.7375,
      "step": 373
    },
    {
      "epoch": 0.384180790960452,
      "grad_norm": 2.4534966945648193,
      "learning_rate": 4.8724781119147315e-06,
      "loss": 1.6603,
      "step": 374
    },
    {
      "epoch": 0.3852080123266564,
      "grad_norm": 2.7308576107025146,
      "learning_rate": 4.870574800152265e-06,
      "loss": 1.6999,
      "step": 375
    },
    {
      "epoch": 0.38623523369286084,
      "grad_norm": 2.551203727722168,
      "learning_rate": 4.868671488389798e-06,
      "loss": 1.6193,
      "step": 376
    },
    {
      "epoch": 0.3872624550590652,
      "grad_norm": 2.78924822807312,
      "learning_rate": 4.8667681766273316e-06,
      "loss": 1.6359,
      "step": 377
    },
    {
      "epoch": 0.3882896764252696,
      "grad_norm": 2.7972333431243896,
      "learning_rate": 4.864864864864866e-06,
      "loss": 1.6057,
      "step": 378
    },
    {
      "epoch": 0.38931689779147405,
      "grad_norm": 2.721069097518921,
      "learning_rate": 4.862961553102399e-06,
      "loss": 1.6661,
      "step": 379
    },
    {
      "epoch": 0.39034411915767847,
      "grad_norm": 2.817729949951172,
      "learning_rate": 4.8610582413399325e-06,
      "loss": 1.7053,
      "step": 380
    },
    {
      "epoch": 0.39034411915767847,
      "eval_loss": 1.9177663326263428,
      "eval_runtime": 37.2158,
      "eval_samples_per_second": 26.87,
      "eval_steps_per_second": 1.129,
      "step": 380
    },
    {
      "epoch": 0.3913713405238829,
      "grad_norm": 2.800873279571533,
      "learning_rate": 4.859154929577465e-06,
      "loss": 1.6584,
      "step": 381
    },
    {
      "epoch": 0.3923985618900873,
      "grad_norm": 2.6885769367218018,
      "learning_rate": 4.857251617814998e-06,
      "loss": 1.661,
      "step": 382
    },
    {
      "epoch": 0.39342578325629174,
      "grad_norm": 2.56797456741333,
      "learning_rate": 4.855348306052532e-06,
      "loss": 1.7058,
      "step": 383
    },
    {
      "epoch": 0.39445300462249616,
      "grad_norm": 2.5235650539398193,
      "learning_rate": 4.853444994290065e-06,
      "loss": 1.6977,
      "step": 384
    },
    {
      "epoch": 0.3954802259887006,
      "grad_norm": 2.713711738586426,
      "learning_rate": 4.8515416825275985e-06,
      "loss": 1.6449,
      "step": 385
    },
    {
      "epoch": 0.396507447354905,
      "grad_norm": 2.7639107704162598,
      "learning_rate": 4.849638370765132e-06,
      "loss": 1.7169,
      "step": 386
    },
    {
      "epoch": 0.3975346687211094,
      "grad_norm": 2.4775755405426025,
      "learning_rate": 4.847735059002665e-06,
      "loss": 1.6423,
      "step": 387
    },
    {
      "epoch": 0.3985618900873138,
      "grad_norm": 2.574907064437866,
      "learning_rate": 4.8458317472401985e-06,
      "loss": 1.5911,
      "step": 388
    },
    {
      "epoch": 0.3995891114535182,
      "grad_norm": 2.667907953262329,
      "learning_rate": 4.843928435477732e-06,
      "loss": 1.6883,
      "step": 389
    },
    {
      "epoch": 0.40061633281972264,
      "grad_norm": 2.592353343963623,
      "learning_rate": 4.842025123715265e-06,
      "loss": 1.639,
      "step": 390
    },
    {
      "epoch": 0.40061633281972264,
      "eval_loss": 1.9100708961486816,
      "eval_runtime": 37.2379,
      "eval_samples_per_second": 26.854,
      "eval_steps_per_second": 1.128,
      "step": 390
    },
    {
      "epoch": 0.40164355418592707,
      "grad_norm": 2.8238866329193115,
      "learning_rate": 4.840121811952798e-06,
      "loss": 1.6835,
      "step": 391
    },
    {
      "epoch": 0.4026707755521315,
      "grad_norm": 2.769272804260254,
      "learning_rate": 4.838218500190331e-06,
      "loss": 1.6651,
      "step": 392
    },
    {
      "epoch": 0.4036979969183359,
      "grad_norm": 2.749093532562256,
      "learning_rate": 4.8363151884278645e-06,
      "loss": 1.6815,
      "step": 393
    },
    {
      "epoch": 0.40472521828454033,
      "grad_norm": 2.7952616214752197,
      "learning_rate": 4.834411876665398e-06,
      "loss": 1.6622,
      "step": 394
    },
    {
      "epoch": 0.40575243965074476,
      "grad_norm": 2.5004849433898926,
      "learning_rate": 4.832508564902931e-06,
      "loss": 1.6768,
      "step": 395
    },
    {
      "epoch": 0.4067796610169492,
      "grad_norm": 2.938015937805176,
      "learning_rate": 4.830605253140465e-06,
      "loss": 1.6171,
      "step": 396
    },
    {
      "epoch": 0.40780688238315355,
      "grad_norm": 2.737666130065918,
      "learning_rate": 4.828701941377998e-06,
      "loss": 1.6572,
      "step": 397
    },
    {
      "epoch": 0.40883410374935797,
      "grad_norm": 2.6246249675750732,
      "learning_rate": 4.826798629615531e-06,
      "loss": 1.609,
      "step": 398
    },
    {
      "epoch": 0.4098613251155624,
      "grad_norm": 2.6501305103302,
      "learning_rate": 4.824895317853065e-06,
      "loss": 1.6235,
      "step": 399
    },
    {
      "epoch": 0.4108885464817668,
      "grad_norm": 2.89320707321167,
      "learning_rate": 4.822992006090598e-06,
      "loss": 1.6494,
      "step": 400
    },
    {
      "epoch": 0.4108885464817668,
      "eval_loss": 1.8916113376617432,
      "eval_runtime": 37.1849,
      "eval_samples_per_second": 26.893,
      "eval_steps_per_second": 1.129,
      "step": 400
    },
    {
      "epoch": 0.41191576784797124,
      "grad_norm": 2.4857001304626465,
      "learning_rate": 4.821088694328131e-06,
      "loss": 1.594,
      "step": 401
    },
    {
      "epoch": 0.41294298921417566,
      "grad_norm": 2.903836250305176,
      "learning_rate": 4.819185382565665e-06,
      "loss": 1.6319,
      "step": 402
    },
    {
      "epoch": 0.4139702105803801,
      "grad_norm": 2.69596266746521,
      "learning_rate": 4.817282070803198e-06,
      "loss": 1.6843,
      "step": 403
    },
    {
      "epoch": 0.4149974319465845,
      "grad_norm": 2.5105295181274414,
      "learning_rate": 4.8153787590407315e-06,
      "loss": 1.6725,
      "step": 404
    },
    {
      "epoch": 0.41602465331278893,
      "grad_norm": 2.7997353076934814,
      "learning_rate": 4.813475447278265e-06,
      "loss": 1.6017,
      "step": 405
    },
    {
      "epoch": 0.41705187467899335,
      "grad_norm": 2.6215803623199463,
      "learning_rate": 4.811572135515798e-06,
      "loss": 1.6824,
      "step": 406
    },
    {
      "epoch": 0.4180790960451977,
      "grad_norm": 2.76534104347229,
      "learning_rate": 4.809668823753332e-06,
      "loss": 1.6111,
      "step": 407
    },
    {
      "epoch": 0.41910631741140214,
      "grad_norm": 2.6977133750915527,
      "learning_rate": 4.807765511990864e-06,
      "loss": 1.6165,
      "step": 408
    },
    {
      "epoch": 0.42013353877760656,
      "grad_norm": 2.6771748065948486,
      "learning_rate": 4.8058622002283975e-06,
      "loss": 1.648,
      "step": 409
    },
    {
      "epoch": 0.421160760143811,
      "grad_norm": 2.746731996536255,
      "learning_rate": 4.803958888465931e-06,
      "loss": 1.6946,
      "step": 410
    },
    {
      "epoch": 0.421160760143811,
      "eval_loss": 1.878269076347351,
      "eval_runtime": 37.0935,
      "eval_samples_per_second": 26.959,
      "eval_steps_per_second": 1.132,
      "step": 410
    },
    {
      "epoch": 0.4221879815100154,
      "grad_norm": 2.688074827194214,
      "learning_rate": 4.802055576703464e-06,
      "loss": 1.6545,
      "step": 411
    },
    {
      "epoch": 0.42321520287621983,
      "grad_norm": 2.7336084842681885,
      "learning_rate": 4.8001522649409976e-06,
      "loss": 1.7305,
      "step": 412
    },
    {
      "epoch": 0.42424242424242425,
      "grad_norm": 2.6902427673339844,
      "learning_rate": 4.798248953178531e-06,
      "loss": 1.612,
      "step": 413
    },
    {
      "epoch": 0.4252696456086287,
      "grad_norm": 2.5339174270629883,
      "learning_rate": 4.796345641416064e-06,
      "loss": 1.5947,
      "step": 414
    },
    {
      "epoch": 0.4262968669748331,
      "grad_norm": 2.6804542541503906,
      "learning_rate": 4.794442329653598e-06,
      "loss": 1.634,
      "step": 415
    },
    {
      "epoch": 0.42732408834103747,
      "grad_norm": 2.5764853954315186,
      "learning_rate": 4.792539017891131e-06,
      "loss": 1.6947,
      "step": 416
    },
    {
      "epoch": 0.4283513097072419,
      "grad_norm": 2.4067094326019287,
      "learning_rate": 4.790635706128664e-06,
      "loss": 1.616,
      "step": 417
    },
    {
      "epoch": 0.4293785310734463,
      "grad_norm": 2.576206684112549,
      "learning_rate": 4.788732394366197e-06,
      "loss": 1.6529,
      "step": 418
    },
    {
      "epoch": 0.43040575243965074,
      "grad_norm": 2.57826828956604,
      "learning_rate": 4.786829082603731e-06,
      "loss": 1.5771,
      "step": 419
    },
    {
      "epoch": 0.43143297380585516,
      "grad_norm": 2.5002858638763428,
      "learning_rate": 4.7849257708412645e-06,
      "loss": 1.5814,
      "step": 420
    },
    {
      "epoch": 0.43143297380585516,
      "eval_loss": 1.8940727710723877,
      "eval_runtime": 37.0932,
      "eval_samples_per_second": 26.959,
      "eval_steps_per_second": 1.132,
      "step": 420
    },
    {
      "epoch": 0.4324601951720596,
      "grad_norm": 2.618067502975464,
      "learning_rate": 4.783022459078798e-06,
      "loss": 1.6645,
      "step": 421
    },
    {
      "epoch": 0.433487416538264,
      "grad_norm": 2.5745837688446045,
      "learning_rate": 4.781119147316331e-06,
      "loss": 1.602,
      "step": 422
    },
    {
      "epoch": 0.4345146379044684,
      "grad_norm": 2.8009719848632812,
      "learning_rate": 4.7792158355538645e-06,
      "loss": 1.6269,
      "step": 423
    },
    {
      "epoch": 0.43554185927067285,
      "grad_norm": 2.8486008644104004,
      "learning_rate": 4.777312523791398e-06,
      "loss": 1.6402,
      "step": 424
    },
    {
      "epoch": 0.43656908063687727,
      "grad_norm": 2.5588459968566895,
      "learning_rate": 4.77540921202893e-06,
      "loss": 1.6122,
      "step": 425
    },
    {
      "epoch": 0.43759630200308164,
      "grad_norm": 3.1455588340759277,
      "learning_rate": 4.773505900266464e-06,
      "loss": 1.6542,
      "step": 426
    },
    {
      "epoch": 0.43862352336928606,
      "grad_norm": 2.6221370697021484,
      "learning_rate": 4.771602588503997e-06,
      "loss": 1.6339,
      "step": 427
    },
    {
      "epoch": 0.4396507447354905,
      "grad_norm": 3.006816864013672,
      "learning_rate": 4.7696992767415305e-06,
      "loss": 1.6381,
      "step": 428
    },
    {
      "epoch": 0.4406779661016949,
      "grad_norm": 2.7317590713500977,
      "learning_rate": 4.767795964979064e-06,
      "loss": 1.6689,
      "step": 429
    },
    {
      "epoch": 0.44170518746789933,
      "grad_norm": 2.9259109497070312,
      "learning_rate": 4.765892653216597e-06,
      "loss": 1.6793,
      "step": 430
    },
    {
      "epoch": 0.44170518746789933,
      "eval_loss": 1.8887399435043335,
      "eval_runtime": 37.1334,
      "eval_samples_per_second": 26.93,
      "eval_steps_per_second": 1.131,
      "step": 430
    },
    {
      "epoch": 0.44273240883410375,
      "grad_norm": 2.8546087741851807,
      "learning_rate": 4.763989341454131e-06,
      "loss": 1.6706,
      "step": 431
    },
    {
      "epoch": 0.4437596302003082,
      "grad_norm": 2.6085422039031982,
      "learning_rate": 4.762086029691664e-06,
      "loss": 1.5855,
      "step": 432
    },
    {
      "epoch": 0.4447868515665126,
      "grad_norm": 2.622277021408081,
      "learning_rate": 4.760182717929197e-06,
      "loss": 1.6068,
      "step": 433
    },
    {
      "epoch": 0.445814072932717,
      "grad_norm": 2.5327494144439697,
      "learning_rate": 4.758279406166731e-06,
      "loss": 1.6303,
      "step": 434
    },
    {
      "epoch": 0.44684129429892144,
      "grad_norm": 2.7626633644104004,
      "learning_rate": 4.756376094404263e-06,
      "loss": 1.5567,
      "step": 435
    },
    {
      "epoch": 0.4478685156651258,
      "grad_norm": 2.613476514816284,
      "learning_rate": 4.7544727826417966e-06,
      "loss": 1.6345,
      "step": 436
    },
    {
      "epoch": 0.44889573703133023,
      "grad_norm": 2.58027720451355,
      "learning_rate": 4.75256947087933e-06,
      "loss": 1.675,
      "step": 437
    },
    {
      "epoch": 0.44992295839753466,
      "grad_norm": 2.6709048748016357,
      "learning_rate": 4.750666159116863e-06,
      "loss": 1.6335,
      "step": 438
    },
    {
      "epoch": 0.4509501797637391,
      "grad_norm": 2.7493093013763428,
      "learning_rate": 4.7487628473543975e-06,
      "loss": 1.6061,
      "step": 439
    },
    {
      "epoch": 0.4519774011299435,
      "grad_norm": 2.725672960281372,
      "learning_rate": 4.746859535591931e-06,
      "loss": 1.5515,
      "step": 440
    },
    {
      "epoch": 0.4519774011299435,
      "eval_loss": 1.8779852390289307,
      "eval_runtime": 37.3135,
      "eval_samples_per_second": 26.8,
      "eval_steps_per_second": 1.126,
      "step": 440
    },
    {
      "epoch": 0.4530046224961479,
      "grad_norm": 2.784477472305298,
      "learning_rate": 4.744956223829464e-06,
      "loss": 1.6146,
      "step": 441
    },
    {
      "epoch": 0.45403184386235235,
      "grad_norm": 2.758119821548462,
      "learning_rate": 4.743052912066997e-06,
      "loss": 1.5945,
      "step": 442
    },
    {
      "epoch": 0.45505906522855677,
      "grad_norm": 2.9006264209747314,
      "learning_rate": 4.74114960030453e-06,
      "loss": 1.6461,
      "step": 443
    },
    {
      "epoch": 0.4560862865947612,
      "grad_norm": 2.8473031520843506,
      "learning_rate": 4.7392462885420635e-06,
      "loss": 1.6565,
      "step": 444
    },
    {
      "epoch": 0.4571135079609656,
      "grad_norm": 2.715806722640991,
      "learning_rate": 4.737342976779597e-06,
      "loss": 1.5414,
      "step": 445
    },
    {
      "epoch": 0.45814072932717,
      "grad_norm": 2.6588337421417236,
      "learning_rate": 4.73543966501713e-06,
      "loss": 1.6284,
      "step": 446
    },
    {
      "epoch": 0.4591679506933744,
      "grad_norm": 3.1152243614196777,
      "learning_rate": 4.7335363532546636e-06,
      "loss": 1.6237,
      "step": 447
    },
    {
      "epoch": 0.4601951720595788,
      "grad_norm": 2.545490264892578,
      "learning_rate": 4.731633041492197e-06,
      "loss": 1.6149,
      "step": 448
    },
    {
      "epoch": 0.46122239342578325,
      "grad_norm": 3.0324277877807617,
      "learning_rate": 4.72972972972973e-06,
      "loss": 1.6277,
      "step": 449
    },
    {
      "epoch": 0.4622496147919877,
      "grad_norm": 2.885970115661621,
      "learning_rate": 4.727826417967264e-06,
      "loss": 1.6625,
      "step": 450
    },
    {
      "epoch": 0.4622496147919877,
      "eval_loss": 1.8651624917984009,
      "eval_runtime": 37.3486,
      "eval_samples_per_second": 26.775,
      "eval_steps_per_second": 1.125,
      "step": 450
    },
    {
      "epoch": 0.4632768361581921,
      "grad_norm": 2.7258739471435547,
      "learning_rate": 4.725923106204797e-06,
      "loss": 1.6263,
      "step": 451
    },
    {
      "epoch": 0.4643040575243965,
      "grad_norm": 2.631122589111328,
      "learning_rate": 4.72401979444233e-06,
      "loss": 1.6081,
      "step": 452
    },
    {
      "epoch": 0.46533127889060094,
      "grad_norm": 2.5882441997528076,
      "learning_rate": 4.722116482679863e-06,
      "loss": 1.6237,
      "step": 453
    },
    {
      "epoch": 0.46635850025680536,
      "grad_norm": 2.66310715675354,
      "learning_rate": 4.720213170917396e-06,
      "loss": 1.6306,
      "step": 454
    },
    {
      "epoch": 0.46738572162300973,
      "grad_norm": 2.6906657218933105,
      "learning_rate": 4.71830985915493e-06,
      "loss": 1.6492,
      "step": 455
    },
    {
      "epoch": 0.46841294298921415,
      "grad_norm": 2.6957409381866455,
      "learning_rate": 4.716406547392463e-06,
      "loss": 1.5993,
      "step": 456
    },
    {
      "epoch": 0.4694401643554186,
      "grad_norm": 2.904311418533325,
      "learning_rate": 4.714503235629996e-06,
      "loss": 1.6633,
      "step": 457
    },
    {
      "epoch": 0.470467385721623,
      "grad_norm": 2.6853256225585938,
      "learning_rate": 4.71259992386753e-06,
      "loss": 1.5824,
      "step": 458
    },
    {
      "epoch": 0.4714946070878274,
      "grad_norm": 2.4767656326293945,
      "learning_rate": 4.710696612105063e-06,
      "loss": 1.6145,
      "step": 459
    },
    {
      "epoch": 0.47252182845403184,
      "grad_norm": 2.5569417476654053,
      "learning_rate": 4.708793300342596e-06,
      "loss": 1.6116,
      "step": 460
    },
    {
      "epoch": 0.47252182845403184,
      "eval_loss": 1.8785003423690796,
      "eval_runtime": 37.401,
      "eval_samples_per_second": 26.737,
      "eval_steps_per_second": 1.123,
      "step": 460
    },
    {
      "epoch": 0.47354904982023627,
      "grad_norm": 2.6568074226379395,
      "learning_rate": 4.70688998858013e-06,
      "loss": 1.6097,
      "step": 461
    },
    {
      "epoch": 0.4745762711864407,
      "grad_norm": 2.6002280712127686,
      "learning_rate": 4.704986676817663e-06,
      "loss": 1.6025,
      "step": 462
    },
    {
      "epoch": 0.4756034925526451,
      "grad_norm": 2.6055355072021484,
      "learning_rate": 4.7030833650551965e-06,
      "loss": 1.6322,
      "step": 463
    },
    {
      "epoch": 0.47663071391884954,
      "grad_norm": 2.6288487911224365,
      "learning_rate": 4.70118005329273e-06,
      "loss": 1.594,
      "step": 464
    },
    {
      "epoch": 0.4776579352850539,
      "grad_norm": 2.6981899738311768,
      "learning_rate": 4.699276741530263e-06,
      "loss": 1.6707,
      "step": 465
    },
    {
      "epoch": 0.4786851566512583,
      "grad_norm": 2.6832222938537598,
      "learning_rate": 4.697373429767797e-06,
      "loss": 1.6435,
      "step": 466
    },
    {
      "epoch": 0.47971237801746275,
      "grad_norm": 2.563159227371216,
      "learning_rate": 4.69547011800533e-06,
      "loss": 1.6249,
      "step": 467
    },
    {
      "epoch": 0.48073959938366717,
      "grad_norm": 2.545778751373291,
      "learning_rate": 4.693566806242863e-06,
      "loss": 1.6309,
      "step": 468
    },
    {
      "epoch": 0.4817668207498716,
      "grad_norm": 2.555227518081665,
      "learning_rate": 4.691663494480397e-06,
      "loss": 1.6128,
      "step": 469
    },
    {
      "epoch": 0.482794042116076,
      "grad_norm": 2.612576723098755,
      "learning_rate": 4.689760182717929e-06,
      "loss": 1.6484,
      "step": 470
    },
    {
      "epoch": 0.482794042116076,
      "eval_loss": 1.8684974908828735,
      "eval_runtime": 37.2607,
      "eval_samples_per_second": 26.838,
      "eval_steps_per_second": 1.127,
      "step": 470
    },
    {
      "epoch": 0.48382126348228044,
      "grad_norm": 2.4981396198272705,
      "learning_rate": 4.6878568709554626e-06,
      "loss": 1.5393,
      "step": 471
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 2.793700695037842,
      "learning_rate": 4.685953559192996e-06,
      "loss": 1.608,
      "step": 472
    },
    {
      "epoch": 0.4858757062146893,
      "grad_norm": 2.7605607509613037,
      "learning_rate": 4.684050247430529e-06,
      "loss": 1.6236,
      "step": 473
    },
    {
      "epoch": 0.4869029275808937,
      "grad_norm": 2.413719892501831,
      "learning_rate": 4.682146935668063e-06,
      "loss": 1.5555,
      "step": 474
    },
    {
      "epoch": 0.4879301489470981,
      "grad_norm": 2.9032249450683594,
      "learning_rate": 4.680243623905596e-06,
      "loss": 1.6042,
      "step": 475
    },
    {
      "epoch": 0.4889573703133025,
      "grad_norm": 2.5445175170898438,
      "learning_rate": 4.678340312143129e-06,
      "loss": 1.6265,
      "step": 476
    },
    {
      "epoch": 0.4899845916795069,
      "grad_norm": 2.6955716609954834,
      "learning_rate": 4.676437000380663e-06,
      "loss": 1.5879,
      "step": 477
    },
    {
      "epoch": 0.49101181304571134,
      "grad_norm": 2.8399500846862793,
      "learning_rate": 4.674533688618196e-06,
      "loss": 1.7009,
      "step": 478
    },
    {
      "epoch": 0.49203903441191577,
      "grad_norm": 2.635035991668701,
      "learning_rate": 4.6726303768557295e-06,
      "loss": 1.6381,
      "step": 479
    },
    {
      "epoch": 0.4930662557781202,
      "grad_norm": 2.7609052658081055,
      "learning_rate": 4.670727065093263e-06,
      "loss": 1.5492,
      "step": 480
    },
    {
      "epoch": 0.4930662557781202,
      "eval_loss": 1.8574503660202026,
      "eval_runtime": 37.2173,
      "eval_samples_per_second": 26.869,
      "eval_steps_per_second": 1.129,
      "step": 480
    },
    {
      "epoch": 0.4940934771443246,
      "grad_norm": 2.593766927719116,
      "learning_rate": 4.668823753330796e-06,
      "loss": 1.5693,
      "step": 481
    },
    {
      "epoch": 0.49512069851052903,
      "grad_norm": 2.8000242710113525,
      "learning_rate": 4.6669204415683295e-06,
      "loss": 1.5998,
      "step": 482
    },
    {
      "epoch": 0.49614791987673346,
      "grad_norm": 2.522688627243042,
      "learning_rate": 4.665017129805863e-06,
      "loss": 1.6653,
      "step": 483
    },
    {
      "epoch": 0.4971751412429379,
      "grad_norm": 2.854534864425659,
      "learning_rate": 4.663113818043396e-06,
      "loss": 1.6393,
      "step": 484
    },
    {
      "epoch": 0.49820236260914225,
      "grad_norm": 2.5333967208862305,
      "learning_rate": 4.66121050628093e-06,
      "loss": 1.606,
      "step": 485
    },
    {
      "epoch": 0.49922958397534667,
      "grad_norm": 2.565260171890259,
      "learning_rate": 4.659307194518463e-06,
      "loss": 1.6471,
      "step": 486
    },
    {
      "epoch": 0.5002568053415511,
      "grad_norm": 2.570220947265625,
      "learning_rate": 4.6574038827559955e-06,
      "loss": 1.6023,
      "step": 487
    },
    {
      "epoch": 0.5012840267077555,
      "grad_norm": 2.460740089416504,
      "learning_rate": 4.655500570993529e-06,
      "loss": 1.6208,
      "step": 488
    },
    {
      "epoch": 0.50231124807396,
      "grad_norm": 2.6339426040649414,
      "learning_rate": 4.653597259231062e-06,
      "loss": 1.6533,
      "step": 489
    },
    {
      "epoch": 0.5033384694401644,
      "grad_norm": 2.681044578552246,
      "learning_rate": 4.651693947468596e-06,
      "loss": 1.5437,
      "step": 490
    },
    {
      "epoch": 0.5033384694401644,
      "eval_loss": 1.8701337575912476,
      "eval_runtime": 37.1714,
      "eval_samples_per_second": 26.902,
      "eval_steps_per_second": 1.13,
      "step": 490
    },
    {
      "epoch": 0.5043656908063687,
      "grad_norm": 2.6900649070739746,
      "learning_rate": 4.649790635706129e-06,
      "loss": 1.6816,
      "step": 491
    },
    {
      "epoch": 0.5053929121725732,
      "grad_norm": 2.6106340885162354,
      "learning_rate": 4.647887323943662e-06,
      "loss": 1.5784,
      "step": 492
    },
    {
      "epoch": 0.5064201335387776,
      "grad_norm": 3.0704469680786133,
      "learning_rate": 4.645984012181196e-06,
      "loss": 1.6065,
      "step": 493
    },
    {
      "epoch": 0.507447354904982,
      "grad_norm": 2.648425817489624,
      "learning_rate": 4.644080700418729e-06,
      "loss": 1.5134,
      "step": 494
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 2.509066343307495,
      "learning_rate": 4.642177388656262e-06,
      "loss": 1.6324,
      "step": 495
    },
    {
      "epoch": 0.5095017976373909,
      "grad_norm": 2.8628857135772705,
      "learning_rate": 4.640274076893796e-06,
      "loss": 1.5957,
      "step": 496
    },
    {
      "epoch": 0.5105290190035953,
      "grad_norm": 2.83579683303833,
      "learning_rate": 4.638370765131328e-06,
      "loss": 1.6556,
      "step": 497
    },
    {
      "epoch": 0.5115562403697997,
      "grad_norm": 2.6618151664733887,
      "learning_rate": 4.636467453368862e-06,
      "loss": 1.5738,
      "step": 498
    },
    {
      "epoch": 0.5125834617360041,
      "grad_norm": 2.936943769454956,
      "learning_rate": 4.634564141606396e-06,
      "loss": 1.5597,
      "step": 499
    },
    {
      "epoch": 0.5136106831022085,
      "grad_norm": 2.8520913124084473,
      "learning_rate": 4.632660829843929e-06,
      "loss": 1.6316,
      "step": 500
    },
    {
      "epoch": 0.5136106831022085,
      "eval_loss": 1.8548446893692017,
      "eval_runtime": 37.1878,
      "eval_samples_per_second": 26.891,
      "eval_steps_per_second": 1.129,
      "step": 500
    },
    {
      "epoch": 0.514637904468413,
      "grad_norm": 2.446373462677002,
      "learning_rate": 4.630757518081463e-06,
      "loss": 1.5533,
      "step": 501
    },
    {
      "epoch": 0.5156651258346173,
      "grad_norm": 2.8339035511016846,
      "learning_rate": 4.628854206318996e-06,
      "loss": 1.6904,
      "step": 502
    },
    {
      "epoch": 0.5166923472008218,
      "grad_norm": 2.664249897003174,
      "learning_rate": 4.626950894556529e-06,
      "loss": 1.5926,
      "step": 503
    },
    {
      "epoch": 0.5177195685670262,
      "grad_norm": 2.681185483932495,
      "learning_rate": 4.625047582794062e-06,
      "loss": 1.6348,
      "step": 504
    },
    {
      "epoch": 0.5187467899332306,
      "grad_norm": 2.468263864517212,
      "learning_rate": 4.623144271031595e-06,
      "loss": 1.5477,
      "step": 505
    },
    {
      "epoch": 0.519774011299435,
      "grad_norm": 2.539066791534424,
      "learning_rate": 4.6212409592691286e-06,
      "loss": 1.5615,
      "step": 506
    },
    {
      "epoch": 0.5208012326656395,
      "grad_norm": 2.822235345840454,
      "learning_rate": 4.619337647506662e-06,
      "loss": 1.6398,
      "step": 507
    },
    {
      "epoch": 0.5218284540318439,
      "grad_norm": 2.6063523292541504,
      "learning_rate": 4.617434335744195e-06,
      "loss": 1.5992,
      "step": 508
    },
    {
      "epoch": 0.5228556753980482,
      "grad_norm": 2.624810218811035,
      "learning_rate": 4.615531023981729e-06,
      "loss": 1.5708,
      "step": 509
    },
    {
      "epoch": 0.5238828967642527,
      "grad_norm": 2.6324198246002197,
      "learning_rate": 4.613627712219262e-06,
      "loss": 1.5865,
      "step": 510
    },
    {
      "epoch": 0.5238828967642527,
      "eval_loss": 1.8442509174346924,
      "eval_runtime": 36.9899,
      "eval_samples_per_second": 27.034,
      "eval_steps_per_second": 1.135,
      "step": 510
    },
    {
      "epoch": 0.5249101181304571,
      "grad_norm": 2.5496973991394043,
      "learning_rate": 4.611724400456795e-06,
      "loss": 1.6048,
      "step": 511
    },
    {
      "epoch": 0.5259373394966615,
      "grad_norm": 2.4150168895721436,
      "learning_rate": 4.609821088694329e-06,
      "loss": 1.6009,
      "step": 512
    },
    {
      "epoch": 0.5269645608628659,
      "grad_norm": 2.8799378871917725,
      "learning_rate": 4.607917776931862e-06,
      "loss": 1.5761,
      "step": 513
    },
    {
      "epoch": 0.5279917822290704,
      "grad_norm": 2.749016761779785,
      "learning_rate": 4.606014465169395e-06,
      "loss": 1.6202,
      "step": 514
    },
    {
      "epoch": 0.5290190035952748,
      "grad_norm": 2.6993141174316406,
      "learning_rate": 4.604111153406928e-06,
      "loss": 1.632,
      "step": 515
    },
    {
      "epoch": 0.5300462249614792,
      "grad_norm": 2.9027445316314697,
      "learning_rate": 4.602207841644461e-06,
      "loss": 1.5352,
      "step": 516
    },
    {
      "epoch": 0.5310734463276836,
      "grad_norm": 2.874267816543579,
      "learning_rate": 4.600304529881995e-06,
      "loss": 1.642,
      "step": 517
    },
    {
      "epoch": 0.5321006676938881,
      "grad_norm": 2.6057357788085938,
      "learning_rate": 4.598401218119528e-06,
      "loss": 1.6179,
      "step": 518
    },
    {
      "epoch": 0.5331278890600925,
      "grad_norm": 2.658812999725342,
      "learning_rate": 4.5964979063570614e-06,
      "loss": 1.6285,
      "step": 519
    },
    {
      "epoch": 0.5341551104262968,
      "grad_norm": 2.667215585708618,
      "learning_rate": 4.594594594594596e-06,
      "loss": 1.5694,
      "step": 520
    },
    {
      "epoch": 0.5341551104262968,
      "eval_loss": 1.8490455150604248,
      "eval_runtime": 37.0914,
      "eval_samples_per_second": 26.96,
      "eval_steps_per_second": 1.132,
      "step": 520
    },
    {
      "epoch": 0.5351823317925013,
      "grad_norm": 2.5979607105255127,
      "learning_rate": 4.592691282832128e-06,
      "loss": 1.6091,
      "step": 521
    },
    {
      "epoch": 0.5362095531587057,
      "grad_norm": 2.552816152572632,
      "learning_rate": 4.5907879710696615e-06,
      "loss": 1.6145,
      "step": 522
    },
    {
      "epoch": 0.5372367745249101,
      "grad_norm": 2.575045347213745,
      "learning_rate": 4.588884659307195e-06,
      "loss": 1.5444,
      "step": 523
    },
    {
      "epoch": 0.5382639958911145,
      "grad_norm": 2.650202751159668,
      "learning_rate": 4.586981347544728e-06,
      "loss": 1.6181,
      "step": 524
    },
    {
      "epoch": 0.539291217257319,
      "grad_norm": 2.7869739532470703,
      "learning_rate": 4.585078035782262e-06,
      "loss": 1.563,
      "step": 525
    },
    {
      "epoch": 0.5403184386235234,
      "grad_norm": 2.6828651428222656,
      "learning_rate": 4.583174724019795e-06,
      "loss": 1.6004,
      "step": 526
    },
    {
      "epoch": 0.5413456599897278,
      "grad_norm": 2.552387237548828,
      "learning_rate": 4.581271412257328e-06,
      "loss": 1.5351,
      "step": 527
    },
    {
      "epoch": 0.5423728813559322,
      "grad_norm": 2.785661458969116,
      "learning_rate": 4.579368100494862e-06,
      "loss": 1.5264,
      "step": 528
    },
    {
      "epoch": 0.5434001027221366,
      "grad_norm": 2.832939863204956,
      "learning_rate": 4.577464788732395e-06,
      "loss": 1.6261,
      "step": 529
    },
    {
      "epoch": 0.544427324088341,
      "grad_norm": 2.641766309738159,
      "learning_rate": 4.575561476969928e-06,
      "loss": 1.6019,
      "step": 530
    },
    {
      "epoch": 0.544427324088341,
      "eval_loss": 1.8477164506912231,
      "eval_runtime": 37.1444,
      "eval_samples_per_second": 26.922,
      "eval_steps_per_second": 1.131,
      "step": 530
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 3.0502524375915527,
      "learning_rate": 4.573658165207461e-06,
      "loss": 1.6436,
      "step": 531
    },
    {
      "epoch": 0.5464817668207499,
      "grad_norm": 2.64772891998291,
      "learning_rate": 4.571754853444994e-06,
      "loss": 1.6085,
      "step": 532
    },
    {
      "epoch": 0.5475089881869543,
      "grad_norm": 2.549692392349243,
      "learning_rate": 4.569851541682528e-06,
      "loss": 1.5977,
      "step": 533
    },
    {
      "epoch": 0.5485362095531587,
      "grad_norm": 2.719322681427002,
      "learning_rate": 4.567948229920061e-06,
      "loss": 1.5659,
      "step": 534
    },
    {
      "epoch": 0.5495634309193631,
      "grad_norm": 2.5005698204040527,
      "learning_rate": 4.566044918157594e-06,
      "loss": 1.5301,
      "step": 535
    },
    {
      "epoch": 0.5505906522855676,
      "grad_norm": 2.7379472255706787,
      "learning_rate": 4.564141606395128e-06,
      "loss": 1.5721,
      "step": 536
    },
    {
      "epoch": 0.551617873651772,
      "grad_norm": 2.6673905849456787,
      "learning_rate": 4.562238294632661e-06,
      "loss": 1.6105,
      "step": 537
    },
    {
      "epoch": 0.5526450950179764,
      "grad_norm": 2.6745669841766357,
      "learning_rate": 4.5603349828701945e-06,
      "loss": 1.577,
      "step": 538
    },
    {
      "epoch": 0.5536723163841808,
      "grad_norm": 2.593475341796875,
      "learning_rate": 4.558431671107728e-06,
      "loss": 1.6084,
      "step": 539
    },
    {
      "epoch": 0.5546995377503852,
      "grad_norm": 2.6577041149139404,
      "learning_rate": 4.556528359345261e-06,
      "loss": 1.6417,
      "step": 540
    },
    {
      "epoch": 0.5546995377503852,
      "eval_loss": 1.839761734008789,
      "eval_runtime": 37.1765,
      "eval_samples_per_second": 26.899,
      "eval_steps_per_second": 1.13,
      "step": 540
    },
    {
      "epoch": 0.5557267591165896,
      "grad_norm": 2.6834933757781982,
      "learning_rate": 4.5546250475827946e-06,
      "loss": 1.5803,
      "step": 541
    },
    {
      "epoch": 0.556753980482794,
      "grad_norm": 2.464665174484253,
      "learning_rate": 4.552721735820328e-06,
      "loss": 1.5324,
      "step": 542
    },
    {
      "epoch": 0.5577812018489985,
      "grad_norm": 2.4488964080810547,
      "learning_rate": 4.550818424057861e-06,
      "loss": 1.5767,
      "step": 543
    },
    {
      "epoch": 0.5588084232152029,
      "grad_norm": 2.763124465942383,
      "learning_rate": 4.548915112295395e-06,
      "loss": 1.5903,
      "step": 544
    },
    {
      "epoch": 0.5598356445814073,
      "grad_norm": 2.7473502159118652,
      "learning_rate": 4.547011800532928e-06,
      "loss": 1.5966,
      "step": 545
    },
    {
      "epoch": 0.5608628659476117,
      "grad_norm": 3.050267219543457,
      "learning_rate": 4.545108488770461e-06,
      "loss": 1.5492,
      "step": 546
    },
    {
      "epoch": 0.5618900873138162,
      "grad_norm": 2.910439968109131,
      "learning_rate": 4.543205177007995e-06,
      "loss": 1.6357,
      "step": 547
    },
    {
      "epoch": 0.5629173086800205,
      "grad_norm": 2.5621109008789062,
      "learning_rate": 4.541301865245527e-06,
      "loss": 1.5693,
      "step": 548
    },
    {
      "epoch": 0.5639445300462249,
      "grad_norm": 2.7492010593414307,
      "learning_rate": 4.539398553483061e-06,
      "loss": 1.615,
      "step": 549
    },
    {
      "epoch": 0.5649717514124294,
      "grad_norm": 2.6651856899261475,
      "learning_rate": 4.537495241720594e-06,
      "loss": 1.5804,
      "step": 550
    },
    {
      "epoch": 0.5649717514124294,
      "eval_loss": 1.8344007730484009,
      "eval_runtime": 37.1218,
      "eval_samples_per_second": 26.938,
      "eval_steps_per_second": 1.131,
      "step": 550
    },
    {
      "epoch": 0.5659989727786338,
      "grad_norm": 2.5933356285095215,
      "learning_rate": 4.535591929958127e-06,
      "loss": 1.6108,
      "step": 551
    },
    {
      "epoch": 0.5670261941448382,
      "grad_norm": 2.660522222518921,
      "learning_rate": 4.533688618195661e-06,
      "loss": 1.5977,
      "step": 552
    },
    {
      "epoch": 0.5680534155110426,
      "grad_norm": 2.5636332035064697,
      "learning_rate": 4.531785306433194e-06,
      "loss": 1.5802,
      "step": 553
    },
    {
      "epoch": 0.5690806368772471,
      "grad_norm": 2.759533166885376,
      "learning_rate": 4.529881994670727e-06,
      "loss": 1.5656,
      "step": 554
    },
    {
      "epoch": 0.5701078582434514,
      "grad_norm": 2.6530680656433105,
      "learning_rate": 4.527978682908261e-06,
      "loss": 1.5693,
      "step": 555
    },
    {
      "epoch": 0.5711350796096559,
      "grad_norm": 2.688427448272705,
      "learning_rate": 4.526075371145794e-06,
      "loss": 1.5769,
      "step": 556
    },
    {
      "epoch": 0.5721623009758603,
      "grad_norm": 2.5163164138793945,
      "learning_rate": 4.5241720593833275e-06,
      "loss": 1.593,
      "step": 557
    },
    {
      "epoch": 0.5731895223420647,
      "grad_norm": 2.612612724304199,
      "learning_rate": 4.52226874762086e-06,
      "loss": 1.5854,
      "step": 558
    },
    {
      "epoch": 0.5742167437082691,
      "grad_norm": 2.647853374481201,
      "learning_rate": 4.520365435858393e-06,
      "loss": 1.6085,
      "step": 559
    },
    {
      "epoch": 0.5752439650744735,
      "grad_norm": 2.7160937786102295,
      "learning_rate": 4.518462124095928e-06,
      "loss": 1.6353,
      "step": 560
    },
    {
      "epoch": 0.5752439650744735,
      "eval_loss": 1.8585809469223022,
      "eval_runtime": 37.328,
      "eval_samples_per_second": 26.79,
      "eval_steps_per_second": 1.125,
      "step": 560
    },
    {
      "epoch": 0.576271186440678,
      "grad_norm": 2.9025261402130127,
      "learning_rate": 4.516558812333461e-06,
      "loss": 1.5903,
      "step": 561
    },
    {
      "epoch": 0.5772984078068824,
      "grad_norm": 2.5665547847747803,
      "learning_rate": 4.514655500570994e-06,
      "loss": 1.5775,
      "step": 562
    },
    {
      "epoch": 0.5783256291730868,
      "grad_norm": 2.591381311416626,
      "learning_rate": 4.512752188808528e-06,
      "loss": 1.6387,
      "step": 563
    },
    {
      "epoch": 0.5793528505392912,
      "grad_norm": 2.7540411949157715,
      "learning_rate": 4.510848877046061e-06,
      "loss": 1.6432,
      "step": 564
    },
    {
      "epoch": 0.5803800719054957,
      "grad_norm": 2.6911778450012207,
      "learning_rate": 4.5089455652835936e-06,
      "loss": 1.6078,
      "step": 565
    },
    {
      "epoch": 0.5814072932717,
      "grad_norm": 2.609619140625,
      "learning_rate": 4.507042253521127e-06,
      "loss": 1.5709,
      "step": 566
    },
    {
      "epoch": 0.5824345146379045,
      "grad_norm": 2.4930672645568848,
      "learning_rate": 4.50513894175866e-06,
      "loss": 1.5305,
      "step": 567
    },
    {
      "epoch": 0.5834617360041089,
      "grad_norm": 2.490992307662964,
      "learning_rate": 4.503235629996194e-06,
      "loss": 1.61,
      "step": 568
    },
    {
      "epoch": 0.5844889573703133,
      "grad_norm": 2.7229626178741455,
      "learning_rate": 4.501332318233727e-06,
      "loss": 1.5996,
      "step": 569
    },
    {
      "epoch": 0.5855161787365177,
      "grad_norm": 2.5289995670318604,
      "learning_rate": 4.49942900647126e-06,
      "loss": 1.5624,
      "step": 570
    },
    {
      "epoch": 0.5855161787365177,
      "eval_loss": 1.867126226425171,
      "eval_runtime": 37.4117,
      "eval_samples_per_second": 26.73,
      "eval_steps_per_second": 1.123,
      "step": 570
    },
    {
      "epoch": 0.5865434001027221,
      "grad_norm": 2.6244945526123047,
      "learning_rate": 4.497525694708794e-06,
      "loss": 1.6154,
      "step": 571
    },
    {
      "epoch": 0.5875706214689266,
      "grad_norm": 2.4343698024749756,
      "learning_rate": 4.495622382946327e-06,
      "loss": 1.5967,
      "step": 572
    },
    {
      "epoch": 0.588597842835131,
      "grad_norm": 2.5899314880371094,
      "learning_rate": 4.4937190711838605e-06,
      "loss": 1.5601,
      "step": 573
    },
    {
      "epoch": 0.5896250642013354,
      "grad_norm": 2.829028367996216,
      "learning_rate": 4.491815759421394e-06,
      "loss": 1.6315,
      "step": 574
    },
    {
      "epoch": 0.5906522855675398,
      "grad_norm": 2.748849868774414,
      "learning_rate": 4.489912447658926e-06,
      "loss": 1.5591,
      "step": 575
    },
    {
      "epoch": 0.5916795069337443,
      "grad_norm": 2.7016541957855225,
      "learning_rate": 4.48800913589646e-06,
      "loss": 1.5575,
      "step": 576
    },
    {
      "epoch": 0.5927067282999486,
      "grad_norm": 2.6260931491851807,
      "learning_rate": 4.486105824133993e-06,
      "loss": 1.5823,
      "step": 577
    },
    {
      "epoch": 0.593733949666153,
      "grad_norm": 2.797224998474121,
      "learning_rate": 4.4842025123715264e-06,
      "loss": 1.5902,
      "step": 578
    },
    {
      "epoch": 0.5947611710323575,
      "grad_norm": 2.671342611312866,
      "learning_rate": 4.48229920060906e-06,
      "loss": 1.551,
      "step": 579
    },
    {
      "epoch": 0.5957883923985619,
      "grad_norm": 2.626988172531128,
      "learning_rate": 4.480395888846594e-06,
      "loss": 1.5312,
      "step": 580
    },
    {
      "epoch": 0.5957883923985619,
      "eval_loss": 1.8685871362686157,
      "eval_runtime": 37.388,
      "eval_samples_per_second": 26.747,
      "eval_steps_per_second": 1.123,
      "step": 580
    },
    {
      "epoch": 0.5968156137647663,
      "grad_norm": 2.8422539234161377,
      "learning_rate": 4.478492577084127e-06,
      "loss": 1.6393,
      "step": 581
    },
    {
      "epoch": 0.5978428351309707,
      "grad_norm": 2.6877822875976562,
      "learning_rate": 4.47658926532166e-06,
      "loss": 1.5376,
      "step": 582
    },
    {
      "epoch": 0.5988700564971752,
      "grad_norm": 2.736452341079712,
      "learning_rate": 4.474685953559193e-06,
      "loss": 1.5798,
      "step": 583
    },
    {
      "epoch": 0.5998972778633795,
      "grad_norm": 2.732513904571533,
      "learning_rate": 4.472782641796727e-06,
      "loss": 1.6897,
      "step": 584
    },
    {
      "epoch": 0.600924499229584,
      "grad_norm": 2.8732001781463623,
      "learning_rate": 4.47087933003426e-06,
      "loss": 1.533,
      "step": 585
    },
    {
      "epoch": 0.6019517205957884,
      "grad_norm": 2.6099846363067627,
      "learning_rate": 4.468976018271793e-06,
      "loss": 1.5481,
      "step": 586
    },
    {
      "epoch": 0.6029789419619929,
      "grad_norm": 2.754856586456299,
      "learning_rate": 4.467072706509327e-06,
      "loss": 1.5497,
      "step": 587
    },
    {
      "epoch": 0.6040061633281972,
      "grad_norm": 2.6755611896514893,
      "learning_rate": 4.46516939474686e-06,
      "loss": 1.5882,
      "step": 588
    },
    {
      "epoch": 0.6050333846944016,
      "grad_norm": 2.632248640060425,
      "learning_rate": 4.463266082984393e-06,
      "loss": 1.5481,
      "step": 589
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.6425793170928955,
      "learning_rate": 4.461362771221927e-06,
      "loss": 1.6607,
      "step": 590
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 1.863004446029663,
      "eval_runtime": 37.2064,
      "eval_samples_per_second": 26.877,
      "eval_steps_per_second": 1.129,
      "step": 590
    },
    {
      "epoch": 0.6070878274268104,
      "grad_norm": 2.6340153217315674,
      "learning_rate": 4.45945945945946e-06,
      "loss": 1.5474,
      "step": 591
    },
    {
      "epoch": 0.6081150487930149,
      "grad_norm": 2.7601606845855713,
      "learning_rate": 4.457556147696993e-06,
      "loss": 1.5569,
      "step": 592
    },
    {
      "epoch": 0.6091422701592193,
      "grad_norm": 2.8151822090148926,
      "learning_rate": 4.455652835934526e-06,
      "loss": 1.6441,
      "step": 593
    },
    {
      "epoch": 0.6101694915254238,
      "grad_norm": 2.757577419281006,
      "learning_rate": 4.453749524172059e-06,
      "loss": 1.6339,
      "step": 594
    },
    {
      "epoch": 0.6111967128916281,
      "grad_norm": 2.6828997135162354,
      "learning_rate": 4.451846212409593e-06,
      "loss": 1.5885,
      "step": 595
    },
    {
      "epoch": 0.6122239342578326,
      "grad_norm": 2.6645700931549072,
      "learning_rate": 4.449942900647126e-06,
      "loss": 1.5866,
      "step": 596
    },
    {
      "epoch": 0.613251155624037,
      "grad_norm": 2.572789430618286,
      "learning_rate": 4.4480395888846595e-06,
      "loss": 1.5255,
      "step": 597
    },
    {
      "epoch": 0.6142783769902413,
      "grad_norm": 2.6281981468200684,
      "learning_rate": 4.446136277122193e-06,
      "loss": 1.5905,
      "step": 598
    },
    {
      "epoch": 0.6153055983564458,
      "grad_norm": 2.6164042949676514,
      "learning_rate": 4.444232965359726e-06,
      "loss": 1.5274,
      "step": 599
    },
    {
      "epoch": 0.6163328197226502,
      "grad_norm": 2.606343984603882,
      "learning_rate": 4.4423296535972596e-06,
      "loss": 1.5846,
      "step": 600
    },
    {
      "epoch": 0.6163328197226502,
      "eval_loss": 1.8634170293807983,
      "eval_runtime": 37.1568,
      "eval_samples_per_second": 26.913,
      "eval_steps_per_second": 1.13,
      "step": 600
    },
    {
      "epoch": 0.6173600410888547,
      "grad_norm": 2.8743226528167725,
      "learning_rate": 4.440426341834793e-06,
      "loss": 1.538,
      "step": 601
    },
    {
      "epoch": 0.618387262455059,
      "grad_norm": 2.5392212867736816,
      "learning_rate": 4.438523030072326e-06,
      "loss": 1.585,
      "step": 602
    },
    {
      "epoch": 0.6194144838212635,
      "grad_norm": 2.715684413909912,
      "learning_rate": 4.43661971830986e-06,
      "loss": 1.6028,
      "step": 603
    },
    {
      "epoch": 0.6204417051874679,
      "grad_norm": 2.514934778213501,
      "learning_rate": 4.434716406547393e-06,
      "loss": 1.5525,
      "step": 604
    },
    {
      "epoch": 0.6214689265536724,
      "grad_norm": 2.685807943344116,
      "learning_rate": 4.432813094784926e-06,
      "loss": 1.5456,
      "step": 605
    },
    {
      "epoch": 0.6224961479198767,
      "grad_norm": 2.848557949066162,
      "learning_rate": 4.43090978302246e-06,
      "loss": 1.6017,
      "step": 606
    },
    {
      "epoch": 0.6235233692860811,
      "grad_norm": 2.6820950508117676,
      "learning_rate": 4.429006471259993e-06,
      "loss": 1.573,
      "step": 607
    },
    {
      "epoch": 0.6245505906522856,
      "grad_norm": 2.493544340133667,
      "learning_rate": 4.4271031594975265e-06,
      "loss": 1.5576,
      "step": 608
    },
    {
      "epoch": 0.6255778120184899,
      "grad_norm": 2.6240968704223633,
      "learning_rate": 4.425199847735059e-06,
      "loss": 1.5312,
      "step": 609
    },
    {
      "epoch": 0.6266050333846944,
      "grad_norm": 2.516113042831421,
      "learning_rate": 4.423296535972592e-06,
      "loss": 1.5284,
      "step": 610
    },
    {
      "epoch": 0.6266050333846944,
      "eval_loss": 1.8370482921600342,
      "eval_runtime": 36.9401,
      "eval_samples_per_second": 27.071,
      "eval_steps_per_second": 1.137,
      "step": 610
    },
    {
      "epoch": 0.6276322547508988,
      "grad_norm": 2.8341944217681885,
      "learning_rate": 4.421393224210126e-06,
      "loss": 1.5609,
      "step": 611
    },
    {
      "epoch": 0.6286594761171033,
      "grad_norm": 2.5433242321014404,
      "learning_rate": 4.419489912447659e-06,
      "loss": 1.5253,
      "step": 612
    },
    {
      "epoch": 0.6296866974833076,
      "grad_norm": 2.6761138439178467,
      "learning_rate": 4.4175866006851924e-06,
      "loss": 1.6171,
      "step": 613
    },
    {
      "epoch": 0.6307139188495121,
      "grad_norm": 2.570707321166992,
      "learning_rate": 4.415683288922726e-06,
      "loss": 1.5766,
      "step": 614
    },
    {
      "epoch": 0.6317411402157165,
      "grad_norm": 2.5421109199523926,
      "learning_rate": 4.413779977160259e-06,
      "loss": 1.5616,
      "step": 615
    },
    {
      "epoch": 0.632768361581921,
      "grad_norm": 2.6767091751098633,
      "learning_rate": 4.4118766653977925e-06,
      "loss": 1.5531,
      "step": 616
    },
    {
      "epoch": 0.6337955829481253,
      "grad_norm": 2.7730000019073486,
      "learning_rate": 4.409973353635326e-06,
      "loss": 1.6224,
      "step": 617
    },
    {
      "epoch": 0.6348228043143297,
      "grad_norm": 2.7248494625091553,
      "learning_rate": 4.408070041872859e-06,
      "loss": 1.5752,
      "step": 618
    },
    {
      "epoch": 0.6358500256805342,
      "grad_norm": 2.5071568489074707,
      "learning_rate": 4.406166730110393e-06,
      "loss": 1.5178,
      "step": 619
    },
    {
      "epoch": 0.6368772470467385,
      "grad_norm": 2.5583243370056152,
      "learning_rate": 4.404263418347925e-06,
      "loss": 1.5298,
      "step": 620
    },
    {
      "epoch": 0.6368772470467385,
      "eval_loss": 1.833156943321228,
      "eval_runtime": 37.1108,
      "eval_samples_per_second": 26.946,
      "eval_steps_per_second": 1.132,
      "step": 620
    },
    {
      "epoch": 0.637904468412943,
      "grad_norm": 2.492332696914673,
      "learning_rate": 4.402360106585459e-06,
      "loss": 1.5443,
      "step": 621
    },
    {
      "epoch": 0.6389316897791474,
      "grad_norm": 2.5734798908233643,
      "learning_rate": 4.400456794822993e-06,
      "loss": 1.5362,
      "step": 622
    },
    {
      "epoch": 0.6399589111453519,
      "grad_norm": 2.7597742080688477,
      "learning_rate": 4.398553483060526e-06,
      "loss": 1.56,
      "step": 623
    },
    {
      "epoch": 0.6409861325115562,
      "grad_norm": 2.6911630630493164,
      "learning_rate": 4.396650171298059e-06,
      "loss": 1.5138,
      "step": 624
    },
    {
      "epoch": 0.6420133538777607,
      "grad_norm": 2.596557378768921,
      "learning_rate": 4.394746859535593e-06,
      "loss": 1.5381,
      "step": 625
    },
    {
      "epoch": 0.6430405752439651,
      "grad_norm": 2.5787081718444824,
      "learning_rate": 4.392843547773125e-06,
      "loss": 1.5284,
      "step": 626
    },
    {
      "epoch": 0.6440677966101694,
      "grad_norm": 2.5975289344787598,
      "learning_rate": 4.390940236010659e-06,
      "loss": 1.5359,
      "step": 627
    },
    {
      "epoch": 0.6450950179763739,
      "grad_norm": 2.5175442695617676,
      "learning_rate": 4.389036924248192e-06,
      "loss": 1.5608,
      "step": 628
    },
    {
      "epoch": 0.6461222393425783,
      "grad_norm": 2.535139322280884,
      "learning_rate": 4.387133612485725e-06,
      "loss": 1.5856,
      "step": 629
    },
    {
      "epoch": 0.6471494607087828,
      "grad_norm": 2.5596396923065186,
      "learning_rate": 4.385230300723259e-06,
      "loss": 1.5467,
      "step": 630
    },
    {
      "epoch": 0.6471494607087828,
      "eval_loss": 1.835809588432312,
      "eval_runtime": 37.1073,
      "eval_samples_per_second": 26.949,
      "eval_steps_per_second": 1.132,
      "step": 630
    },
    {
      "epoch": 0.6481766820749871,
      "grad_norm": 2.691441297531128,
      "learning_rate": 4.383326988960792e-06,
      "loss": 1.5516,
      "step": 631
    },
    {
      "epoch": 0.6492039034411916,
      "grad_norm": 2.534902811050415,
      "learning_rate": 4.3814236771983255e-06,
      "loss": 1.5215,
      "step": 632
    },
    {
      "epoch": 0.650231124807396,
      "grad_norm": 2.626211643218994,
      "learning_rate": 4.379520365435859e-06,
      "loss": 1.5811,
      "step": 633
    },
    {
      "epoch": 0.6512583461736005,
      "grad_norm": 2.6646480560302734,
      "learning_rate": 4.377617053673392e-06,
      "loss": 1.5674,
      "step": 634
    },
    {
      "epoch": 0.6522855675398048,
      "grad_norm": 2.5728375911712646,
      "learning_rate": 4.3757137419109256e-06,
      "loss": 1.5655,
      "step": 635
    },
    {
      "epoch": 0.6533127889060092,
      "grad_norm": 2.72817325592041,
      "learning_rate": 4.373810430148459e-06,
      "loss": 1.6283,
      "step": 636
    },
    {
      "epoch": 0.6543400102722137,
      "grad_norm": 2.5395050048828125,
      "learning_rate": 4.3719071183859914e-06,
      "loss": 1.5191,
      "step": 637
    },
    {
      "epoch": 0.655367231638418,
      "grad_norm": 2.964163064956665,
      "learning_rate": 4.370003806623525e-06,
      "loss": 1.5839,
      "step": 638
    },
    {
      "epoch": 0.6563944530046225,
      "grad_norm": 2.517214298248291,
      "learning_rate": 4.368100494861058e-06,
      "loss": 1.5278,
      "step": 639
    },
    {
      "epoch": 0.6574216743708269,
      "grad_norm": 2.765145778656006,
      "learning_rate": 4.3661971830985915e-06,
      "loss": 1.5718,
      "step": 640
    },
    {
      "epoch": 0.6574216743708269,
      "eval_loss": 1.8324981927871704,
      "eval_runtime": 37.0883,
      "eval_samples_per_second": 26.963,
      "eval_steps_per_second": 1.132,
      "step": 640
    },
    {
      "epoch": 0.6584488957370314,
      "grad_norm": 2.463439464569092,
      "learning_rate": 4.364293871336126e-06,
      "loss": 1.5378,
      "step": 641
    },
    {
      "epoch": 0.6594761171032357,
      "grad_norm": 2.7756192684173584,
      "learning_rate": 4.362390559573659e-06,
      "loss": 1.5601,
      "step": 642
    },
    {
      "epoch": 0.6605033384694402,
      "grad_norm": 2.4134342670440674,
      "learning_rate": 4.3604872478111925e-06,
      "loss": 1.5252,
      "step": 643
    },
    {
      "epoch": 0.6615305598356446,
      "grad_norm": 2.4549670219421387,
      "learning_rate": 4.358583936048725e-06,
      "loss": 1.4602,
      "step": 644
    },
    {
      "epoch": 0.662557781201849,
      "grad_norm": 2.6815311908721924,
      "learning_rate": 4.356680624286258e-06,
      "loss": 1.5865,
      "step": 645
    },
    {
      "epoch": 0.6635850025680534,
      "grad_norm": 2.5216286182403564,
      "learning_rate": 4.354777312523792e-06,
      "loss": 1.5432,
      "step": 646
    },
    {
      "epoch": 0.6646122239342578,
      "grad_norm": 2.5547335147857666,
      "learning_rate": 4.352874000761325e-06,
      "loss": 1.59,
      "step": 647
    },
    {
      "epoch": 0.6656394453004623,
      "grad_norm": 2.587646484375,
      "learning_rate": 4.3509706889988584e-06,
      "loss": 1.5351,
      "step": 648
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.530768871307373,
      "learning_rate": 4.349067377236392e-06,
      "loss": 1.603,
      "step": 649
    },
    {
      "epoch": 0.6676938880328711,
      "grad_norm": 2.727213144302368,
      "learning_rate": 4.347164065473925e-06,
      "loss": 1.6032,
      "step": 650
    },
    {
      "epoch": 0.6676938880328711,
      "eval_loss": 1.8441001176834106,
      "eval_runtime": 37.0797,
      "eval_samples_per_second": 26.969,
      "eval_steps_per_second": 1.133,
      "step": 650
    },
    {
      "epoch": 0.6687211093990755,
      "grad_norm": 2.750441312789917,
      "learning_rate": 4.3452607537114585e-06,
      "loss": 1.6172,
      "step": 651
    },
    {
      "epoch": 0.66974833076528,
      "grad_norm": 2.571692705154419,
      "learning_rate": 4.343357441948992e-06,
      "loss": 1.528,
      "step": 652
    },
    {
      "epoch": 0.6707755521314843,
      "grad_norm": 2.5415565967559814,
      "learning_rate": 4.341454130186525e-06,
      "loss": 1.5629,
      "step": 653
    },
    {
      "epoch": 0.6718027734976888,
      "grad_norm": 2.4970591068267822,
      "learning_rate": 4.339550818424058e-06,
      "loss": 1.5299,
      "step": 654
    },
    {
      "epoch": 0.6728299948638932,
      "grad_norm": 2.5283010005950928,
      "learning_rate": 4.337647506661591e-06,
      "loss": 1.5149,
      "step": 655
    },
    {
      "epoch": 0.6738572162300975,
      "grad_norm": 2.5205166339874268,
      "learning_rate": 4.3357441948991245e-06,
      "loss": 1.5486,
      "step": 656
    },
    {
      "epoch": 0.674884437596302,
      "grad_norm": 2.4621481895446777,
      "learning_rate": 4.333840883136658e-06,
      "loss": 1.5769,
      "step": 657
    },
    {
      "epoch": 0.6759116589625064,
      "grad_norm": 2.5954174995422363,
      "learning_rate": 4.331937571374191e-06,
      "loss": 1.5242,
      "step": 658
    },
    {
      "epoch": 0.6769388803287109,
      "grad_norm": 2.553523063659668,
      "learning_rate": 4.3300342596117246e-06,
      "loss": 1.5339,
      "step": 659
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 2.5289323329925537,
      "learning_rate": 4.328130947849258e-06,
      "loss": 1.5371,
      "step": 660
    },
    {
      "epoch": 0.6779661016949152,
      "eval_loss": 1.8562791347503662,
      "eval_runtime": 37.1299,
      "eval_samples_per_second": 26.932,
      "eval_steps_per_second": 1.131,
      "step": 660
    },
    {
      "epoch": 0.6789933230611197,
      "grad_norm": 2.537010908126831,
      "learning_rate": 4.326227636086791e-06,
      "loss": 1.5005,
      "step": 661
    },
    {
      "epoch": 0.6800205444273241,
      "grad_norm": 2.6716437339782715,
      "learning_rate": 4.324324324324325e-06,
      "loss": 1.5501,
      "step": 662
    },
    {
      "epoch": 0.6810477657935285,
      "grad_norm": 2.4969863891601562,
      "learning_rate": 4.322421012561858e-06,
      "loss": 1.5474,
      "step": 663
    },
    {
      "epoch": 0.6820749871597329,
      "grad_norm": 2.6803722381591797,
      "learning_rate": 4.320517700799391e-06,
      "loss": 1.5397,
      "step": 664
    },
    {
      "epoch": 0.6831022085259374,
      "grad_norm": 2.6370725631713867,
      "learning_rate": 4.318614389036925e-06,
      "loss": 1.6274,
      "step": 665
    },
    {
      "epoch": 0.6841294298921418,
      "grad_norm": 2.6329426765441895,
      "learning_rate": 4.316711077274458e-06,
      "loss": 1.6015,
      "step": 666
    },
    {
      "epoch": 0.6851566512583461,
      "grad_norm": 2.6113548278808594,
      "learning_rate": 4.3148077655119915e-06,
      "loss": 1.5015,
      "step": 667
    },
    {
      "epoch": 0.6861838726245506,
      "grad_norm": 2.718801498413086,
      "learning_rate": 4.312904453749525e-06,
      "loss": 1.597,
      "step": 668
    },
    {
      "epoch": 0.687211093990755,
      "grad_norm": 2.638871908187866,
      "learning_rate": 4.311001141987058e-06,
      "loss": 1.5528,
      "step": 669
    },
    {
      "epoch": 0.6882383153569595,
      "grad_norm": 2.74849534034729,
      "learning_rate": 4.3090978302245916e-06,
      "loss": 1.5627,
      "step": 670
    },
    {
      "epoch": 0.6882383153569595,
      "eval_loss": 1.8493062257766724,
      "eval_runtime": 37.2964,
      "eval_samples_per_second": 26.812,
      "eval_steps_per_second": 1.126,
      "step": 670
    },
    {
      "epoch": 0.6892655367231638,
      "grad_norm": 2.6839728355407715,
      "learning_rate": 4.307194518462124e-06,
      "loss": 1.552,
      "step": 671
    },
    {
      "epoch": 0.6902927580893683,
      "grad_norm": 2.844496488571167,
      "learning_rate": 4.3052912066996574e-06,
      "loss": 1.6218,
      "step": 672
    },
    {
      "epoch": 0.6913199794555727,
      "grad_norm": 2.4792141914367676,
      "learning_rate": 4.303387894937191e-06,
      "loss": 1.592,
      "step": 673
    },
    {
      "epoch": 0.6923472008217771,
      "grad_norm": 2.4852263927459717,
      "learning_rate": 4.301484583174724e-06,
      "loss": 1.5165,
      "step": 674
    },
    {
      "epoch": 0.6933744221879815,
      "grad_norm": 2.5438406467437744,
      "learning_rate": 4.2995812714122575e-06,
      "loss": 1.5109,
      "step": 675
    },
    {
      "epoch": 0.6944016435541859,
      "grad_norm": 2.5431931018829346,
      "learning_rate": 4.297677959649791e-06,
      "loss": 1.5345,
      "step": 676
    },
    {
      "epoch": 0.6954288649203904,
      "grad_norm": 2.650897979736328,
      "learning_rate": 4.295774647887324e-06,
      "loss": 1.5607,
      "step": 677
    },
    {
      "epoch": 0.6964560862865947,
      "grad_norm": 2.721167802810669,
      "learning_rate": 4.293871336124858e-06,
      "loss": 1.5764,
      "step": 678
    },
    {
      "epoch": 0.6974833076527992,
      "grad_norm": 2.5402960777282715,
      "learning_rate": 4.291968024362391e-06,
      "loss": 1.5505,
      "step": 679
    },
    {
      "epoch": 0.6985105290190036,
      "grad_norm": 2.6878864765167236,
      "learning_rate": 4.290064712599924e-06,
      "loss": 1.5821,
      "step": 680
    },
    {
      "epoch": 0.6985105290190036,
      "eval_loss": 1.8400572538375854,
      "eval_runtime": 37.3885,
      "eval_samples_per_second": 26.746,
      "eval_steps_per_second": 1.123,
      "step": 680
    },
    {
      "epoch": 0.699537750385208,
      "grad_norm": 2.6703765392303467,
      "learning_rate": 4.288161400837458e-06,
      "loss": 1.5794,
      "step": 681
    },
    {
      "epoch": 0.7005649717514124,
      "grad_norm": 2.626725196838379,
      "learning_rate": 4.286258089074991e-06,
      "loss": 1.4986,
      "step": 682
    },
    {
      "epoch": 0.7015921931176169,
      "grad_norm": 2.6062862873077393,
      "learning_rate": 4.284354777312524e-06,
      "loss": 1.5606,
      "step": 683
    },
    {
      "epoch": 0.7026194144838213,
      "grad_norm": 2.6359503269195557,
      "learning_rate": 4.282451465550058e-06,
      "loss": 1.5174,
      "step": 684
    },
    {
      "epoch": 0.7036466358500256,
      "grad_norm": 2.8453221321105957,
      "learning_rate": 4.280548153787591e-06,
      "loss": 1.543,
      "step": 685
    },
    {
      "epoch": 0.7046738572162301,
      "grad_norm": 2.5195140838623047,
      "learning_rate": 4.2786448420251245e-06,
      "loss": 1.5462,
      "step": 686
    },
    {
      "epoch": 0.7057010785824345,
      "grad_norm": 2.757230281829834,
      "learning_rate": 4.276741530262658e-06,
      "loss": 1.5033,
      "step": 687
    },
    {
      "epoch": 0.706728299948639,
      "grad_norm": 2.708148241043091,
      "learning_rate": 4.27483821850019e-06,
      "loss": 1.5286,
      "step": 688
    },
    {
      "epoch": 0.7077555213148433,
      "grad_norm": 2.6836555004119873,
      "learning_rate": 4.272934906737724e-06,
      "loss": 1.5973,
      "step": 689
    },
    {
      "epoch": 0.7087827426810478,
      "grad_norm": 2.6970176696777344,
      "learning_rate": 4.271031594975257e-06,
      "loss": 1.5844,
      "step": 690
    },
    {
      "epoch": 0.7087827426810478,
      "eval_loss": 1.8414137363433838,
      "eval_runtime": 37.3847,
      "eval_samples_per_second": 26.749,
      "eval_steps_per_second": 1.123,
      "step": 690
    },
    {
      "epoch": 0.7098099640472522,
      "grad_norm": 2.69212007522583,
      "learning_rate": 4.2691282832127905e-06,
      "loss": 1.5817,
      "step": 691
    },
    {
      "epoch": 0.7108371854134566,
      "grad_norm": 2.6203925609588623,
      "learning_rate": 4.267224971450324e-06,
      "loss": 1.5411,
      "step": 692
    },
    {
      "epoch": 0.711864406779661,
      "grad_norm": 2.418916940689087,
      "learning_rate": 4.265321659687857e-06,
      "loss": 1.483,
      "step": 693
    },
    {
      "epoch": 0.7128916281458655,
      "grad_norm": 2.784196615219116,
      "learning_rate": 4.2634183479253906e-06,
      "loss": 1.5286,
      "step": 694
    },
    {
      "epoch": 0.7139188495120699,
      "grad_norm": 2.675043821334839,
      "learning_rate": 4.261515036162924e-06,
      "loss": 1.5067,
      "step": 695
    },
    {
      "epoch": 0.7149460708782742,
      "grad_norm": 2.7075769901275635,
      "learning_rate": 4.259611724400457e-06,
      "loss": 1.5316,
      "step": 696
    },
    {
      "epoch": 0.7159732922444787,
      "grad_norm": 2.911898612976074,
      "learning_rate": 4.257708412637991e-06,
      "loss": 1.5203,
      "step": 697
    },
    {
      "epoch": 0.7170005136106831,
      "grad_norm": 2.748126983642578,
      "learning_rate": 4.255805100875523e-06,
      "loss": 1.465,
      "step": 698
    },
    {
      "epoch": 0.7180277349768875,
      "grad_norm": 2.6017746925354004,
      "learning_rate": 4.2539017891130565e-06,
      "loss": 1.4956,
      "step": 699
    },
    {
      "epoch": 0.7190549563430919,
      "grad_norm": 3.0877811908721924,
      "learning_rate": 4.25199847735059e-06,
      "loss": 1.587,
      "step": 700
    },
    {
      "epoch": 0.7190549563430919,
      "eval_loss": 1.8209773302078247,
      "eval_runtime": 37.1796,
      "eval_samples_per_second": 26.896,
      "eval_steps_per_second": 1.13,
      "step": 700
    },
    {
      "epoch": 0.7200821777092964,
      "grad_norm": 2.6982834339141846,
      "learning_rate": 4.250095165588123e-06,
      "loss": 1.4984,
      "step": 701
    },
    {
      "epoch": 0.7211093990755008,
      "grad_norm": 2.7425835132598877,
      "learning_rate": 4.2481918538256575e-06,
      "loss": 1.5596,
      "step": 702
    },
    {
      "epoch": 0.7221366204417052,
      "grad_norm": 2.838890314102173,
      "learning_rate": 4.246288542063191e-06,
      "loss": 1.5207,
      "step": 703
    },
    {
      "epoch": 0.7231638418079096,
      "grad_norm": 2.694871664047241,
      "learning_rate": 4.244385230300724e-06,
      "loss": 1.5592,
      "step": 704
    },
    {
      "epoch": 0.724191063174114,
      "grad_norm": 2.630206823348999,
      "learning_rate": 4.242481918538257e-06,
      "loss": 1.5005,
      "step": 705
    },
    {
      "epoch": 0.7252182845403184,
      "grad_norm": 2.9697742462158203,
      "learning_rate": 4.24057860677579e-06,
      "loss": 1.5329,
      "step": 706
    },
    {
      "epoch": 0.7262455059065228,
      "grad_norm": 2.8384499549865723,
      "learning_rate": 4.2386752950133234e-06,
      "loss": 1.5731,
      "step": 707
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.530134916305542,
      "learning_rate": 4.236771983250857e-06,
      "loss": 1.5187,
      "step": 708
    },
    {
      "epoch": 0.7282999486389317,
      "grad_norm": 2.601384162902832,
      "learning_rate": 4.23486867148839e-06,
      "loss": 1.5069,
      "step": 709
    },
    {
      "epoch": 0.7293271700051361,
      "grad_norm": 2.7746055126190186,
      "learning_rate": 4.2329653597259235e-06,
      "loss": 1.5377,
      "step": 710
    },
    {
      "epoch": 0.7293271700051361,
      "eval_loss": 1.8144561052322388,
      "eval_runtime": 36.9983,
      "eval_samples_per_second": 27.028,
      "eval_steps_per_second": 1.135,
      "step": 710
    },
    {
      "epoch": 0.7303543913713405,
      "grad_norm": 2.860414505004883,
      "learning_rate": 4.231062047963457e-06,
      "loss": 1.5604,
      "step": 711
    },
    {
      "epoch": 0.731381612737545,
      "grad_norm": 2.7479686737060547,
      "learning_rate": 4.22915873620099e-06,
      "loss": 1.578,
      "step": 712
    },
    {
      "epoch": 0.7324088341037494,
      "grad_norm": 3.093319892883301,
      "learning_rate": 4.227255424438524e-06,
      "loss": 1.5609,
      "step": 713
    },
    {
      "epoch": 0.7334360554699538,
      "grad_norm": 2.611778497695923,
      "learning_rate": 4.225352112676057e-06,
      "loss": 1.5775,
      "step": 714
    },
    {
      "epoch": 0.7344632768361582,
      "grad_norm": 2.5740373134613037,
      "learning_rate": 4.2234488009135895e-06,
      "loss": 1.5059,
      "step": 715
    },
    {
      "epoch": 0.7354904982023626,
      "grad_norm": 2.5564701557159424,
      "learning_rate": 4.221545489151123e-06,
      "loss": 1.4687,
      "step": 716
    },
    {
      "epoch": 0.736517719568567,
      "grad_norm": 2.566359758377075,
      "learning_rate": 4.219642177388656e-06,
      "loss": 1.5084,
      "step": 717
    },
    {
      "epoch": 0.7375449409347714,
      "grad_norm": 2.632016897201538,
      "learning_rate": 4.2177388656261896e-06,
      "loss": 1.5449,
      "step": 718
    },
    {
      "epoch": 0.7385721623009759,
      "grad_norm": 2.525616407394409,
      "learning_rate": 4.215835553863723e-06,
      "loss": 1.4882,
      "step": 719
    },
    {
      "epoch": 0.7395993836671803,
      "grad_norm": 2.674612522125244,
      "learning_rate": 4.213932242101256e-06,
      "loss": 1.4576,
      "step": 720
    },
    {
      "epoch": 0.7395993836671803,
      "eval_loss": 1.8278547525405884,
      "eval_runtime": 37.0938,
      "eval_samples_per_second": 26.959,
      "eval_steps_per_second": 1.132,
      "step": 720
    },
    {
      "epoch": 0.7406266050333847,
      "grad_norm": 2.8636388778686523,
      "learning_rate": 4.21202893033879e-06,
      "loss": 1.5149,
      "step": 721
    },
    {
      "epoch": 0.7416538263995891,
      "grad_norm": 2.7943782806396484,
      "learning_rate": 4.210125618576323e-06,
      "loss": 1.5409,
      "step": 722
    },
    {
      "epoch": 0.7426810477657936,
      "grad_norm": 2.5911595821380615,
      "learning_rate": 4.208222306813856e-06,
      "loss": 1.5324,
      "step": 723
    },
    {
      "epoch": 0.743708269131998,
      "grad_norm": 2.7497329711914062,
      "learning_rate": 4.20631899505139e-06,
      "loss": 1.5842,
      "step": 724
    },
    {
      "epoch": 0.7447354904982023,
      "grad_norm": 2.649540424346924,
      "learning_rate": 4.204415683288923e-06,
      "loss": 1.438,
      "step": 725
    },
    {
      "epoch": 0.7457627118644068,
      "grad_norm": 2.696852684020996,
      "learning_rate": 4.2025123715264565e-06,
      "loss": 1.5538,
      "step": 726
    },
    {
      "epoch": 0.7467899332306112,
      "grad_norm": 2.553107738494873,
      "learning_rate": 4.20060905976399e-06,
      "loss": 1.5115,
      "step": 727
    },
    {
      "epoch": 0.7478171545968156,
      "grad_norm": 2.559718370437622,
      "learning_rate": 4.198705748001523e-06,
      "loss": 1.4613,
      "step": 728
    },
    {
      "epoch": 0.74884437596302,
      "grad_norm": 2.615692377090454,
      "learning_rate": 4.1968024362390566e-06,
      "loss": 1.5655,
      "step": 729
    },
    {
      "epoch": 0.7498715973292245,
      "grad_norm": 2.432598114013672,
      "learning_rate": 4.19489912447659e-06,
      "loss": 1.5192,
      "step": 730
    },
    {
      "epoch": 0.7498715973292245,
      "eval_loss": 1.8231704235076904,
      "eval_runtime": 37.1648,
      "eval_samples_per_second": 26.907,
      "eval_steps_per_second": 1.13,
      "step": 730
    },
    {
      "epoch": 0.7508988186954288,
      "grad_norm": 2.63838529586792,
      "learning_rate": 4.192995812714123e-06,
      "loss": 1.5888,
      "step": 731
    },
    {
      "epoch": 0.7519260400616333,
      "grad_norm": 2.711975574493408,
      "learning_rate": 4.191092500951656e-06,
      "loss": 1.5309,
      "step": 732
    },
    {
      "epoch": 0.7529532614278377,
      "grad_norm": 2.459561347961426,
      "learning_rate": 4.189189189189189e-06,
      "loss": 1.5141,
      "step": 733
    },
    {
      "epoch": 0.7539804827940421,
      "grad_norm": 2.7000269889831543,
      "learning_rate": 4.1872858774267225e-06,
      "loss": 1.5148,
      "step": 734
    },
    {
      "epoch": 0.7550077041602465,
      "grad_norm": 2.8983139991760254,
      "learning_rate": 4.185382565664256e-06,
      "loss": 1.5766,
      "step": 735
    },
    {
      "epoch": 0.7560349255264509,
      "grad_norm": 2.600132942199707,
      "learning_rate": 4.183479253901789e-06,
      "loss": 1.4987,
      "step": 736
    },
    {
      "epoch": 0.7570621468926554,
      "grad_norm": 2.579481363296509,
      "learning_rate": 4.181575942139323e-06,
      "loss": 1.4761,
      "step": 737
    },
    {
      "epoch": 0.7580893682588598,
      "grad_norm": 2.708799123764038,
      "learning_rate": 4.179672630376856e-06,
      "loss": 1.5278,
      "step": 738
    },
    {
      "epoch": 0.7591165896250642,
      "grad_norm": 2.4836950302124023,
      "learning_rate": 4.177769318614389e-06,
      "loss": 1.5056,
      "step": 739
    },
    {
      "epoch": 0.7601438109912686,
      "grad_norm": 2.4556121826171875,
      "learning_rate": 4.175866006851923e-06,
      "loss": 1.5637,
      "step": 740
    },
    {
      "epoch": 0.7601438109912686,
      "eval_loss": 1.8386361598968506,
      "eval_runtime": 37.1304,
      "eval_samples_per_second": 26.932,
      "eval_steps_per_second": 1.131,
      "step": 740
    },
    {
      "epoch": 0.7611710323574731,
      "grad_norm": 2.504653215408325,
      "learning_rate": 4.173962695089456e-06,
      "loss": 1.5519,
      "step": 741
    },
    {
      "epoch": 0.7621982537236774,
      "grad_norm": 2.50435733795166,
      "learning_rate": 4.1720593833269894e-06,
      "loss": 1.5379,
      "step": 742
    },
    {
      "epoch": 0.7632254750898819,
      "grad_norm": 2.5818099975585938,
      "learning_rate": 4.170156071564523e-06,
      "loss": 1.5462,
      "step": 743
    },
    {
      "epoch": 0.7642526964560863,
      "grad_norm": 2.6530778408050537,
      "learning_rate": 4.168252759802056e-06,
      "loss": 1.6112,
      "step": 744
    },
    {
      "epoch": 0.7652799178222907,
      "grad_norm": 2.463993549346924,
      "learning_rate": 4.1663494480395895e-06,
      "loss": 1.5174,
      "step": 745
    },
    {
      "epoch": 0.7663071391884951,
      "grad_norm": 2.6928186416625977,
      "learning_rate": 4.164446136277123e-06,
      "loss": 1.5639,
      "step": 746
    },
    {
      "epoch": 0.7673343605546995,
      "grad_norm": 2.5539324283599854,
      "learning_rate": 4.162542824514656e-06,
      "loss": 1.5557,
      "step": 747
    },
    {
      "epoch": 0.768361581920904,
      "grad_norm": 2.6068854331970215,
      "learning_rate": 4.16063951275219e-06,
      "loss": 1.5309,
      "step": 748
    },
    {
      "epoch": 0.7693888032871083,
      "grad_norm": 2.5830016136169434,
      "learning_rate": 4.158736200989722e-06,
      "loss": 1.4876,
      "step": 749
    },
    {
      "epoch": 0.7704160246533128,
      "grad_norm": 2.6156973838806152,
      "learning_rate": 4.1568328892272555e-06,
      "loss": 1.5284,
      "step": 750
    },
    {
      "epoch": 0.7704160246533128,
      "eval_loss": 1.8401216268539429,
      "eval_runtime": 37.1657,
      "eval_samples_per_second": 26.907,
      "eval_steps_per_second": 1.13,
      "step": 750
    },
    {
      "epoch": 0.7714432460195172,
      "grad_norm": 2.6343140602111816,
      "learning_rate": 4.154929577464789e-06,
      "loss": 1.5348,
      "step": 751
    },
    {
      "epoch": 0.7724704673857217,
      "grad_norm": 2.483319044113159,
      "learning_rate": 4.153026265702322e-06,
      "loss": 1.5344,
      "step": 752
    },
    {
      "epoch": 0.773497688751926,
      "grad_norm": 2.755953550338745,
      "learning_rate": 4.1511229539398556e-06,
      "loss": 1.4715,
      "step": 753
    },
    {
      "epoch": 0.7745249101181304,
      "grad_norm": 2.5264573097229004,
      "learning_rate": 4.149219642177389e-06,
      "loss": 1.5236,
      "step": 754
    },
    {
      "epoch": 0.7755521314843349,
      "grad_norm": 2.793328046798706,
      "learning_rate": 4.147316330414922e-06,
      "loss": 1.5316,
      "step": 755
    },
    {
      "epoch": 0.7765793528505393,
      "grad_norm": 2.7607345581054688,
      "learning_rate": 4.145413018652456e-06,
      "loss": 1.4518,
      "step": 756
    },
    {
      "epoch": 0.7776065742167437,
      "grad_norm": 2.5960099697113037,
      "learning_rate": 4.143509706889989e-06,
      "loss": 1.5499,
      "step": 757
    },
    {
      "epoch": 0.7786337955829481,
      "grad_norm": 2.752039909362793,
      "learning_rate": 4.141606395127522e-06,
      "loss": 1.5433,
      "step": 758
    },
    {
      "epoch": 0.7796610169491526,
      "grad_norm": 2.724228620529175,
      "learning_rate": 4.139703083365056e-06,
      "loss": 1.5242,
      "step": 759
    },
    {
      "epoch": 0.7806882383153569,
      "grad_norm": 2.668635368347168,
      "learning_rate": 4.137799771602588e-06,
      "loss": 1.563,
      "step": 760
    },
    {
      "epoch": 0.7806882383153569,
      "eval_loss": 1.8386328220367432,
      "eval_runtime": 37.1498,
      "eval_samples_per_second": 26.918,
      "eval_steps_per_second": 1.131,
      "step": 760
    },
    {
      "epoch": 0.7817154596815614,
      "grad_norm": 2.5619916915893555,
      "learning_rate": 4.135896459840122e-06,
      "loss": 1.5398,
      "step": 761
    },
    {
      "epoch": 0.7827426810477658,
      "grad_norm": 2.642075538635254,
      "learning_rate": 4.133993148077656e-06,
      "loss": 1.4938,
      "step": 762
    },
    {
      "epoch": 0.7837699024139703,
      "grad_norm": 2.700350522994995,
      "learning_rate": 4.132089836315189e-06,
      "loss": 1.5196,
      "step": 763
    },
    {
      "epoch": 0.7847971237801746,
      "grad_norm": 2.9511470794677734,
      "learning_rate": 4.1301865245527226e-06,
      "loss": 1.531,
      "step": 764
    },
    {
      "epoch": 0.785824345146379,
      "grad_norm": 2.681643486022949,
      "learning_rate": 4.128283212790256e-06,
      "loss": 1.4747,
      "step": 765
    },
    {
      "epoch": 0.7868515665125835,
      "grad_norm": 2.5386908054351807,
      "learning_rate": 4.1263799010277884e-06,
      "loss": 1.4424,
      "step": 766
    },
    {
      "epoch": 0.7878787878787878,
      "grad_norm": 2.8831303119659424,
      "learning_rate": 4.124476589265322e-06,
      "loss": 1.5677,
      "step": 767
    },
    {
      "epoch": 0.7889060092449923,
      "grad_norm": 2.593442678451538,
      "learning_rate": 4.122573277502855e-06,
      "loss": 1.5043,
      "step": 768
    },
    {
      "epoch": 0.7899332306111967,
      "grad_norm": 2.6778204441070557,
      "learning_rate": 4.1206699657403885e-06,
      "loss": 1.5053,
      "step": 769
    },
    {
      "epoch": 0.7909604519774012,
      "grad_norm": 2.6590428352355957,
      "learning_rate": 4.118766653977922e-06,
      "loss": 1.5338,
      "step": 770
    },
    {
      "epoch": 0.7909604519774012,
      "eval_loss": 1.8266831636428833,
      "eval_runtime": 37.1428,
      "eval_samples_per_second": 26.923,
      "eval_steps_per_second": 1.131,
      "step": 770
    },
    {
      "epoch": 0.7919876733436055,
      "grad_norm": 2.6895735263824463,
      "learning_rate": 4.116863342215455e-06,
      "loss": 1.5067,
      "step": 771
    },
    {
      "epoch": 0.79301489470981,
      "grad_norm": 2.5284390449523926,
      "learning_rate": 4.114960030452989e-06,
      "loss": 1.4775,
      "step": 772
    },
    {
      "epoch": 0.7940421160760144,
      "grad_norm": 2.610442876815796,
      "learning_rate": 4.113056718690522e-06,
      "loss": 1.519,
      "step": 773
    },
    {
      "epoch": 0.7950693374422187,
      "grad_norm": 2.5558695793151855,
      "learning_rate": 4.111153406928055e-06,
      "loss": 1.4565,
      "step": 774
    },
    {
      "epoch": 0.7960965588084232,
      "grad_norm": 2.6313698291778564,
      "learning_rate": 4.109250095165589e-06,
      "loss": 1.5626,
      "step": 775
    },
    {
      "epoch": 0.7971237801746276,
      "grad_norm": 2.553133726119995,
      "learning_rate": 4.107346783403122e-06,
      "loss": 1.5591,
      "step": 776
    },
    {
      "epoch": 0.7981510015408321,
      "grad_norm": 2.6599066257476807,
      "learning_rate": 4.105443471640655e-06,
      "loss": 1.5064,
      "step": 777
    },
    {
      "epoch": 0.7991782229070364,
      "grad_norm": 2.5582847595214844,
      "learning_rate": 4.103540159878188e-06,
      "loss": 1.4824,
      "step": 778
    },
    {
      "epoch": 0.8002054442732409,
      "grad_norm": 2.658006429672241,
      "learning_rate": 4.101636848115721e-06,
      "loss": 1.5187,
      "step": 779
    },
    {
      "epoch": 0.8012326656394453,
      "grad_norm": 2.5531060695648193,
      "learning_rate": 4.099733536353255e-06,
      "loss": 1.4862,
      "step": 780
    },
    {
      "epoch": 0.8012326656394453,
      "eval_loss": 1.8293416500091553,
      "eval_runtime": 37.2532,
      "eval_samples_per_second": 26.843,
      "eval_steps_per_second": 1.127,
      "step": 780
    },
    {
      "epoch": 0.8022598870056498,
      "grad_norm": 2.7279789447784424,
      "learning_rate": 4.097830224590788e-06,
      "loss": 1.5299,
      "step": 781
    },
    {
      "epoch": 0.8032871083718541,
      "grad_norm": 2.7395598888397217,
      "learning_rate": 4.095926912828322e-06,
      "loss": 1.6016,
      "step": 782
    },
    {
      "epoch": 0.8043143297380585,
      "grad_norm": 2.6175215244293213,
      "learning_rate": 4.094023601065855e-06,
      "loss": 1.5657,
      "step": 783
    },
    {
      "epoch": 0.805341551104263,
      "grad_norm": 2.4423458576202393,
      "learning_rate": 4.092120289303388e-06,
      "loss": 1.5085,
      "step": 784
    },
    {
      "epoch": 0.8063687724704673,
      "grad_norm": 2.569363594055176,
      "learning_rate": 4.0902169775409215e-06,
      "loss": 1.5272,
      "step": 785
    },
    {
      "epoch": 0.8073959938366718,
      "grad_norm": 2.5281481742858887,
      "learning_rate": 4.088313665778455e-06,
      "loss": 1.473,
      "step": 786
    },
    {
      "epoch": 0.8084232152028762,
      "grad_norm": 2.527724504470825,
      "learning_rate": 4.086410354015988e-06,
      "loss": 1.4978,
      "step": 787
    },
    {
      "epoch": 0.8094504365690807,
      "grad_norm": 2.524348497390747,
      "learning_rate": 4.0845070422535216e-06,
      "loss": 1.4818,
      "step": 788
    },
    {
      "epoch": 0.810477657935285,
      "grad_norm": 2.5176353454589844,
      "learning_rate": 4.082603730491055e-06,
      "loss": 1.4946,
      "step": 789
    },
    {
      "epoch": 0.8115048793014895,
      "grad_norm": 2.63879132270813,
      "learning_rate": 4.080700418728588e-06,
      "loss": 1.5411,
      "step": 790
    },
    {
      "epoch": 0.8115048793014895,
      "eval_loss": 1.8180036544799805,
      "eval_runtime": 37.4282,
      "eval_samples_per_second": 26.718,
      "eval_steps_per_second": 1.122,
      "step": 790
    },
    {
      "epoch": 0.8125321006676939,
      "grad_norm": 2.5492401123046875,
      "learning_rate": 4.078797106966122e-06,
      "loss": 1.5185,
      "step": 791
    },
    {
      "epoch": 0.8135593220338984,
      "grad_norm": 2.668663501739502,
      "learning_rate": 4.076893795203655e-06,
      "loss": 1.4187,
      "step": 792
    },
    {
      "epoch": 0.8145865434001027,
      "grad_norm": 2.500401496887207,
      "learning_rate": 4.074990483441188e-06,
      "loss": 1.4902,
      "step": 793
    },
    {
      "epoch": 0.8156137647663071,
      "grad_norm": 2.7383482456207275,
      "learning_rate": 4.073087171678721e-06,
      "loss": 1.5454,
      "step": 794
    },
    {
      "epoch": 0.8166409861325116,
      "grad_norm": 2.7144806385040283,
      "learning_rate": 4.071183859916254e-06,
      "loss": 1.5336,
      "step": 795
    },
    {
      "epoch": 0.8176682074987159,
      "grad_norm": 2.8088390827178955,
      "learning_rate": 4.069280548153788e-06,
      "loss": 1.5107,
      "step": 796
    },
    {
      "epoch": 0.8186954288649204,
      "grad_norm": 2.786731004714966,
      "learning_rate": 4.067377236391321e-06,
      "loss": 1.4762,
      "step": 797
    },
    {
      "epoch": 0.8197226502311248,
      "grad_norm": 2.7185654640197754,
      "learning_rate": 4.065473924628854e-06,
      "loss": 1.5129,
      "step": 798
    },
    {
      "epoch": 0.8207498715973293,
      "grad_norm": 2.5182769298553467,
      "learning_rate": 4.063570612866388e-06,
      "loss": 1.5392,
      "step": 799
    },
    {
      "epoch": 0.8217770929635336,
      "grad_norm": 2.594088554382324,
      "learning_rate": 4.061667301103921e-06,
      "loss": 1.5389,
      "step": 800
    },
    {
      "epoch": 0.8217770929635336,
      "eval_loss": 1.8078373670578003,
      "eval_runtime": 37.4191,
      "eval_samples_per_second": 26.724,
      "eval_steps_per_second": 1.122,
      "step": 800
    },
    {
      "epoch": 0.8228043143297381,
      "grad_norm": 2.5099830627441406,
      "learning_rate": 4.0597639893414544e-06,
      "loss": 1.5033,
      "step": 801
    },
    {
      "epoch": 0.8238315356959425,
      "grad_norm": 2.514415740966797,
      "learning_rate": 4.057860677578988e-06,
      "loss": 1.5152,
      "step": 802
    },
    {
      "epoch": 0.8248587570621468,
      "grad_norm": 2.738563060760498,
      "learning_rate": 4.055957365816521e-06,
      "loss": 1.5302,
      "step": 803
    },
    {
      "epoch": 0.8258859784283513,
      "grad_norm": 2.633472204208374,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 1.4735,
      "step": 804
    },
    {
      "epoch": 0.8269131997945557,
      "grad_norm": 2.6635546684265137,
      "learning_rate": 4.052150742291588e-06,
      "loss": 1.4675,
      "step": 805
    },
    {
      "epoch": 0.8279404211607602,
      "grad_norm": 2.515171527862549,
      "learning_rate": 4.050247430529121e-06,
      "loss": 1.4682,
      "step": 806
    },
    {
      "epoch": 0.8289676425269645,
      "grad_norm": 2.689464569091797,
      "learning_rate": 4.048344118766655e-06,
      "loss": 1.5499,
      "step": 807
    },
    {
      "epoch": 0.829994863893169,
      "grad_norm": 2.639822244644165,
      "learning_rate": 4.046440807004188e-06,
      "loss": 1.5297,
      "step": 808
    },
    {
      "epoch": 0.8310220852593734,
      "grad_norm": 2.6470797061920166,
      "learning_rate": 4.044537495241721e-06,
      "loss": 1.5215,
      "step": 809
    },
    {
      "epoch": 0.8320493066255779,
      "grad_norm": 2.4769506454467773,
      "learning_rate": 4.042634183479255e-06,
      "loss": 1.4859,
      "step": 810
    },
    {
      "epoch": 0.8320493066255779,
      "eval_loss": 1.8010139465332031,
      "eval_runtime": 36.974,
      "eval_samples_per_second": 27.046,
      "eval_steps_per_second": 1.136,
      "step": 810
    },
    {
      "epoch": 0.8330765279917822,
      "grad_norm": 2.72485613822937,
      "learning_rate": 4.040730871716787e-06,
      "loss": 1.5301,
      "step": 811
    },
    {
      "epoch": 0.8341037493579867,
      "grad_norm": 2.522390365600586,
      "learning_rate": 4.038827559954321e-06,
      "loss": 1.5199,
      "step": 812
    },
    {
      "epoch": 0.8351309707241911,
      "grad_norm": 2.6865367889404297,
      "learning_rate": 4.036924248191854e-06,
      "loss": 1.5605,
      "step": 813
    },
    {
      "epoch": 0.8361581920903954,
      "grad_norm": 2.6478004455566406,
      "learning_rate": 4.035020936429387e-06,
      "loss": 1.5291,
      "step": 814
    },
    {
      "epoch": 0.8371854134565999,
      "grad_norm": 2.632477045059204,
      "learning_rate": 4.033117624666921e-06,
      "loss": 1.5134,
      "step": 815
    },
    {
      "epoch": 0.8382126348228043,
      "grad_norm": 2.683198928833008,
      "learning_rate": 4.031214312904454e-06,
      "loss": 1.4787,
      "step": 816
    },
    {
      "epoch": 0.8392398561890088,
      "grad_norm": 2.536470651626587,
      "learning_rate": 4.029311001141987e-06,
      "loss": 1.4717,
      "step": 817
    },
    {
      "epoch": 0.8402670775552131,
      "grad_norm": 2.5733802318573,
      "learning_rate": 4.027407689379521e-06,
      "loss": 1.4507,
      "step": 818
    },
    {
      "epoch": 0.8412942989214176,
      "grad_norm": 2.779752016067505,
      "learning_rate": 4.025504377617054e-06,
      "loss": 1.503,
      "step": 819
    },
    {
      "epoch": 0.842321520287622,
      "grad_norm": 2.6932976245880127,
      "learning_rate": 4.0236010658545875e-06,
      "loss": 1.4815,
      "step": 820
    },
    {
      "epoch": 0.842321520287622,
      "eval_loss": 1.7960872650146484,
      "eval_runtime": 37.0481,
      "eval_samples_per_second": 26.992,
      "eval_steps_per_second": 1.134,
      "step": 820
    },
    {
      "epoch": 0.8433487416538265,
      "grad_norm": 2.7531607151031494,
      "learning_rate": 4.02169775409212e-06,
      "loss": 1.5286,
      "step": 821
    },
    {
      "epoch": 0.8443759630200308,
      "grad_norm": 2.8817851543426514,
      "learning_rate": 4.019794442329653e-06,
      "loss": 1.5765,
      "step": 822
    },
    {
      "epoch": 0.8454031843862352,
      "grad_norm": 2.544451951980591,
      "learning_rate": 4.0178911305671876e-06,
      "loss": 1.452,
      "step": 823
    },
    {
      "epoch": 0.8464304057524397,
      "grad_norm": 2.8142952919006348,
      "learning_rate": 4.015987818804721e-06,
      "loss": 1.5374,
      "step": 824
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 2.5938656330108643,
      "learning_rate": 4.014084507042254e-06,
      "loss": 1.4756,
      "step": 825
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 2.7007956504821777,
      "learning_rate": 4.012181195279788e-06,
      "loss": 1.445,
      "step": 826
    },
    {
      "epoch": 0.8495120698510529,
      "grad_norm": 2.9885144233703613,
      "learning_rate": 4.010277883517321e-06,
      "loss": 1.5516,
      "step": 827
    },
    {
      "epoch": 0.8505392912172574,
      "grad_norm": 2.814607620239258,
      "learning_rate": 4.0083745717548535e-06,
      "loss": 1.5375,
      "step": 828
    },
    {
      "epoch": 0.8515665125834617,
      "grad_norm": 2.9187042713165283,
      "learning_rate": 4.006471259992387e-06,
      "loss": 1.496,
      "step": 829
    },
    {
      "epoch": 0.8525937339496662,
      "grad_norm": 2.7408974170684814,
      "learning_rate": 4.00456794822992e-06,
      "loss": 1.5195,
      "step": 830
    },
    {
      "epoch": 0.8525937339496662,
      "eval_loss": 1.795318365097046,
      "eval_runtime": 37.1058,
      "eval_samples_per_second": 26.95,
      "eval_steps_per_second": 1.132,
      "step": 830
    },
    {
      "epoch": 0.8536209553158706,
      "grad_norm": 2.6824872493743896,
      "learning_rate": 4.002664636467454e-06,
      "loss": 1.5568,
      "step": 831
    },
    {
      "epoch": 0.8546481766820749,
      "grad_norm": 2.7598671913146973,
      "learning_rate": 4.000761324704987e-06,
      "loss": 1.5134,
      "step": 832
    },
    {
      "epoch": 0.8556753980482794,
      "grad_norm": 2.728449583053589,
      "learning_rate": 3.99885801294252e-06,
      "loss": 1.4675,
      "step": 833
    },
    {
      "epoch": 0.8567026194144838,
      "grad_norm": 2.900674343109131,
      "learning_rate": 3.996954701180054e-06,
      "loss": 1.5259,
      "step": 834
    },
    {
      "epoch": 0.8577298407806883,
      "grad_norm": 2.818105697631836,
      "learning_rate": 3.995051389417587e-06,
      "loss": 1.4533,
      "step": 835
    },
    {
      "epoch": 0.8587570621468926,
      "grad_norm": 2.791207790374756,
      "learning_rate": 3.9931480776551204e-06,
      "loss": 1.4692,
      "step": 836
    },
    {
      "epoch": 0.8597842835130971,
      "grad_norm": 2.584268808364868,
      "learning_rate": 3.991244765892654e-06,
      "loss": 1.554,
      "step": 837
    },
    {
      "epoch": 0.8608115048793015,
      "grad_norm": 2.5646660327911377,
      "learning_rate": 3.989341454130186e-06,
      "loss": 1.4782,
      "step": 838
    },
    {
      "epoch": 0.861838726245506,
      "grad_norm": 2.4912099838256836,
      "learning_rate": 3.98743814236772e-06,
      "loss": 1.4494,
      "step": 839
    },
    {
      "epoch": 0.8628659476117103,
      "grad_norm": 2.4632041454315186,
      "learning_rate": 3.985534830605253e-06,
      "loss": 1.4911,
      "step": 840
    },
    {
      "epoch": 0.8628659476117103,
      "eval_loss": 1.7955427169799805,
      "eval_runtime": 37.0188,
      "eval_samples_per_second": 27.013,
      "eval_steps_per_second": 1.135,
      "step": 840
    },
    {
      "epoch": 0.8638931689779148,
      "grad_norm": 2.6388566493988037,
      "learning_rate": 3.983631518842786e-06,
      "loss": 1.5135,
      "step": 841
    },
    {
      "epoch": 0.8649203903441192,
      "grad_norm": 2.7619292736053467,
      "learning_rate": 3.98172820708032e-06,
      "loss": 1.5279,
      "step": 842
    },
    {
      "epoch": 0.8659476117103235,
      "grad_norm": 2.510636806488037,
      "learning_rate": 3.979824895317854e-06,
      "loss": 1.4815,
      "step": 843
    },
    {
      "epoch": 0.866974833076528,
      "grad_norm": 2.6847214698791504,
      "learning_rate": 3.977921583555387e-06,
      "loss": 1.4556,
      "step": 844
    },
    {
      "epoch": 0.8680020544427324,
      "grad_norm": 2.5366628170013428,
      "learning_rate": 3.97601827179292e-06,
      "loss": 1.5297,
      "step": 845
    },
    {
      "epoch": 0.8690292758089369,
      "grad_norm": 2.6018271446228027,
      "learning_rate": 3.974114960030453e-06,
      "loss": 1.4642,
      "step": 846
    },
    {
      "epoch": 0.8700564971751412,
      "grad_norm": 2.612107276916504,
      "learning_rate": 3.9722116482679866e-06,
      "loss": 1.5065,
      "step": 847
    },
    {
      "epoch": 0.8710837185413457,
      "grad_norm": 2.6163978576660156,
      "learning_rate": 3.97030833650552e-06,
      "loss": 1.4913,
      "step": 848
    },
    {
      "epoch": 0.8721109399075501,
      "grad_norm": 2.516697645187378,
      "learning_rate": 3.968405024743053e-06,
      "loss": 1.5006,
      "step": 849
    },
    {
      "epoch": 0.8731381612737545,
      "grad_norm": 2.6012990474700928,
      "learning_rate": 3.966501712980587e-06,
      "loss": 1.4276,
      "step": 850
    },
    {
      "epoch": 0.8731381612737545,
      "eval_loss": 1.807686448097229,
      "eval_runtime": 37.0357,
      "eval_samples_per_second": 27.001,
      "eval_steps_per_second": 1.134,
      "step": 850
    },
    {
      "epoch": 0.8741653826399589,
      "grad_norm": 2.6806578636169434,
      "learning_rate": 3.96459840121812e-06,
      "loss": 1.4849,
      "step": 851
    },
    {
      "epoch": 0.8751926040061633,
      "grad_norm": 2.5139150619506836,
      "learning_rate": 3.962695089455653e-06,
      "loss": 1.4919,
      "step": 852
    },
    {
      "epoch": 0.8762198253723678,
      "grad_norm": 2.779398202896118,
      "learning_rate": 3.960791777693187e-06,
      "loss": 1.5446,
      "step": 853
    },
    {
      "epoch": 0.8772470467385721,
      "grad_norm": 2.7080724239349365,
      "learning_rate": 3.95888846593072e-06,
      "loss": 1.5215,
      "step": 854
    },
    {
      "epoch": 0.8782742681047766,
      "grad_norm": 2.658066749572754,
      "learning_rate": 3.956985154168253e-06,
      "loss": 1.5315,
      "step": 855
    },
    {
      "epoch": 0.879301489470981,
      "grad_norm": 2.6813952922821045,
      "learning_rate": 3.955081842405786e-06,
      "loss": 1.549,
      "step": 856
    },
    {
      "epoch": 0.8803287108371854,
      "grad_norm": 2.744917869567871,
      "learning_rate": 3.953178530643319e-06,
      "loss": 1.5717,
      "step": 857
    },
    {
      "epoch": 0.8813559322033898,
      "grad_norm": 2.869288921356201,
      "learning_rate": 3.951275218880853e-06,
      "loss": 1.5882,
      "step": 858
    },
    {
      "epoch": 0.8823831535695943,
      "grad_norm": 2.633666515350342,
      "learning_rate": 3.949371907118386e-06,
      "loss": 1.4332,
      "step": 859
    },
    {
      "epoch": 0.8834103749357987,
      "grad_norm": 2.7296335697174072,
      "learning_rate": 3.9474685953559194e-06,
      "loss": 1.5349,
      "step": 860
    },
    {
      "epoch": 0.8834103749357987,
      "eval_loss": 1.80648934841156,
      "eval_runtime": 37.0387,
      "eval_samples_per_second": 26.999,
      "eval_steps_per_second": 1.134,
      "step": 860
    },
    {
      "epoch": 0.884437596302003,
      "grad_norm": 3.0827677249908447,
      "learning_rate": 3.945565283593453e-06,
      "loss": 1.5108,
      "step": 861
    },
    {
      "epoch": 0.8854648176682075,
      "grad_norm": 2.81630539894104,
      "learning_rate": 3.943661971830986e-06,
      "loss": 1.493,
      "step": 862
    },
    {
      "epoch": 0.8864920390344119,
      "grad_norm": 2.804215908050537,
      "learning_rate": 3.9417586600685195e-06,
      "loss": 1.5432,
      "step": 863
    },
    {
      "epoch": 0.8875192604006163,
      "grad_norm": 2.9585464000701904,
      "learning_rate": 3.939855348306053e-06,
      "loss": 1.5202,
      "step": 864
    },
    {
      "epoch": 0.8885464817668207,
      "grad_norm": 2.717324733734131,
      "learning_rate": 3.937952036543586e-06,
      "loss": 1.5172,
      "step": 865
    },
    {
      "epoch": 0.8895737031330252,
      "grad_norm": 2.62406325340271,
      "learning_rate": 3.93604872478112e-06,
      "loss": 1.5064,
      "step": 866
    },
    {
      "epoch": 0.8906009244992296,
      "grad_norm": 3.2518985271453857,
      "learning_rate": 3.934145413018653e-06,
      "loss": 1.5159,
      "step": 867
    },
    {
      "epoch": 0.891628145865434,
      "grad_norm": 2.6304733753204346,
      "learning_rate": 3.932242101256186e-06,
      "loss": 1.4723,
      "step": 868
    },
    {
      "epoch": 0.8926553672316384,
      "grad_norm": 2.6788930892944336,
      "learning_rate": 3.93033878949372e-06,
      "loss": 1.4759,
      "step": 869
    },
    {
      "epoch": 0.8936825885978429,
      "grad_norm": 2.8033249378204346,
      "learning_rate": 3.928435477731253e-06,
      "loss": 1.4435,
      "step": 870
    },
    {
      "epoch": 0.8936825885978429,
      "eval_loss": 1.8105196952819824,
      "eval_runtime": 37.062,
      "eval_samples_per_second": 26.982,
      "eval_steps_per_second": 1.133,
      "step": 870
    },
    {
      "epoch": 0.8947098099640473,
      "grad_norm": 2.8569960594177246,
      "learning_rate": 3.9265321659687864e-06,
      "loss": 1.5424,
      "step": 871
    },
    {
      "epoch": 0.8957370313302516,
      "grad_norm": 2.8037121295928955,
      "learning_rate": 3.924628854206319e-06,
      "loss": 1.5368,
      "step": 872
    },
    {
      "epoch": 0.8967642526964561,
      "grad_norm": 2.7444164752960205,
      "learning_rate": 3.922725542443852e-06,
      "loss": 1.5102,
      "step": 873
    },
    {
      "epoch": 0.8977914740626605,
      "grad_norm": 2.8689560890197754,
      "learning_rate": 3.920822230681386e-06,
      "loss": 1.4649,
      "step": 874
    },
    {
      "epoch": 0.8988186954288649,
      "grad_norm": 2.677800178527832,
      "learning_rate": 3.918918918918919e-06,
      "loss": 1.4815,
      "step": 875
    },
    {
      "epoch": 0.8998459167950693,
      "grad_norm": 2.8248631954193115,
      "learning_rate": 3.917015607156452e-06,
      "loss": 1.5193,
      "step": 876
    },
    {
      "epoch": 0.9008731381612738,
      "grad_norm": 2.8272030353546143,
      "learning_rate": 3.915112295393986e-06,
      "loss": 1.4968,
      "step": 877
    },
    {
      "epoch": 0.9019003595274782,
      "grad_norm": 3.0890679359436035,
      "learning_rate": 3.913208983631519e-06,
      "loss": 1.5705,
      "step": 878
    },
    {
      "epoch": 0.9029275808936826,
      "grad_norm": 2.7617108821868896,
      "learning_rate": 3.9113056718690525e-06,
      "loss": 1.5598,
      "step": 879
    },
    {
      "epoch": 0.903954802259887,
      "grad_norm": 2.7723312377929688,
      "learning_rate": 3.909402360106586e-06,
      "loss": 1.4781,
      "step": 880
    },
    {
      "epoch": 0.903954802259887,
      "eval_loss": 1.8097552061080933,
      "eval_runtime": 37.0572,
      "eval_samples_per_second": 26.985,
      "eval_steps_per_second": 1.133,
      "step": 880
    },
    {
      "epoch": 0.9049820236260914,
      "grad_norm": 2.82560133934021,
      "learning_rate": 3.907499048344119e-06,
      "loss": 1.5224,
      "step": 881
    },
    {
      "epoch": 0.9060092449922958,
      "grad_norm": 2.881899118423462,
      "learning_rate": 3.905595736581652e-06,
      "loss": 1.5169,
      "step": 882
    },
    {
      "epoch": 0.9070364663585002,
      "grad_norm": 2.8723788261413574,
      "learning_rate": 3.903692424819186e-06,
      "loss": 1.5225,
      "step": 883
    },
    {
      "epoch": 0.9080636877247047,
      "grad_norm": 2.5932493209838867,
      "learning_rate": 3.901789113056719e-06,
      "loss": 1.4796,
      "step": 884
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.6493358612060547,
      "learning_rate": 3.899885801294253e-06,
      "loss": 1.5324,
      "step": 885
    },
    {
      "epoch": 0.9101181304571135,
      "grad_norm": 2.630830764770508,
      "learning_rate": 3.897982489531786e-06,
      "loss": 1.5113,
      "step": 886
    },
    {
      "epoch": 0.9111453518233179,
      "grad_norm": 2.8183815479278564,
      "learning_rate": 3.896079177769319e-06,
      "loss": 1.5677,
      "step": 887
    },
    {
      "epoch": 0.9121725731895224,
      "grad_norm": 2.811241865158081,
      "learning_rate": 3.894175866006853e-06,
      "loss": 1.5357,
      "step": 888
    },
    {
      "epoch": 0.9131997945557268,
      "grad_norm": 2.6256868839263916,
      "learning_rate": 3.892272554244385e-06,
      "loss": 1.5603,
      "step": 889
    },
    {
      "epoch": 0.9142270159219312,
      "grad_norm": 2.677891254425049,
      "learning_rate": 3.890369242481919e-06,
      "loss": 1.4748,
      "step": 890
    },
    {
      "epoch": 0.9142270159219312,
      "eval_loss": 1.7937747240066528,
      "eval_runtime": 37.3281,
      "eval_samples_per_second": 26.789,
      "eval_steps_per_second": 1.125,
      "step": 890
    },
    {
      "epoch": 0.9152542372881356,
      "grad_norm": 2.639749765396118,
      "learning_rate": 3.888465930719452e-06,
      "loss": 1.5477,
      "step": 891
    },
    {
      "epoch": 0.91628145865434,
      "grad_norm": 2.622887134552002,
      "learning_rate": 3.886562618956985e-06,
      "loss": 1.481,
      "step": 892
    },
    {
      "epoch": 0.9173086800205444,
      "grad_norm": 2.659916639328003,
      "learning_rate": 3.884659307194519e-06,
      "loss": 1.5044,
      "step": 893
    },
    {
      "epoch": 0.9183359013867488,
      "grad_norm": 2.524515151977539,
      "learning_rate": 3.882755995432052e-06,
      "loss": 1.4378,
      "step": 894
    },
    {
      "epoch": 0.9193631227529533,
      "grad_norm": 2.5722432136535645,
      "learning_rate": 3.8808526836695854e-06,
      "loss": 1.5173,
      "step": 895
    },
    {
      "epoch": 0.9203903441191577,
      "grad_norm": 2.6036641597747803,
      "learning_rate": 3.878949371907119e-06,
      "loss": 1.4823,
      "step": 896
    },
    {
      "epoch": 0.9214175654853621,
      "grad_norm": 2.671830415725708,
      "learning_rate": 3.877046060144652e-06,
      "loss": 1.4679,
      "step": 897
    },
    {
      "epoch": 0.9224447868515665,
      "grad_norm": 2.649601936340332,
      "learning_rate": 3.8751427483821855e-06,
      "loss": 1.4651,
      "step": 898
    },
    {
      "epoch": 0.923472008217771,
      "grad_norm": 2.6315200328826904,
      "learning_rate": 3.873239436619718e-06,
      "loss": 1.5029,
      "step": 899
    },
    {
      "epoch": 0.9244992295839753,
      "grad_norm": 2.7867085933685303,
      "learning_rate": 3.871336124857251e-06,
      "loss": 1.4635,
      "step": 900
    },
    {
      "epoch": 0.9244992295839753,
      "eval_loss": 1.8004201650619507,
      "eval_runtime": 37.3928,
      "eval_samples_per_second": 26.743,
      "eval_steps_per_second": 1.123,
      "step": 900
    },
    {
      "epoch": 0.9255264509501797,
      "grad_norm": 2.6903293132781982,
      "learning_rate": 3.869432813094785e-06,
      "loss": 1.5043,
      "step": 901
    },
    {
      "epoch": 0.9265536723163842,
      "grad_norm": 2.6979103088378906,
      "learning_rate": 3.867529501332318e-06,
      "loss": 1.5168,
      "step": 902
    },
    {
      "epoch": 0.9275808936825886,
      "grad_norm": 2.594578266143799,
      "learning_rate": 3.8656261895698515e-06,
      "loss": 1.5141,
      "step": 903
    },
    {
      "epoch": 0.928608115048793,
      "grad_norm": 2.774486780166626,
      "learning_rate": 3.863722877807386e-06,
      "loss": 1.4982,
      "step": 904
    },
    {
      "epoch": 0.9296353364149974,
      "grad_norm": 2.6801764965057373,
      "learning_rate": 3.861819566044919e-06,
      "loss": 1.5294,
      "step": 905
    },
    {
      "epoch": 0.9306625577812019,
      "grad_norm": 2.681169033050537,
      "learning_rate": 3.859916254282452e-06,
      "loss": 1.5687,
      "step": 906
    },
    {
      "epoch": 0.9316897791474062,
      "grad_norm": 2.6592564582824707,
      "learning_rate": 3.858012942519985e-06,
      "loss": 1.4967,
      "step": 907
    },
    {
      "epoch": 0.9327170005136107,
      "grad_norm": 2.636106491088867,
      "learning_rate": 3.856109630757518e-06,
      "loss": 1.5108,
      "step": 908
    },
    {
      "epoch": 0.9337442218798151,
      "grad_norm": 2.6918840408325195,
      "learning_rate": 3.854206318995052e-06,
      "loss": 1.4821,
      "step": 909
    },
    {
      "epoch": 0.9347714432460195,
      "grad_norm": 2.7018930912017822,
      "learning_rate": 3.852303007232585e-06,
      "loss": 1.5336,
      "step": 910
    },
    {
      "epoch": 0.9347714432460195,
      "eval_loss": 1.789487600326538,
      "eval_runtime": 36.9792,
      "eval_samples_per_second": 27.042,
      "eval_steps_per_second": 1.136,
      "step": 910
    },
    {
      "epoch": 0.9357986646122239,
      "grad_norm": 2.5881776809692383,
      "learning_rate": 3.850399695470118e-06,
      "loss": 1.4632,
      "step": 911
    },
    {
      "epoch": 0.9368258859784283,
      "grad_norm": 2.55487322807312,
      "learning_rate": 3.848496383707652e-06,
      "loss": 1.4825,
      "step": 912
    },
    {
      "epoch": 0.9378531073446328,
      "grad_norm": 2.5630884170532227,
      "learning_rate": 3.846593071945185e-06,
      "loss": 1.463,
      "step": 913
    },
    {
      "epoch": 0.9388803287108372,
      "grad_norm": 2.6740293502807617,
      "learning_rate": 3.8446897601827185e-06,
      "loss": 1.4499,
      "step": 914
    },
    {
      "epoch": 0.9399075500770416,
      "grad_norm": 2.641127109527588,
      "learning_rate": 3.842786448420252e-06,
      "loss": 1.5618,
      "step": 915
    },
    {
      "epoch": 0.940934771443246,
      "grad_norm": 2.669304370880127,
      "learning_rate": 3.840883136657785e-06,
      "loss": 1.5725,
      "step": 916
    },
    {
      "epoch": 0.9419619928094505,
      "grad_norm": 2.7020649909973145,
      "learning_rate": 3.838979824895318e-06,
      "loss": 1.5587,
      "step": 917
    },
    {
      "epoch": 0.9429892141756548,
      "grad_norm": 2.6244914531707764,
      "learning_rate": 3.837076513132851e-06,
      "loss": 1.4363,
      "step": 918
    },
    {
      "epoch": 0.9440164355418593,
      "grad_norm": 2.7254905700683594,
      "learning_rate": 3.8351732013703844e-06,
      "loss": 1.5205,
      "step": 919
    },
    {
      "epoch": 0.9450436569080637,
      "grad_norm": 2.4565985202789307,
      "learning_rate": 3.833269889607918e-06,
      "loss": 1.4785,
      "step": 920
    },
    {
      "epoch": 0.9450436569080637,
      "eval_loss": 1.8004498481750488,
      "eval_runtime": 37.0945,
      "eval_samples_per_second": 26.958,
      "eval_steps_per_second": 1.132,
      "step": 920
    },
    {
      "epoch": 0.9460708782742681,
      "grad_norm": 2.8681352138519287,
      "learning_rate": 3.831366577845451e-06,
      "loss": 1.4598,
      "step": 921
    },
    {
      "epoch": 0.9470980996404725,
      "grad_norm": 2.6068029403686523,
      "learning_rate": 3.8294632660829845e-06,
      "loss": 1.4565,
      "step": 922
    },
    {
      "epoch": 0.9481253210066769,
      "grad_norm": 2.764873504638672,
      "learning_rate": 3.827559954320518e-06,
      "loss": 1.4882,
      "step": 923
    },
    {
      "epoch": 0.9491525423728814,
      "grad_norm": 2.655008316040039,
      "learning_rate": 3.825656642558051e-06,
      "loss": 1.5004,
      "step": 924
    },
    {
      "epoch": 0.9501797637390857,
      "grad_norm": 2.7953972816467285,
      "learning_rate": 3.823753330795585e-06,
      "loss": 1.5172,
      "step": 925
    },
    {
      "epoch": 0.9512069851052902,
      "grad_norm": 2.698564052581787,
      "learning_rate": 3.821850019033118e-06,
      "loss": 1.5022,
      "step": 926
    },
    {
      "epoch": 0.9522342064714946,
      "grad_norm": 2.4997222423553467,
      "learning_rate": 3.819946707270651e-06,
      "loss": 1.4373,
      "step": 927
    },
    {
      "epoch": 0.9532614278376991,
      "grad_norm": 2.6832621097564697,
      "learning_rate": 3.818043395508185e-06,
      "loss": 1.49,
      "step": 928
    },
    {
      "epoch": 0.9542886492039034,
      "grad_norm": 2.734297513961792,
      "learning_rate": 3.816140083745718e-06,
      "loss": 1.499,
      "step": 929
    },
    {
      "epoch": 0.9553158705701078,
      "grad_norm": 2.6073720455169678,
      "learning_rate": 3.8142367719832514e-06,
      "loss": 1.5701,
      "step": 930
    },
    {
      "epoch": 0.9553158705701078,
      "eval_loss": 1.8036372661590576,
      "eval_runtime": 37.141,
      "eval_samples_per_second": 26.924,
      "eval_steps_per_second": 1.131,
      "step": 930
    },
    {
      "epoch": 0.9563430919363123,
      "grad_norm": 2.6502838134765625,
      "learning_rate": 3.8123334602207844e-06,
      "loss": 1.4666,
      "step": 931
    },
    {
      "epoch": 0.9573703133025167,
      "grad_norm": 2.6244993209838867,
      "learning_rate": 3.8104301484583177e-06,
      "loss": 1.4659,
      "step": 932
    },
    {
      "epoch": 0.9583975346687211,
      "grad_norm": 2.5014829635620117,
      "learning_rate": 3.808526836695851e-06,
      "loss": 1.4562,
      "step": 933
    },
    {
      "epoch": 0.9594247560349255,
      "grad_norm": 2.6359620094299316,
      "learning_rate": 3.8066235249333845e-06,
      "loss": 1.5223,
      "step": 934
    },
    {
      "epoch": 0.96045197740113,
      "grad_norm": 2.648331642150879,
      "learning_rate": 3.804720213170918e-06,
      "loss": 1.4924,
      "step": 935
    },
    {
      "epoch": 0.9614791987673343,
      "grad_norm": 2.731069564819336,
      "learning_rate": 3.8028169014084508e-06,
      "loss": 1.4959,
      "step": 936
    },
    {
      "epoch": 0.9625064201335388,
      "grad_norm": 2.5789313316345215,
      "learning_rate": 3.800913589645984e-06,
      "loss": 1.5125,
      "step": 937
    },
    {
      "epoch": 0.9635336414997432,
      "grad_norm": 2.6203935146331787,
      "learning_rate": 3.7990102778835175e-06,
      "loss": 1.5064,
      "step": 938
    },
    {
      "epoch": 0.9645608628659477,
      "grad_norm": 2.7040021419525146,
      "learning_rate": 3.797106966121051e-06,
      "loss": 1.5092,
      "step": 939
    },
    {
      "epoch": 0.965588084232152,
      "grad_norm": 2.575338125228882,
      "learning_rate": 3.7952036543585842e-06,
      "loss": 1.5653,
      "step": 940
    },
    {
      "epoch": 0.965588084232152,
      "eval_loss": 1.7945882081985474,
      "eval_runtime": 37.1973,
      "eval_samples_per_second": 26.884,
      "eval_steps_per_second": 1.129,
      "step": 940
    },
    {
      "epoch": 0.9666153055983564,
      "grad_norm": 2.597348690032959,
      "learning_rate": 3.793300342596117e-06,
      "loss": 1.5312,
      "step": 941
    },
    {
      "epoch": 0.9676425269645609,
      "grad_norm": 2.590437650680542,
      "learning_rate": 3.7913970308336505e-06,
      "loss": 1.5347,
      "step": 942
    },
    {
      "epoch": 0.9686697483307652,
      "grad_norm": 2.6253085136413574,
      "learning_rate": 3.789493719071184e-06,
      "loss": 1.5265,
      "step": 943
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 2.5652995109558105,
      "learning_rate": 3.7875904073087177e-06,
      "loss": 1.4738,
      "step": 944
    },
    {
      "epoch": 0.9707241910631741,
      "grad_norm": 2.552037239074707,
      "learning_rate": 3.785687095546251e-06,
      "loss": 1.4311,
      "step": 945
    },
    {
      "epoch": 0.9717514124293786,
      "grad_norm": 2.806792974472046,
      "learning_rate": 3.7837837837837844e-06,
      "loss": 1.4857,
      "step": 946
    },
    {
      "epoch": 0.9727786337955829,
      "grad_norm": 2.696113348007202,
      "learning_rate": 3.7818804720213178e-06,
      "loss": 1.4604,
      "step": 947
    },
    {
      "epoch": 0.9738058551617874,
      "grad_norm": 2.972555637359619,
      "learning_rate": 3.7799771602588507e-06,
      "loss": 1.4328,
      "step": 948
    },
    {
      "epoch": 0.9748330765279918,
      "grad_norm": 2.7260053157806396,
      "learning_rate": 3.778073848496384e-06,
      "loss": 1.4315,
      "step": 949
    },
    {
      "epoch": 0.9758602978941961,
      "grad_norm": 2.501840353012085,
      "learning_rate": 3.7761705367339174e-06,
      "loss": 1.4913,
      "step": 950
    },
    {
      "epoch": 0.9758602978941961,
      "eval_loss": 1.7844858169555664,
      "eval_runtime": 37.1439,
      "eval_samples_per_second": 26.922,
      "eval_steps_per_second": 1.131,
      "step": 950
    },
    {
      "epoch": 0.9768875192604006,
      "grad_norm": 2.704096794128418,
      "learning_rate": 3.7742672249714508e-06,
      "loss": 1.5138,
      "step": 951
    },
    {
      "epoch": 0.977914740626605,
      "grad_norm": 2.7086358070373535,
      "learning_rate": 3.772363913208984e-06,
      "loss": 1.5228,
      "step": 952
    },
    {
      "epoch": 0.9789419619928095,
      "grad_norm": 2.6690590381622314,
      "learning_rate": 3.770460601446517e-06,
      "loss": 1.4828,
      "step": 953
    },
    {
      "epoch": 0.9799691833590138,
      "grad_norm": 2.6134684085845947,
      "learning_rate": 3.7685572896840504e-06,
      "loss": 1.509,
      "step": 954
    },
    {
      "epoch": 0.9809964047252183,
      "grad_norm": 2.7675423622131348,
      "learning_rate": 3.766653977921584e-06,
      "loss": 1.4544,
      "step": 955
    },
    {
      "epoch": 0.9820236260914227,
      "grad_norm": 2.8518829345703125,
      "learning_rate": 3.764750666159117e-06,
      "loss": 1.4042,
      "step": 956
    },
    {
      "epoch": 0.9830508474576272,
      "grad_norm": 2.6791179180145264,
      "learning_rate": 3.7628473543966505e-06,
      "loss": 1.4961,
      "step": 957
    },
    {
      "epoch": 0.9840780688238315,
      "grad_norm": 3.014099597930908,
      "learning_rate": 3.7609440426341835e-06,
      "loss": 1.4582,
      "step": 958
    },
    {
      "epoch": 0.9851052901900359,
      "grad_norm": 2.7100412845611572,
      "learning_rate": 3.759040730871717e-06,
      "loss": 1.5491,
      "step": 959
    },
    {
      "epoch": 0.9861325115562404,
      "grad_norm": 2.565263271331787,
      "learning_rate": 3.75713741910925e-06,
      "loss": 1.4437,
      "step": 960
    },
    {
      "epoch": 0.9861325115562404,
      "eval_loss": 1.7909018993377686,
      "eval_runtime": 37.1828,
      "eval_samples_per_second": 26.894,
      "eval_steps_per_second": 1.13,
      "step": 960
    },
    {
      "epoch": 0.9871597329224447,
      "grad_norm": 2.8322439193725586,
      "learning_rate": 3.7552341073467836e-06,
      "loss": 1.4322,
      "step": 961
    },
    {
      "epoch": 0.9881869542886492,
      "grad_norm": 2.738586902618408,
      "learning_rate": 3.753330795584317e-06,
      "loss": 1.4797,
      "step": 962
    },
    {
      "epoch": 0.9892141756548536,
      "grad_norm": 2.711247205734253,
      "learning_rate": 3.75142748382185e-06,
      "loss": 1.5409,
      "step": 963
    },
    {
      "epoch": 0.9902413970210581,
      "grad_norm": 2.5288846492767334,
      "learning_rate": 3.749524172059384e-06,
      "loss": 1.4408,
      "step": 964
    },
    {
      "epoch": 0.9912686183872624,
      "grad_norm": 2.5907785892486572,
      "learning_rate": 3.747620860296917e-06,
      "loss": 1.4872,
      "step": 965
    },
    {
      "epoch": 0.9922958397534669,
      "grad_norm": 2.4952504634857178,
      "learning_rate": 3.7457175485344504e-06,
      "loss": 1.5277,
      "step": 966
    },
    {
      "epoch": 0.9933230611196713,
      "grad_norm": 2.633167266845703,
      "learning_rate": 3.7438142367719837e-06,
      "loss": 1.5258,
      "step": 967
    },
    {
      "epoch": 0.9943502824858758,
      "grad_norm": 2.6295969486236572,
      "learning_rate": 3.741910925009517e-06,
      "loss": 1.5013,
      "step": 968
    },
    {
      "epoch": 0.9953775038520801,
      "grad_norm": 2.4780771732330322,
      "learning_rate": 3.7400076132470505e-06,
      "loss": 1.4646,
      "step": 969
    },
    {
      "epoch": 0.9964047252182845,
      "grad_norm": 2.585749626159668,
      "learning_rate": 3.7381043014845834e-06,
      "loss": 1.4814,
      "step": 970
    },
    {
      "epoch": 0.9964047252182845,
      "eval_loss": 1.801538109779358,
      "eval_runtime": 37.1396,
      "eval_samples_per_second": 26.925,
      "eval_steps_per_second": 1.131,
      "step": 970
    },
    {
      "epoch": 0.997431946584489,
      "grad_norm": 2.626293182373047,
      "learning_rate": 3.7362009897221168e-06,
      "loss": 1.5071,
      "step": 971
    },
    {
      "epoch": 0.9984591679506933,
      "grad_norm": 2.7129111289978027,
      "learning_rate": 3.73429767795965e-06,
      "loss": 1.4911,
      "step": 972
    },
    {
      "epoch": 0.9994863893168978,
      "grad_norm": 2.623265027999878,
      "learning_rate": 3.7323943661971835e-06,
      "loss": 1.506,
      "step": 973
    },
    {
      "epoch": 1.0005136106831023,
      "grad_norm": 2.497767925262451,
      "learning_rate": 3.730491054434717e-06,
      "loss": 1.4119,
      "step": 974
    },
    {
      "epoch": 1.0015408320493067,
      "grad_norm": 2.8463239669799805,
      "learning_rate": 3.72858774267225e-06,
      "loss": 1.3677,
      "step": 975
    },
    {
      "epoch": 1.002568053415511,
      "grad_norm": 2.643601417541504,
      "learning_rate": 3.726684430909783e-06,
      "loss": 1.4573,
      "step": 976
    },
    {
      "epoch": 1.0035952747817154,
      "grad_norm": 2.751330614089966,
      "learning_rate": 3.7247811191473165e-06,
      "loss": 1.4821,
      "step": 977
    },
    {
      "epoch": 1.00462249614792,
      "grad_norm": 2.714946985244751,
      "learning_rate": 3.72287780738485e-06,
      "loss": 1.4898,
      "step": 978
    },
    {
      "epoch": 1.0056497175141244,
      "grad_norm": 3.066772699356079,
      "learning_rate": 3.7209744956223832e-06,
      "loss": 1.464,
      "step": 979
    },
    {
      "epoch": 1.0066769388803287,
      "grad_norm": 2.818673610687256,
      "learning_rate": 3.719071183859916e-06,
      "loss": 1.4743,
      "step": 980
    },
    {
      "epoch": 1.0066769388803287,
      "eval_loss": 1.7895241975784302,
      "eval_runtime": 37.1443,
      "eval_samples_per_second": 26.922,
      "eval_steps_per_second": 1.131,
      "step": 980
    },
    {
      "epoch": 1.007704160246533,
      "grad_norm": 2.6721136569976807,
      "learning_rate": 3.7171678720974495e-06,
      "loss": 1.3996,
      "step": 981
    },
    {
      "epoch": 1.0087313816127375,
      "grad_norm": 2.757756471633911,
      "learning_rate": 3.715264560334983e-06,
      "loss": 1.5009,
      "step": 982
    },
    {
      "epoch": 1.009758602978942,
      "grad_norm": 2.8042495250701904,
      "learning_rate": 3.7133612485725163e-06,
      "loss": 1.4672,
      "step": 983
    },
    {
      "epoch": 1.0107858243451464,
      "grad_norm": 2.662808418273926,
      "learning_rate": 3.71145793681005e-06,
      "loss": 1.4799,
      "step": 984
    },
    {
      "epoch": 1.0118130457113508,
      "grad_norm": 2.6021053791046143,
      "learning_rate": 3.7095546250475834e-06,
      "loss": 1.4035,
      "step": 985
    },
    {
      "epoch": 1.0128402670775551,
      "grad_norm": 2.7127013206481934,
      "learning_rate": 3.7076513132851168e-06,
      "loss": 1.5192,
      "step": 986
    },
    {
      "epoch": 1.0138674884437597,
      "grad_norm": 2.6402854919433594,
      "learning_rate": 3.7057480015226497e-06,
      "loss": 1.4396,
      "step": 987
    },
    {
      "epoch": 1.014894709809964,
      "grad_norm": 2.5758111476898193,
      "learning_rate": 3.703844689760183e-06,
      "loss": 1.4109,
      "step": 988
    },
    {
      "epoch": 1.0159219311761685,
      "grad_norm": 2.7668240070343018,
      "learning_rate": 3.7019413779977164e-06,
      "loss": 1.4865,
      "step": 989
    },
    {
      "epoch": 1.0169491525423728,
      "grad_norm": 2.840165615081787,
      "learning_rate": 3.70003806623525e-06,
      "loss": 1.4708,
      "step": 990
    },
    {
      "epoch": 1.0169491525423728,
      "eval_loss": 1.7967512607574463,
      "eval_runtime": 37.2322,
      "eval_samples_per_second": 26.858,
      "eval_steps_per_second": 1.128,
      "step": 990
    },
    {
      "epoch": 1.0179763739085772,
      "grad_norm": 2.8218352794647217,
      "learning_rate": 3.698134754472783e-06,
      "loss": 1.4414,
      "step": 991
    },
    {
      "epoch": 1.0190035952747818,
      "grad_norm": 2.742658853530884,
      "learning_rate": 3.696231442710316e-06,
      "loss": 1.5109,
      "step": 992
    },
    {
      "epoch": 1.0200308166409862,
      "grad_norm": 2.6375293731689453,
      "learning_rate": 3.6943281309478495e-06,
      "loss": 1.4433,
      "step": 993
    },
    {
      "epoch": 1.0210580380071905,
      "grad_norm": 2.914246082305908,
      "learning_rate": 3.692424819185383e-06,
      "loss": 1.4656,
      "step": 994
    },
    {
      "epoch": 1.022085259373395,
      "grad_norm": 2.9181385040283203,
      "learning_rate": 3.690521507422916e-06,
      "loss": 1.4735,
      "step": 995
    },
    {
      "epoch": 1.0231124807395995,
      "grad_norm": 2.755748748779297,
      "learning_rate": 3.6886181956604496e-06,
      "loss": 1.4196,
      "step": 996
    },
    {
      "epoch": 1.0241397021058039,
      "grad_norm": 2.9245896339416504,
      "learning_rate": 3.686714883897983e-06,
      "loss": 1.4316,
      "step": 997
    },
    {
      "epoch": 1.0251669234720082,
      "grad_norm": 2.839991569519043,
      "learning_rate": 3.684811572135516e-06,
      "loss": 1.4965,
      "step": 998
    },
    {
      "epoch": 1.0261941448382126,
      "grad_norm": 2.904658079147339,
      "learning_rate": 3.6829082603730492e-06,
      "loss": 1.4962,
      "step": 999
    },
    {
      "epoch": 1.027221366204417,
      "grad_norm": 2.811279535293579,
      "learning_rate": 3.6810049486105826e-06,
      "loss": 1.4514,
      "step": 1000
    },
    {
      "epoch": 1.027221366204417,
      "eval_loss": 1.7917094230651855,
      "eval_runtime": 37.315,
      "eval_samples_per_second": 26.799,
      "eval_steps_per_second": 1.126,
      "step": 1000
    },
    {
      "epoch": 1.0282485875706215,
      "grad_norm": 2.730212926864624,
      "learning_rate": 3.679101636848116e-06,
      "loss": 1.4842,
      "step": 1001
    },
    {
      "epoch": 1.029275808936826,
      "grad_norm": 2.8678042888641357,
      "learning_rate": 3.6771983250856493e-06,
      "loss": 1.4726,
      "step": 1002
    },
    {
      "epoch": 1.0303030303030303,
      "grad_norm": 2.8228330612182617,
      "learning_rate": 3.6752950133231822e-06,
      "loss": 1.4501,
      "step": 1003
    },
    {
      "epoch": 1.0313302516692346,
      "grad_norm": 2.8341903686523438,
      "learning_rate": 3.6733917015607156e-06,
      "loss": 1.4758,
      "step": 1004
    },
    {
      "epoch": 1.0323574730354392,
      "grad_norm": 2.7462005615234375,
      "learning_rate": 3.6714883897982494e-06,
      "loss": 1.4783,
      "step": 1005
    },
    {
      "epoch": 1.0333846944016436,
      "grad_norm": 2.920111894607544,
      "learning_rate": 3.6695850780357828e-06,
      "loss": 1.4393,
      "step": 1006
    },
    {
      "epoch": 1.034411915767848,
      "grad_norm": 2.919542074203491,
      "learning_rate": 3.667681766273316e-06,
      "loss": 1.426,
      "step": 1007
    },
    {
      "epoch": 1.0354391371340523,
      "grad_norm": 2.8268563747406006,
      "learning_rate": 3.6657784545108495e-06,
      "loss": 1.4391,
      "step": 1008
    },
    {
      "epoch": 1.0364663585002567,
      "grad_norm": 2.711895227432251,
      "learning_rate": 3.6638751427483824e-06,
      "loss": 1.5107,
      "step": 1009
    },
    {
      "epoch": 1.0374935798664613,
      "grad_norm": 2.7996091842651367,
      "learning_rate": 3.6619718309859158e-06,
      "loss": 1.4679,
      "step": 1010
    },
    {
      "epoch": 1.0374935798664613,
      "eval_loss": 1.800447940826416,
      "eval_runtime": 36.9218,
      "eval_samples_per_second": 27.084,
      "eval_steps_per_second": 1.138,
      "step": 1010
    },
    {
      "epoch": 1.0385208012326657,
      "grad_norm": 2.8792128562927246,
      "learning_rate": 3.660068519223449e-06,
      "loss": 1.5156,
      "step": 1011
    },
    {
      "epoch": 1.03954802259887,
      "grad_norm": 2.7948005199432373,
      "learning_rate": 3.6581652074609825e-06,
      "loss": 1.4709,
      "step": 1012
    },
    {
      "epoch": 1.0405752439650744,
      "grad_norm": 2.7429840564727783,
      "learning_rate": 3.656261895698516e-06,
      "loss": 1.4414,
      "step": 1013
    },
    {
      "epoch": 1.041602465331279,
      "grad_norm": 2.7512004375457764,
      "learning_rate": 3.6543585839360492e-06,
      "loss": 1.4326,
      "step": 1014
    },
    {
      "epoch": 1.0426296866974833,
      "grad_norm": 2.7879600524902344,
      "learning_rate": 3.652455272173582e-06,
      "loss": 1.4292,
      "step": 1015
    },
    {
      "epoch": 1.0436569080636877,
      "grad_norm": 2.568354368209839,
      "learning_rate": 3.6505519604111155e-06,
      "loss": 1.4383,
      "step": 1016
    },
    {
      "epoch": 1.044684129429892,
      "grad_norm": 2.811404228210449,
      "learning_rate": 3.648648648648649e-06,
      "loss": 1.4573,
      "step": 1017
    },
    {
      "epoch": 1.0457113507960965,
      "grad_norm": 2.909632921218872,
      "learning_rate": 3.6467453368861823e-06,
      "loss": 1.4934,
      "step": 1018
    },
    {
      "epoch": 1.046738572162301,
      "grad_norm": 2.8014776706695557,
      "learning_rate": 3.6448420251237156e-06,
      "loss": 1.4433,
      "step": 1019
    },
    {
      "epoch": 1.0477657935285054,
      "grad_norm": 2.8205742835998535,
      "learning_rate": 3.6429387133612486e-06,
      "loss": 1.4752,
      "step": 1020
    },
    {
      "epoch": 1.0477657935285054,
      "eval_loss": 1.8070552349090576,
      "eval_runtime": 37.1396,
      "eval_samples_per_second": 26.925,
      "eval_steps_per_second": 1.131,
      "step": 1020
    },
    {
      "epoch": 1.0487930148947098,
      "grad_norm": 2.768143653869629,
      "learning_rate": 3.641035401598782e-06,
      "loss": 1.4928,
      "step": 1021
    },
    {
      "epoch": 1.0498202362609141,
      "grad_norm": 2.8027846813201904,
      "learning_rate": 3.6391320898363153e-06,
      "loss": 1.4416,
      "step": 1022
    },
    {
      "epoch": 1.0508474576271187,
      "grad_norm": 2.832526683807373,
      "learning_rate": 3.6372287780738487e-06,
      "loss": 1.4796,
      "step": 1023
    },
    {
      "epoch": 1.051874678993323,
      "grad_norm": 2.8225502967834473,
      "learning_rate": 3.635325466311382e-06,
      "loss": 1.4553,
      "step": 1024
    },
    {
      "epoch": 1.0529019003595275,
      "grad_norm": 2.7793405055999756,
      "learning_rate": 3.633422154548916e-06,
      "loss": 1.4756,
      "step": 1025
    },
    {
      "epoch": 1.0539291217257318,
      "grad_norm": 2.872873544692993,
      "learning_rate": 3.6315188427864487e-06,
      "loss": 1.496,
      "step": 1026
    },
    {
      "epoch": 1.0549563430919364,
      "grad_norm": 2.7574472427368164,
      "learning_rate": 3.629615531023982e-06,
      "loss": 1.464,
      "step": 1027
    },
    {
      "epoch": 1.0559835644581408,
      "grad_norm": 2.7946009635925293,
      "learning_rate": 3.6277122192615155e-06,
      "loss": 1.4149,
      "step": 1028
    },
    {
      "epoch": 1.0570107858243452,
      "grad_norm": 2.764500617980957,
      "learning_rate": 3.625808907499049e-06,
      "loss": 1.4756,
      "step": 1029
    },
    {
      "epoch": 1.0580380071905495,
      "grad_norm": 2.659693479537964,
      "learning_rate": 3.623905595736582e-06,
      "loss": 1.4385,
      "step": 1030
    },
    {
      "epoch": 1.0580380071905495,
      "eval_loss": 1.8091230392456055,
      "eval_runtime": 37.109,
      "eval_samples_per_second": 26.948,
      "eval_steps_per_second": 1.132,
      "step": 1030
    },
    {
      "epoch": 1.059065228556754,
      "grad_norm": 2.7500619888305664,
      "learning_rate": 3.6220022839741156e-06,
      "loss": 1.4334,
      "step": 1031
    },
    {
      "epoch": 1.0600924499229585,
      "grad_norm": 2.7663776874542236,
      "learning_rate": 3.6200989722116485e-06,
      "loss": 1.4194,
      "step": 1032
    },
    {
      "epoch": 1.0611196712891628,
      "grad_norm": 2.820913076400757,
      "learning_rate": 3.618195660449182e-06,
      "loss": 1.4453,
      "step": 1033
    },
    {
      "epoch": 1.0621468926553672,
      "grad_norm": 2.8579916954040527,
      "learning_rate": 3.6162923486867152e-06,
      "loss": 1.4529,
      "step": 1034
    },
    {
      "epoch": 1.0631741140215716,
      "grad_norm": 2.729501247406006,
      "learning_rate": 3.6143890369242486e-06,
      "loss": 1.3907,
      "step": 1035
    },
    {
      "epoch": 1.064201335387776,
      "grad_norm": 2.7084193229675293,
      "learning_rate": 3.612485725161782e-06,
      "loss": 1.4252,
      "step": 1036
    },
    {
      "epoch": 1.0652285567539805,
      "grad_norm": 2.7723402976989746,
      "learning_rate": 3.610582413399315e-06,
      "loss": 1.4339,
      "step": 1037
    },
    {
      "epoch": 1.066255778120185,
      "grad_norm": 2.5925116539001465,
      "learning_rate": 3.6086791016368482e-06,
      "loss": 1.4252,
      "step": 1038
    },
    {
      "epoch": 1.0672829994863893,
      "grad_norm": 2.7101070880889893,
      "learning_rate": 3.6067757898743816e-06,
      "loss": 1.4405,
      "step": 1039
    },
    {
      "epoch": 1.0683102208525936,
      "grad_norm": 2.786576509475708,
      "learning_rate": 3.604872478111915e-06,
      "loss": 1.5049,
      "step": 1040
    },
    {
      "epoch": 1.0683102208525936,
      "eval_loss": 1.8023844957351685,
      "eval_runtime": 37.0823,
      "eval_samples_per_second": 26.967,
      "eval_steps_per_second": 1.133,
      "step": 1040
    },
    {
      "epoch": 1.0693374422187982,
      "grad_norm": 2.8227438926696777,
      "learning_rate": 3.6029691663494483e-06,
      "loss": 1.5062,
      "step": 1041
    },
    {
      "epoch": 1.0703646635850026,
      "grad_norm": 2.722092866897583,
      "learning_rate": 3.6010658545869813e-06,
      "loss": 1.4508,
      "step": 1042
    },
    {
      "epoch": 1.071391884951207,
      "grad_norm": 2.719161033630371,
      "learning_rate": 3.5991625428245146e-06,
      "loss": 1.4791,
      "step": 1043
    },
    {
      "epoch": 1.0724191063174113,
      "grad_norm": 2.8131134510040283,
      "learning_rate": 3.597259231062048e-06,
      "loss": 1.4811,
      "step": 1044
    },
    {
      "epoch": 1.073446327683616,
      "grad_norm": 2.7374107837677,
      "learning_rate": 3.5953559192995818e-06,
      "loss": 1.4063,
      "step": 1045
    },
    {
      "epoch": 1.0744735490498203,
      "grad_norm": 2.7651076316833496,
      "learning_rate": 3.593452607537115e-06,
      "loss": 1.4589,
      "step": 1046
    },
    {
      "epoch": 1.0755007704160247,
      "grad_norm": 2.806370735168457,
      "learning_rate": 3.5915492957746485e-06,
      "loss": 1.5197,
      "step": 1047
    },
    {
      "epoch": 1.076527991782229,
      "grad_norm": 2.8023886680603027,
      "learning_rate": 3.589645984012182e-06,
      "loss": 1.4427,
      "step": 1048
    },
    {
      "epoch": 1.0775552131484334,
      "grad_norm": 2.7536637783050537,
      "learning_rate": 3.587742672249715e-06,
      "loss": 1.4753,
      "step": 1049
    },
    {
      "epoch": 1.078582434514638,
      "grad_norm": 2.6758344173431396,
      "learning_rate": 3.585839360487248e-06,
      "loss": 1.4302,
      "step": 1050
    },
    {
      "epoch": 1.078582434514638,
      "eval_loss": 1.7898778915405273,
      "eval_runtime": 37.04,
      "eval_samples_per_second": 26.998,
      "eval_steps_per_second": 1.134,
      "step": 1050
    },
    {
      "epoch": 1.0796096558808423,
      "grad_norm": 2.688101053237915,
      "learning_rate": 3.5839360487247815e-06,
      "loss": 1.4271,
      "step": 1051
    },
    {
      "epoch": 1.0806368772470467,
      "grad_norm": 2.8667550086975098,
      "learning_rate": 3.582032736962315e-06,
      "loss": 1.4372,
      "step": 1052
    },
    {
      "epoch": 1.081664098613251,
      "grad_norm": 2.9168481826782227,
      "learning_rate": 3.5801294251998483e-06,
      "loss": 1.5063,
      "step": 1053
    },
    {
      "epoch": 1.0826913199794557,
      "grad_norm": 2.743194580078125,
      "learning_rate": 3.578226113437381e-06,
      "loss": 1.4041,
      "step": 1054
    },
    {
      "epoch": 1.08371854134566,
      "grad_norm": 2.6846320629119873,
      "learning_rate": 3.5763228016749146e-06,
      "loss": 1.4427,
      "step": 1055
    },
    {
      "epoch": 1.0847457627118644,
      "grad_norm": 3.1511945724487305,
      "learning_rate": 3.574419489912448e-06,
      "loss": 1.515,
      "step": 1056
    },
    {
      "epoch": 1.0857729840780688,
      "grad_norm": 2.7326550483703613,
      "learning_rate": 3.5725161781499813e-06,
      "loss": 1.4529,
      "step": 1057
    },
    {
      "epoch": 1.0868002054442731,
      "grad_norm": 2.7120213508605957,
      "learning_rate": 3.5706128663875147e-06,
      "loss": 1.4407,
      "step": 1058
    },
    {
      "epoch": 1.0878274268104777,
      "grad_norm": 2.7596473693847656,
      "learning_rate": 3.5687095546250476e-06,
      "loss": 1.4606,
      "step": 1059
    },
    {
      "epoch": 1.088854648176682,
      "grad_norm": 2.9972002506256104,
      "learning_rate": 3.566806242862581e-06,
      "loss": 1.4836,
      "step": 1060
    },
    {
      "epoch": 1.088854648176682,
      "eval_loss": 1.7897655963897705,
      "eval_runtime": 37.0403,
      "eval_samples_per_second": 26.998,
      "eval_steps_per_second": 1.134,
      "step": 1060
    },
    {
      "epoch": 1.0898818695428865,
      "grad_norm": 2.8134756088256836,
      "learning_rate": 3.5649029311001143e-06,
      "loss": 1.4575,
      "step": 1061
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.7311880588531494,
      "learning_rate": 3.5629996193376477e-06,
      "loss": 1.4326,
      "step": 1062
    },
    {
      "epoch": 1.0919363122752954,
      "grad_norm": 2.7033157348632812,
      "learning_rate": 3.561096307575181e-06,
      "loss": 1.4457,
      "step": 1063
    },
    {
      "epoch": 1.0929635336414998,
      "grad_norm": 2.8118138313293457,
      "learning_rate": 3.559192995812714e-06,
      "loss": 1.4869,
      "step": 1064
    },
    {
      "epoch": 1.0939907550077042,
      "grad_norm": 2.8734021186828613,
      "learning_rate": 3.557289684050248e-06,
      "loss": 1.4964,
      "step": 1065
    },
    {
      "epoch": 1.0950179763739085,
      "grad_norm": 2.8652546405792236,
      "learning_rate": 3.555386372287781e-06,
      "loss": 1.4856,
      "step": 1066
    },
    {
      "epoch": 1.0960451977401129,
      "grad_norm": 2.6305630207061768,
      "learning_rate": 3.5534830605253145e-06,
      "loss": 1.49,
      "step": 1067
    },
    {
      "epoch": 1.0970724191063175,
      "grad_norm": 2.9516141414642334,
      "learning_rate": 3.551579748762848e-06,
      "loss": 1.4617,
      "step": 1068
    },
    {
      "epoch": 1.0980996404725218,
      "grad_norm": 2.862701416015625,
      "learning_rate": 3.5496764370003812e-06,
      "loss": 1.4481,
      "step": 1069
    },
    {
      "epoch": 1.0991268618387262,
      "grad_norm": 2.829742193222046,
      "learning_rate": 3.5477731252379146e-06,
      "loss": 1.4309,
      "step": 1070
    },
    {
      "epoch": 1.0991268618387262,
      "eval_loss": 1.7925201654434204,
      "eval_runtime": 37.0708,
      "eval_samples_per_second": 26.975,
      "eval_steps_per_second": 1.133,
      "step": 1070
    },
    {
      "epoch": 1.1001540832049306,
      "grad_norm": 2.786259174346924,
      "learning_rate": 3.5458698134754475e-06,
      "loss": 1.3945,
      "step": 1071
    },
    {
      "epoch": 1.1011813045711352,
      "grad_norm": 2.766643762588501,
      "learning_rate": 3.543966501712981e-06,
      "loss": 1.419,
      "step": 1072
    },
    {
      "epoch": 1.1022085259373395,
      "grad_norm": 2.830751419067383,
      "learning_rate": 3.5420631899505142e-06,
      "loss": 1.4222,
      "step": 1073
    },
    {
      "epoch": 1.103235747303544,
      "grad_norm": 3.1409215927124023,
      "learning_rate": 3.5401598781880476e-06,
      "loss": 1.5262,
      "step": 1074
    },
    {
      "epoch": 1.1042629686697483,
      "grad_norm": 2.678180694580078,
      "learning_rate": 3.538256566425581e-06,
      "loss": 1.3482,
      "step": 1075
    },
    {
      "epoch": 1.1052901900359529,
      "grad_norm": 2.935002088546753,
      "learning_rate": 3.536353254663114e-06,
      "loss": 1.4864,
      "step": 1076
    },
    {
      "epoch": 1.1063174114021572,
      "grad_norm": 2.91631817817688,
      "learning_rate": 3.5344499429006473e-06,
      "loss": 1.4334,
      "step": 1077
    },
    {
      "epoch": 1.1073446327683616,
      "grad_norm": 2.6676135063171387,
      "learning_rate": 3.5325466311381806e-06,
      "loss": 1.4236,
      "step": 1078
    },
    {
      "epoch": 1.108371854134566,
      "grad_norm": 2.779284715652466,
      "learning_rate": 3.530643319375714e-06,
      "loss": 1.463,
      "step": 1079
    },
    {
      "epoch": 1.1093990755007703,
      "grad_norm": 2.9586293697357178,
      "learning_rate": 3.5287400076132474e-06,
      "loss": 1.3953,
      "step": 1080
    },
    {
      "epoch": 1.1093990755007703,
      "eval_loss": 1.8054182529449463,
      "eval_runtime": 37.0591,
      "eval_samples_per_second": 26.984,
      "eval_steps_per_second": 1.133,
      "step": 1080
    },
    {
      "epoch": 1.110426296866975,
      "grad_norm": 2.990781545639038,
      "learning_rate": 3.5268366958507803e-06,
      "loss": 1.4249,
      "step": 1081
    },
    {
      "epoch": 1.1114535182331793,
      "grad_norm": 2.8546760082244873,
      "learning_rate": 3.5249333840883137e-06,
      "loss": 1.4679,
      "step": 1082
    },
    {
      "epoch": 1.1124807395993837,
      "grad_norm": 3.1174535751342773,
      "learning_rate": 3.523030072325847e-06,
      "loss": 1.4587,
      "step": 1083
    },
    {
      "epoch": 1.113507960965588,
      "grad_norm": 3.0766773223876953,
      "learning_rate": 3.5211267605633804e-06,
      "loss": 1.422,
      "step": 1084
    },
    {
      "epoch": 1.1145351823317924,
      "grad_norm": 2.842961311340332,
      "learning_rate": 3.519223448800914e-06,
      "loss": 1.4335,
      "step": 1085
    },
    {
      "epoch": 1.115562403697997,
      "grad_norm": 2.682828426361084,
      "learning_rate": 3.5173201370384475e-06,
      "loss": 1.4401,
      "step": 1086
    },
    {
      "epoch": 1.1165896250642013,
      "grad_norm": 2.9262278079986572,
      "learning_rate": 3.515416825275981e-06,
      "loss": 1.4399,
      "step": 1087
    },
    {
      "epoch": 1.1176168464304057,
      "grad_norm": 3.0017175674438477,
      "learning_rate": 3.513513513513514e-06,
      "loss": 1.4466,
      "step": 1088
    },
    {
      "epoch": 1.11864406779661,
      "grad_norm": 2.6254069805145264,
      "learning_rate": 3.511610201751047e-06,
      "loss": 1.4662,
      "step": 1089
    },
    {
      "epoch": 1.1196712891628147,
      "grad_norm": 2.8206210136413574,
      "learning_rate": 3.5097068899885806e-06,
      "loss": 1.3802,
      "step": 1090
    },
    {
      "epoch": 1.1196712891628147,
      "eval_loss": 1.7929506301879883,
      "eval_runtime": 37.2033,
      "eval_samples_per_second": 26.879,
      "eval_steps_per_second": 1.129,
      "step": 1090
    },
    {
      "epoch": 1.120698510529019,
      "grad_norm": 2.769279718399048,
      "learning_rate": 3.507803578226114e-06,
      "loss": 1.4483,
      "step": 1091
    },
    {
      "epoch": 1.1217257318952234,
      "grad_norm": 2.872576951980591,
      "learning_rate": 3.5059002664636473e-06,
      "loss": 1.5144,
      "step": 1092
    },
    {
      "epoch": 1.1227529532614278,
      "grad_norm": 2.690979480743408,
      "learning_rate": 3.5039969547011802e-06,
      "loss": 1.3938,
      "step": 1093
    },
    {
      "epoch": 1.1237801746276324,
      "grad_norm": 2.7444441318511963,
      "learning_rate": 3.5020936429387136e-06,
      "loss": 1.4295,
      "step": 1094
    },
    {
      "epoch": 1.1248073959938367,
      "grad_norm": 3.083296775817871,
      "learning_rate": 3.500190331176247e-06,
      "loss": 1.4406,
      "step": 1095
    },
    {
      "epoch": 1.125834617360041,
      "grad_norm": 2.7987470626831055,
      "learning_rate": 3.4982870194137803e-06,
      "loss": 1.4937,
      "step": 1096
    },
    {
      "epoch": 1.1268618387262455,
      "grad_norm": 2.9728713035583496,
      "learning_rate": 3.4963837076513137e-06,
      "loss": 1.4253,
      "step": 1097
    },
    {
      "epoch": 1.1278890600924498,
      "grad_norm": 2.9592628479003906,
      "learning_rate": 3.4944803958888466e-06,
      "loss": 1.4339,
      "step": 1098
    },
    {
      "epoch": 1.1289162814586544,
      "grad_norm": 2.6775944232940674,
      "learning_rate": 3.49257708412638e-06,
      "loss": 1.3673,
      "step": 1099
    },
    {
      "epoch": 1.1299435028248588,
      "grad_norm": 2.9349329471588135,
      "learning_rate": 3.4906737723639133e-06,
      "loss": 1.4954,
      "step": 1100
    },
    {
      "epoch": 1.1299435028248588,
      "eval_loss": 1.7869561910629272,
      "eval_runtime": 37.3494,
      "eval_samples_per_second": 26.774,
      "eval_steps_per_second": 1.125,
      "step": 1100
    },
    {
      "epoch": 1.1309707241910631,
      "grad_norm": 3.086017608642578,
      "learning_rate": 3.4887704606014467e-06,
      "loss": 1.5096,
      "step": 1101
    },
    {
      "epoch": 1.1319979455572675,
      "grad_norm": 2.9253454208374023,
      "learning_rate": 3.48686714883898e-06,
      "loss": 1.4466,
      "step": 1102
    },
    {
      "epoch": 1.133025166923472,
      "grad_norm": 2.7312660217285156,
      "learning_rate": 3.484963837076513e-06,
      "loss": 1.4934,
      "step": 1103
    },
    {
      "epoch": 1.1340523882896765,
      "grad_norm": 2.9656598567962646,
      "learning_rate": 3.4830605253140464e-06,
      "loss": 1.4439,
      "step": 1104
    },
    {
      "epoch": 1.1350796096558808,
      "grad_norm": 3.083383321762085,
      "learning_rate": 3.4811572135515797e-06,
      "loss": 1.3937,
      "step": 1105
    },
    {
      "epoch": 1.1361068310220852,
      "grad_norm": 2.7466018199920654,
      "learning_rate": 3.4792539017891135e-06,
      "loss": 1.4244,
      "step": 1106
    },
    {
      "epoch": 1.1371340523882898,
      "grad_norm": 2.755654811859131,
      "learning_rate": 3.477350590026647e-06,
      "loss": 1.3984,
      "step": 1107
    },
    {
      "epoch": 1.1381612737544942,
      "grad_norm": 3.2099268436431885,
      "learning_rate": 3.4754472782641802e-06,
      "loss": 1.4775,
      "step": 1108
    },
    {
      "epoch": 1.1391884951206985,
      "grad_norm": 2.843669891357422,
      "learning_rate": 3.4735439665017136e-06,
      "loss": 1.461,
      "step": 1109
    },
    {
      "epoch": 1.140215716486903,
      "grad_norm": 2.816732883453369,
      "learning_rate": 3.4716406547392465e-06,
      "loss": 1.429,
      "step": 1110
    },
    {
      "epoch": 1.140215716486903,
      "eval_loss": 1.7837378978729248,
      "eval_runtime": 36.9198,
      "eval_samples_per_second": 27.086,
      "eval_steps_per_second": 1.138,
      "step": 1110
    },
    {
      "epoch": 1.1412429378531073,
      "grad_norm": 2.885383367538452,
      "learning_rate": 3.46973734297678e-06,
      "loss": 1.4462,
      "step": 1111
    },
    {
      "epoch": 1.1422701592193119,
      "grad_norm": 2.8761868476867676,
      "learning_rate": 3.4678340312143133e-06,
      "loss": 1.4468,
      "step": 1112
    },
    {
      "epoch": 1.1432973805855162,
      "grad_norm": 2.9817779064178467,
      "learning_rate": 3.4659307194518466e-06,
      "loss": 1.4919,
      "step": 1113
    },
    {
      "epoch": 1.1443246019517206,
      "grad_norm": 2.931018352508545,
      "learning_rate": 3.46402740768938e-06,
      "loss": 1.4836,
      "step": 1114
    },
    {
      "epoch": 1.145351823317925,
      "grad_norm": 2.708481788635254,
      "learning_rate": 3.462124095926913e-06,
      "loss": 1.4092,
      "step": 1115
    },
    {
      "epoch": 1.1463790446841293,
      "grad_norm": 2.8369429111480713,
      "learning_rate": 3.4602207841644463e-06,
      "loss": 1.4135,
      "step": 1116
    },
    {
      "epoch": 1.147406266050334,
      "grad_norm": 2.8598222732543945,
      "learning_rate": 3.4583174724019797e-06,
      "loss": 1.4578,
      "step": 1117
    },
    {
      "epoch": 1.1484334874165383,
      "grad_norm": 2.7808401584625244,
      "learning_rate": 3.456414160639513e-06,
      "loss": 1.5097,
      "step": 1118
    },
    {
      "epoch": 1.1494607087827426,
      "grad_norm": 2.777153253555298,
      "learning_rate": 3.4545108488770464e-06,
      "loss": 1.4687,
      "step": 1119
    },
    {
      "epoch": 1.150487930148947,
      "grad_norm": 2.750174045562744,
      "learning_rate": 3.4526075371145793e-06,
      "loss": 1.4513,
      "step": 1120
    },
    {
      "epoch": 1.150487930148947,
      "eval_loss": 1.8012630939483643,
      "eval_runtime": 37.0093,
      "eval_samples_per_second": 27.02,
      "eval_steps_per_second": 1.135,
      "step": 1120
    },
    {
      "epoch": 1.1515151515151516,
      "grad_norm": 2.7932651042938232,
      "learning_rate": 3.4507042253521127e-06,
      "loss": 1.4574,
      "step": 1121
    },
    {
      "epoch": 1.152542372881356,
      "grad_norm": 2.8366539478302,
      "learning_rate": 3.448800913589646e-06,
      "loss": 1.5257,
      "step": 1122
    },
    {
      "epoch": 1.1535695942475603,
      "grad_norm": 2.7671217918395996,
      "learning_rate": 3.4468976018271794e-06,
      "loss": 1.4256,
      "step": 1123
    },
    {
      "epoch": 1.1545968156137647,
      "grad_norm": 2.845287799835205,
      "learning_rate": 3.4449942900647128e-06,
      "loss": 1.5317,
      "step": 1124
    },
    {
      "epoch": 1.1556240369799693,
      "grad_norm": 2.6672773361206055,
      "learning_rate": 3.443090978302246e-06,
      "loss": 1.4782,
      "step": 1125
    },
    {
      "epoch": 1.1566512583461737,
      "grad_norm": 2.844306230545044,
      "learning_rate": 3.44118766653978e-06,
      "loss": 1.4771,
      "step": 1126
    },
    {
      "epoch": 1.157678479712378,
      "grad_norm": 2.723320484161377,
      "learning_rate": 3.439284354777313e-06,
      "loss": 1.4568,
      "step": 1127
    },
    {
      "epoch": 1.1587057010785824,
      "grad_norm": 2.725198984146118,
      "learning_rate": 3.4373810430148462e-06,
      "loss": 1.4682,
      "step": 1128
    },
    {
      "epoch": 1.1597329224447868,
      "grad_norm": 2.8702471256256104,
      "learning_rate": 3.4354777312523796e-06,
      "loss": 1.4079,
      "step": 1129
    },
    {
      "epoch": 1.1607601438109914,
      "grad_norm": 2.924783945083618,
      "learning_rate": 3.433574419489913e-06,
      "loss": 1.437,
      "step": 1130
    },
    {
      "epoch": 1.1607601438109914,
      "eval_loss": 1.817325234413147,
      "eval_runtime": 37.0048,
      "eval_samples_per_second": 27.024,
      "eval_steps_per_second": 1.135,
      "step": 1130
    },
    {
      "epoch": 1.1617873651771957,
      "grad_norm": 2.8339531421661377,
      "learning_rate": 3.4316711077274463e-06,
      "loss": 1.469,
      "step": 1131
    },
    {
      "epoch": 1.1628145865434,
      "grad_norm": 2.880077362060547,
      "learning_rate": 3.4297677959649792e-06,
      "loss": 1.4364,
      "step": 1132
    },
    {
      "epoch": 1.1638418079096045,
      "grad_norm": 2.96932315826416,
      "learning_rate": 3.4278644842025126e-06,
      "loss": 1.4868,
      "step": 1133
    },
    {
      "epoch": 1.1648690292758088,
      "grad_norm": 2.643693685531616,
      "learning_rate": 3.425961172440046e-06,
      "loss": 1.428,
      "step": 1134
    },
    {
      "epoch": 1.1658962506420134,
      "grad_norm": 2.9450795650482178,
      "learning_rate": 3.4240578606775793e-06,
      "loss": 1.5155,
      "step": 1135
    },
    {
      "epoch": 1.1669234720082178,
      "grad_norm": 2.7274043560028076,
      "learning_rate": 3.4221545489151127e-06,
      "loss": 1.4486,
      "step": 1136
    },
    {
      "epoch": 1.1679506933744221,
      "grad_norm": 2.7501749992370605,
      "learning_rate": 3.4202512371526456e-06,
      "loss": 1.4254,
      "step": 1137
    },
    {
      "epoch": 1.1689779147406265,
      "grad_norm": 3.0331339836120605,
      "learning_rate": 3.418347925390179e-06,
      "loss": 1.5338,
      "step": 1138
    },
    {
      "epoch": 1.170005136106831,
      "grad_norm": 2.9705629348754883,
      "learning_rate": 3.4164446136277124e-06,
      "loss": 1.474,
      "step": 1139
    },
    {
      "epoch": 1.1710323574730355,
      "grad_norm": 2.76196026802063,
      "learning_rate": 3.4145413018652457e-06,
      "loss": 1.4505,
      "step": 1140
    },
    {
      "epoch": 1.1710323574730355,
      "eval_loss": 1.8043485879898071,
      "eval_runtime": 37.0258,
      "eval_samples_per_second": 27.008,
      "eval_steps_per_second": 1.134,
      "step": 1140
    },
    {
      "epoch": 1.1720595788392398,
      "grad_norm": 2.6624460220336914,
      "learning_rate": 3.412637990102779e-06,
      "loss": 1.4739,
      "step": 1141
    },
    {
      "epoch": 1.1730868002054442,
      "grad_norm": 2.7865853309631348,
      "learning_rate": 3.4107346783403125e-06,
      "loss": 1.4923,
      "step": 1142
    },
    {
      "epoch": 1.1741140215716488,
      "grad_norm": 2.782196044921875,
      "learning_rate": 3.4088313665778454e-06,
      "loss": 1.4041,
      "step": 1143
    },
    {
      "epoch": 1.1751412429378532,
      "grad_norm": 2.7796125411987305,
      "learning_rate": 3.4069280548153788e-06,
      "loss": 1.4446,
      "step": 1144
    },
    {
      "epoch": 1.1761684643040575,
      "grad_norm": 2.8180930614471436,
      "learning_rate": 3.405024743052912e-06,
      "loss": 1.424,
      "step": 1145
    },
    {
      "epoch": 1.177195685670262,
      "grad_norm": 2.8180930614471436,
      "learning_rate": 3.405024743052912e-06,
      "loss": 1.4457,
      "step": 1146
    },
    {
      "epoch": 1.1782229070364663,
      "grad_norm": 2.922241449356079,
      "learning_rate": 3.403121431290446e-06,
      "loss": 1.4162,
      "step": 1147
    },
    {
      "epoch": 1.1792501284026708,
      "grad_norm": 2.850069761276245,
      "learning_rate": 3.4012181195279793e-06,
      "loss": 1.432,
      "step": 1148
    },
    {
      "epoch": 1.1802773497688752,
      "grad_norm": 2.863086462020874,
      "learning_rate": 3.3993148077655126e-06,
      "loss": 1.442,
      "step": 1149
    },
    {
      "epoch": 1.1813045711350796,
      "grad_norm": 2.9545702934265137,
      "learning_rate": 3.3974114960030456e-06,
      "loss": 1.3931,
      "step": 1150
    },
    {
      "epoch": 1.1813045711350796,
      "eval_loss": 1.812690258026123,
      "eval_runtime": 37.0328,
      "eval_samples_per_second": 27.003,
      "eval_steps_per_second": 1.134,
      "step": 1150
    },
    {
      "epoch": 1.182331792501284,
      "grad_norm": 2.9074196815490723,
      "learning_rate": 3.395508184240579e-06,
      "loss": 1.4199,
      "step": 1151
    },
    {
      "epoch": 1.1833590138674885,
      "grad_norm": 3.1387181282043457,
      "learning_rate": 3.3936048724781123e-06,
      "loss": 1.509,
      "step": 1152
    },
    {
      "epoch": 1.184386235233693,
      "grad_norm": 3.0677523612976074,
      "learning_rate": 3.3917015607156457e-06,
      "loss": 1.3952,
      "step": 1153
    },
    {
      "epoch": 1.1854134565998973,
      "grad_norm": 2.8462986946105957,
      "learning_rate": 3.389798248953179e-06,
      "loss": 1.415,
      "step": 1154
    },
    {
      "epoch": 1.1864406779661016,
      "grad_norm": 2.9588911533355713,
      "learning_rate": 3.387894937190712e-06,
      "loss": 1.464,
      "step": 1155
    },
    {
      "epoch": 1.1874678993323062,
      "grad_norm": 2.7406837940216064,
      "learning_rate": 3.3859916254282453e-06,
      "loss": 1.4604,
      "step": 1156
    },
    {
      "epoch": 1.1884951206985106,
      "grad_norm": 2.837615489959717,
      "learning_rate": 3.3840883136657787e-06,
      "loss": 1.5108,
      "step": 1157
    },
    {
      "epoch": 1.189522342064715,
      "grad_norm": 2.6457152366638184,
      "learning_rate": 3.382185001903312e-06,
      "loss": 1.4247,
      "step": 1158
    },
    {
      "epoch": 1.1905495634309193,
      "grad_norm": 2.887178659439087,
      "learning_rate": 3.3802816901408454e-06,
      "loss": 1.4609,
      "step": 1159
    },
    {
      "epoch": 1.1915767847971237,
      "grad_norm": 2.6568970680236816,
      "learning_rate": 3.3783783783783788e-06,
      "loss": 1.4393,
      "step": 1160
    },
    {
      "epoch": 1.1915767847971237,
      "eval_loss": 1.7956509590148926,
      "eval_runtime": 37.0256,
      "eval_samples_per_second": 27.008,
      "eval_steps_per_second": 1.134,
      "step": 1160
    },
    {
      "epoch": 1.1926040061633283,
      "grad_norm": 2.731114625930786,
      "learning_rate": 3.3764750666159117e-06,
      "loss": 1.4399,
      "step": 1161
    },
    {
      "epoch": 1.1936312275295327,
      "grad_norm": 2.8750929832458496,
      "learning_rate": 3.374571754853445e-06,
      "loss": 1.4318,
      "step": 1162
    },
    {
      "epoch": 1.194658448895737,
      "grad_norm": 2.697359085083008,
      "learning_rate": 3.3726684430909784e-06,
      "loss": 1.3884,
      "step": 1163
    },
    {
      "epoch": 1.1956856702619414,
      "grad_norm": 2.6850969791412354,
      "learning_rate": 3.370765131328512e-06,
      "loss": 1.4104,
      "step": 1164
    },
    {
      "epoch": 1.1967128916281458,
      "grad_norm": 2.739121913909912,
      "learning_rate": 3.368861819566045e-06,
      "loss": 1.4227,
      "step": 1165
    },
    {
      "epoch": 1.1977401129943503,
      "grad_norm": 2.7206380367279053,
      "learning_rate": 3.366958507803578e-06,
      "loss": 1.4587,
      "step": 1166
    },
    {
      "epoch": 1.1987673343605547,
      "grad_norm": 2.651898145675659,
      "learning_rate": 3.365055196041112e-06,
      "loss": 1.4081,
      "step": 1167
    },
    {
      "epoch": 1.199794555726759,
      "grad_norm": 2.7718563079833984,
      "learning_rate": 3.3631518842786452e-06,
      "loss": 1.4835,
      "step": 1168
    },
    {
      "epoch": 1.2008217770929634,
      "grad_norm": 2.771209716796875,
      "learning_rate": 3.3612485725161786e-06,
      "loss": 1.4036,
      "step": 1169
    },
    {
      "epoch": 1.201848998459168,
      "grad_norm": 2.6938540935516357,
      "learning_rate": 3.359345260753712e-06,
      "loss": 1.4987,
      "step": 1170
    },
    {
      "epoch": 1.201848998459168,
      "eval_loss": 1.784244179725647,
      "eval_runtime": 37.0121,
      "eval_samples_per_second": 27.018,
      "eval_steps_per_second": 1.135,
      "step": 1170
    },
    {
      "epoch": 1.2028762198253724,
      "grad_norm": 2.884730339050293,
      "learning_rate": 3.3574419489912453e-06,
      "loss": 1.4778,
      "step": 1171
    },
    {
      "epoch": 1.2039034411915768,
      "grad_norm": 3.020632266998291,
      "learning_rate": 3.3555386372287783e-06,
      "loss": 1.5133,
      "step": 1172
    },
    {
      "epoch": 1.2049306625577811,
      "grad_norm": 2.8026037216186523,
      "learning_rate": 3.3536353254663116e-06,
      "loss": 1.405,
      "step": 1173
    },
    {
      "epoch": 1.2059578839239857,
      "grad_norm": 2.763852596282959,
      "learning_rate": 3.351732013703845e-06,
      "loss": 1.4437,
      "step": 1174
    },
    {
      "epoch": 1.20698510529019,
      "grad_norm": 2.897768259048462,
      "learning_rate": 3.3498287019413784e-06,
      "loss": 1.4548,
      "step": 1175
    },
    {
      "epoch": 1.2080123266563945,
      "grad_norm": 2.8483922481536865,
      "learning_rate": 3.3479253901789117e-06,
      "loss": 1.4734,
      "step": 1176
    },
    {
      "epoch": 1.2090395480225988,
      "grad_norm": 2.8507916927337646,
      "learning_rate": 3.346022078416445e-06,
      "loss": 1.378,
      "step": 1177
    },
    {
      "epoch": 1.2100667693888032,
      "grad_norm": 2.74048113822937,
      "learning_rate": 3.344118766653978e-06,
      "loss": 1.4493,
      "step": 1178
    },
    {
      "epoch": 1.2110939907550078,
      "grad_norm": 2.8064157962799072,
      "learning_rate": 3.3422154548915114e-06,
      "loss": 1.4221,
      "step": 1179
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 2.721090793609619,
      "learning_rate": 3.3403121431290448e-06,
      "loss": 1.4603,
      "step": 1180
    },
    {
      "epoch": 1.2121212121212122,
      "eval_loss": 1.7907730340957642,
      "eval_runtime": 37.0168,
      "eval_samples_per_second": 27.015,
      "eval_steps_per_second": 1.135,
      "step": 1180
    },
    {
      "epoch": 1.2131484334874165,
      "grad_norm": 2.706892967224121,
      "learning_rate": 3.338408831366578e-06,
      "loss": 1.3903,
      "step": 1181
    },
    {
      "epoch": 1.214175654853621,
      "grad_norm": 3.0193893909454346,
      "learning_rate": 3.3365055196041115e-06,
      "loss": 1.4715,
      "step": 1182
    },
    {
      "epoch": 1.2152028762198253,
      "grad_norm": 2.8713300228118896,
      "learning_rate": 3.3346022078416444e-06,
      "loss": 1.5017,
      "step": 1183
    },
    {
      "epoch": 1.2162300975860298,
      "grad_norm": 2.7250728607177734,
      "learning_rate": 3.3326988960791778e-06,
      "loss": 1.4242,
      "step": 1184
    },
    {
      "epoch": 1.2172573189522342,
      "grad_norm": 2.9367101192474365,
      "learning_rate": 3.330795584316711e-06,
      "loss": 1.3563,
      "step": 1185
    },
    {
      "epoch": 1.2182845403184386,
      "grad_norm": 3.059776544570923,
      "learning_rate": 3.3288922725542445e-06,
      "loss": 1.4334,
      "step": 1186
    },
    {
      "epoch": 1.219311761684643,
      "grad_norm": 2.996394157409668,
      "learning_rate": 3.326988960791778e-06,
      "loss": 1.4457,
      "step": 1187
    },
    {
      "epoch": 1.2203389830508475,
      "grad_norm": 2.8958983421325684,
      "learning_rate": 3.3250856490293117e-06,
      "loss": 1.4904,
      "step": 1188
    },
    {
      "epoch": 1.221366204417052,
      "grad_norm": 2.9497668743133545,
      "learning_rate": 3.323182337266845e-06,
      "loss": 1.4836,
      "step": 1189
    },
    {
      "epoch": 1.2223934257832563,
      "grad_norm": 2.87447452545166,
      "learning_rate": 3.321279025504378e-06,
      "loss": 1.4664,
      "step": 1190
    },
    {
      "epoch": 1.2223934257832563,
      "eval_loss": 1.7865233421325684,
      "eval_runtime": 37.1144,
      "eval_samples_per_second": 26.944,
      "eval_steps_per_second": 1.132,
      "step": 1190
    },
    {
      "epoch": 1.2234206471494606,
      "grad_norm": 2.7893576622009277,
      "learning_rate": 3.3193757137419113e-06,
      "loss": 1.4584,
      "step": 1191
    },
    {
      "epoch": 1.2244478685156652,
      "grad_norm": 2.791938543319702,
      "learning_rate": 3.3174724019794447e-06,
      "loss": 1.4974,
      "step": 1192
    },
    {
      "epoch": 1.2254750898818696,
      "grad_norm": 2.7816781997680664,
      "learning_rate": 3.315569090216978e-06,
      "loss": 1.4103,
      "step": 1193
    },
    {
      "epoch": 1.226502311248074,
      "grad_norm": 2.938931703567505,
      "learning_rate": 3.3136657784545114e-06,
      "loss": 1.498,
      "step": 1194
    },
    {
      "epoch": 1.2275295326142783,
      "grad_norm": 2.7500035762786865,
      "learning_rate": 3.3117624666920443e-06,
      "loss": 1.4044,
      "step": 1195
    },
    {
      "epoch": 1.2285567539804827,
      "grad_norm": 2.758352756500244,
      "learning_rate": 3.3098591549295777e-06,
      "loss": 1.3968,
      "step": 1196
    },
    {
      "epoch": 1.2295839753466873,
      "grad_norm": 3.144857406616211,
      "learning_rate": 3.307955843167111e-06,
      "loss": 1.459,
      "step": 1197
    },
    {
      "epoch": 1.2306111967128917,
      "grad_norm": 2.777985095977783,
      "learning_rate": 3.3060525314046444e-06,
      "loss": 1.4419,
      "step": 1198
    },
    {
      "epoch": 1.231638418079096,
      "grad_norm": 2.812570810317993,
      "learning_rate": 3.304149219642178e-06,
      "loss": 1.446,
      "step": 1199
    },
    {
      "epoch": 1.2326656394453004,
      "grad_norm": 2.6672797203063965,
      "learning_rate": 3.3022459078797107e-06,
      "loss": 1.4615,
      "step": 1200
    },
    {
      "epoch": 1.2326656394453004,
      "eval_loss": 1.7857756614685059,
      "eval_runtime": 37.313,
      "eval_samples_per_second": 26.8,
      "eval_steps_per_second": 1.126,
      "step": 1200
    },
    {
      "epoch": 1.2336928608115048,
      "grad_norm": 2.8981986045837402,
      "learning_rate": 3.300342596117244e-06,
      "loss": 1.4418,
      "step": 1201
    },
    {
      "epoch": 1.2347200821777093,
      "grad_norm": 2.7738826274871826,
      "learning_rate": 3.2984392843547775e-06,
      "loss": 1.3926,
      "step": 1202
    },
    {
      "epoch": 1.2357473035439137,
      "grad_norm": 2.735461950302124,
      "learning_rate": 3.296535972592311e-06,
      "loss": 1.3843,
      "step": 1203
    },
    {
      "epoch": 1.236774524910118,
      "grad_norm": 2.965693950653076,
      "learning_rate": 3.294632660829844e-06,
      "loss": 1.4248,
      "step": 1204
    },
    {
      "epoch": 1.2378017462763227,
      "grad_norm": 2.850008010864258,
      "learning_rate": 3.292729349067377e-06,
      "loss": 1.4277,
      "step": 1205
    },
    {
      "epoch": 1.238828967642527,
      "grad_norm": 2.823054790496826,
      "learning_rate": 3.2908260373049105e-06,
      "loss": 1.4151,
      "step": 1206
    },
    {
      "epoch": 1.2398561890087314,
      "grad_norm": 2.7468149662017822,
      "learning_rate": 3.288922725542444e-06,
      "loss": 1.4661,
      "step": 1207
    },
    {
      "epoch": 1.2408834103749358,
      "grad_norm": 2.7379629611968994,
      "learning_rate": 3.2870194137799776e-06,
      "loss": 1.4219,
      "step": 1208
    },
    {
      "epoch": 1.2419106317411401,
      "grad_norm": 2.7809221744537354,
      "learning_rate": 3.285116102017511e-06,
      "loss": 1.4002,
      "step": 1209
    },
    {
      "epoch": 1.2429378531073447,
      "grad_norm": 2.9209654331207275,
      "learning_rate": 3.2832127902550444e-06,
      "loss": 1.4646,
      "step": 1210
    },
    {
      "epoch": 1.2429378531073447,
      "eval_loss": 1.786431908607483,
      "eval_runtime": 36.9218,
      "eval_samples_per_second": 27.084,
      "eval_steps_per_second": 1.138,
      "step": 1210
    },
    {
      "epoch": 1.243965074473549,
      "grad_norm": 2.9019596576690674,
      "learning_rate": 3.2813094784925777e-06,
      "loss": 1.4526,
      "step": 1211
    },
    {
      "epoch": 1.2449922958397535,
      "grad_norm": 2.7218263149261475,
      "learning_rate": 3.2794061667301107e-06,
      "loss": 1.3911,
      "step": 1212
    },
    {
      "epoch": 1.2460195172059578,
      "grad_norm": 2.8646154403686523,
      "learning_rate": 3.277502854967644e-06,
      "loss": 1.4305,
      "step": 1213
    },
    {
      "epoch": 1.2470467385721622,
      "grad_norm": 2.9065611362457275,
      "learning_rate": 3.2755995432051774e-06,
      "loss": 1.4239,
      "step": 1214
    },
    {
      "epoch": 1.2480739599383668,
      "grad_norm": 2.7945196628570557,
      "learning_rate": 3.2736962314427107e-06,
      "loss": 1.4321,
      "step": 1215
    },
    {
      "epoch": 1.2491011813045712,
      "grad_norm": 2.644244909286499,
      "learning_rate": 3.271792919680244e-06,
      "loss": 1.4411,
      "step": 1216
    },
    {
      "epoch": 1.2501284026707755,
      "grad_norm": 2.9058117866516113,
      "learning_rate": 3.269889607917777e-06,
      "loss": 1.4942,
      "step": 1217
    },
    {
      "epoch": 1.25115562403698,
      "grad_norm": 2.8242623805999756,
      "learning_rate": 3.2679862961553104e-06,
      "loss": 1.4033,
      "step": 1218
    },
    {
      "epoch": 1.2521828454031843,
      "grad_norm": 2.859428644180298,
      "learning_rate": 3.2660829843928438e-06,
      "loss": 1.4298,
      "step": 1219
    },
    {
      "epoch": 1.2532100667693888,
      "grad_norm": 2.7864537239074707,
      "learning_rate": 3.264179672630377e-06,
      "loss": 1.4827,
      "step": 1220
    },
    {
      "epoch": 1.2532100667693888,
      "eval_loss": 1.7798999547958374,
      "eval_runtime": 37.0394,
      "eval_samples_per_second": 26.998,
      "eval_steps_per_second": 1.134,
      "step": 1220
    },
    {
      "epoch": 1.2542372881355932,
      "grad_norm": 2.7938263416290283,
      "learning_rate": 3.2622763608679105e-06,
      "loss": 1.4412,
      "step": 1221
    },
    {
      "epoch": 1.2552645095017976,
      "grad_norm": 2.811281204223633,
      "learning_rate": 3.2603730491054434e-06,
      "loss": 1.4139,
      "step": 1222
    },
    {
      "epoch": 1.2562917308680022,
      "grad_norm": 2.7984230518341064,
      "learning_rate": 3.258469737342977e-06,
      "loss": 1.4205,
      "step": 1223
    },
    {
      "epoch": 1.2573189522342065,
      "grad_norm": 2.7457568645477295,
      "learning_rate": 3.25656642558051e-06,
      "loss": 1.396,
      "step": 1224
    },
    {
      "epoch": 1.258346173600411,
      "grad_norm": 2.84529972076416,
      "learning_rate": 3.2546631138180435e-06,
      "loss": 1.4192,
      "step": 1225
    },
    {
      "epoch": 1.2593733949666153,
      "grad_norm": 2.8183720111846924,
      "learning_rate": 3.252759802055577e-06,
      "loss": 1.4104,
      "step": 1226
    },
    {
      "epoch": 1.2604006163328196,
      "grad_norm": 2.8872454166412354,
      "learning_rate": 3.25085649029311e-06,
      "loss": 1.4256,
      "step": 1227
    },
    {
      "epoch": 1.2614278376990242,
      "grad_norm": 2.935094118118286,
      "learning_rate": 3.248953178530644e-06,
      "loss": 1.4558,
      "step": 1228
    },
    {
      "epoch": 1.2624550590652286,
      "grad_norm": 2.69989013671875,
      "learning_rate": 3.247049866768177e-06,
      "loss": 1.4167,
      "step": 1229
    },
    {
      "epoch": 1.263482280431433,
      "grad_norm": 2.781031847000122,
      "learning_rate": 3.2451465550057103e-06,
      "loss": 1.3745,
      "step": 1230
    },
    {
      "epoch": 1.263482280431433,
      "eval_loss": 1.7843073606491089,
      "eval_runtime": 37.0672,
      "eval_samples_per_second": 26.978,
      "eval_steps_per_second": 1.133,
      "step": 1230
    },
    {
      "epoch": 1.2645095017976373,
      "grad_norm": 2.974919080734253,
      "learning_rate": 3.2432432432432437e-06,
      "loss": 1.4221,
      "step": 1231
    },
    {
      "epoch": 1.2655367231638417,
      "grad_norm": 3.095144271850586,
      "learning_rate": 3.241339931480777e-06,
      "loss": 1.4539,
      "step": 1232
    },
    {
      "epoch": 1.2665639445300463,
      "grad_norm": 3.0127196311950684,
      "learning_rate": 3.2394366197183104e-06,
      "loss": 1.4559,
      "step": 1233
    },
    {
      "epoch": 1.2675911658962506,
      "grad_norm": 2.7474801540374756,
      "learning_rate": 3.2375333079558434e-06,
      "loss": 1.4177,
      "step": 1234
    },
    {
      "epoch": 1.268618387262455,
      "grad_norm": 2.743516445159912,
      "learning_rate": 3.2356299961933767e-06,
      "loss": 1.4471,
      "step": 1235
    },
    {
      "epoch": 1.2696456086286596,
      "grad_norm": 2.7801740169525146,
      "learning_rate": 3.23372668443091e-06,
      "loss": 1.4588,
      "step": 1236
    },
    {
      "epoch": 1.270672829994864,
      "grad_norm": 2.697547435760498,
      "learning_rate": 3.2318233726684435e-06,
      "loss": 1.4857,
      "step": 1237
    },
    {
      "epoch": 1.2717000513610683,
      "grad_norm": 2.646890640258789,
      "learning_rate": 3.229920060905977e-06,
      "loss": 1.4068,
      "step": 1238
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 2.694401979446411,
      "learning_rate": 3.2280167491435098e-06,
      "loss": 1.4148,
      "step": 1239
    },
    {
      "epoch": 1.273754494093477,
      "grad_norm": 2.6139938831329346,
      "learning_rate": 3.226113437381043e-06,
      "loss": 1.4214,
      "step": 1240
    },
    {
      "epoch": 1.273754494093477,
      "eval_loss": 1.7777915000915527,
      "eval_runtime": 37.1013,
      "eval_samples_per_second": 26.953,
      "eval_steps_per_second": 1.132,
      "step": 1240
    },
    {
      "epoch": 1.2747817154596817,
      "grad_norm": 2.7611944675445557,
      "learning_rate": 3.2242101256185765e-06,
      "loss": 1.4344,
      "step": 1241
    },
    {
      "epoch": 1.275808936825886,
      "grad_norm": 2.7328102588653564,
      "learning_rate": 3.22230681385611e-06,
      "loss": 1.4891,
      "step": 1242
    },
    {
      "epoch": 1.2768361581920904,
      "grad_norm": 2.822169065475464,
      "learning_rate": 3.220403502093643e-06,
      "loss": 1.4328,
      "step": 1243
    },
    {
      "epoch": 1.2778633795582948,
      "grad_norm": 2.878005027770996,
      "learning_rate": 3.218500190331176e-06,
      "loss": 1.4779,
      "step": 1244
    },
    {
      "epoch": 1.2788906009244991,
      "grad_norm": 2.725376844406128,
      "learning_rate": 3.2165968785687095e-06,
      "loss": 1.424,
      "step": 1245
    },
    {
      "epoch": 1.2799178222907037,
      "grad_norm": 2.7598440647125244,
      "learning_rate": 3.214693566806243e-06,
      "loss": 1.4533,
      "step": 1246
    },
    {
      "epoch": 1.280945043656908,
      "grad_norm": 2.8355820178985596,
      "learning_rate": 3.2127902550437762e-06,
      "loss": 1.4433,
      "step": 1247
    },
    {
      "epoch": 1.2819722650231125,
      "grad_norm": 2.7520253658294678,
      "learning_rate": 3.21088694328131e-06,
      "loss": 1.4318,
      "step": 1248
    },
    {
      "epoch": 1.2829994863893168,
      "grad_norm": 2.7606403827667236,
      "learning_rate": 3.2089836315188434e-06,
      "loss": 1.4242,
      "step": 1249
    },
    {
      "epoch": 1.2840267077555212,
      "grad_norm": 2.784301280975342,
      "learning_rate": 3.2070803197563767e-06,
      "loss": 1.4253,
      "step": 1250
    },
    {
      "epoch": 1.2840267077555212,
      "eval_loss": 1.7868103981018066,
      "eval_runtime": 37.0911,
      "eval_samples_per_second": 26.961,
      "eval_steps_per_second": 1.132,
      "step": 1250
    },
    {
      "epoch": 1.2850539291217258,
      "grad_norm": 2.8295185565948486,
      "learning_rate": 3.2051770079939097e-06,
      "loss": 1.4432,
      "step": 1251
    },
    {
      "epoch": 1.2860811504879301,
      "grad_norm": 2.8488125801086426,
      "learning_rate": 3.203273696231443e-06,
      "loss": 1.4714,
      "step": 1252
    },
    {
      "epoch": 1.2871083718541345,
      "grad_norm": 2.771249294281006,
      "learning_rate": 3.2013703844689764e-06,
      "loss": 1.4093,
      "step": 1253
    },
    {
      "epoch": 1.288135593220339,
      "grad_norm": 2.782425880432129,
      "learning_rate": 3.1994670727065098e-06,
      "loss": 1.4597,
      "step": 1254
    },
    {
      "epoch": 1.2891628145865435,
      "grad_norm": 2.88934588432312,
      "learning_rate": 3.197563760944043e-06,
      "loss": 1.4805,
      "step": 1255
    },
    {
      "epoch": 1.2901900359527478,
      "grad_norm": 2.8694138526916504,
      "learning_rate": 3.195660449181576e-06,
      "loss": 1.4699,
      "step": 1256
    },
    {
      "epoch": 1.2912172573189522,
      "grad_norm": 2.799614667892456,
      "learning_rate": 3.1937571374191094e-06,
      "loss": 1.4044,
      "step": 1257
    },
    {
      "epoch": 1.2922444786851566,
      "grad_norm": 3.001746416091919,
      "learning_rate": 3.191853825656643e-06,
      "loss": 1.4305,
      "step": 1258
    },
    {
      "epoch": 1.2932717000513612,
      "grad_norm": 2.8572680950164795,
      "learning_rate": 3.189950513894176e-06,
      "loss": 1.4252,
      "step": 1259
    },
    {
      "epoch": 1.2942989214175655,
      "grad_norm": 2.942213535308838,
      "learning_rate": 3.1880472021317095e-06,
      "loss": 1.427,
      "step": 1260
    },
    {
      "epoch": 1.2942989214175655,
      "eval_loss": 1.7929412126541138,
      "eval_runtime": 37.087,
      "eval_samples_per_second": 26.964,
      "eval_steps_per_second": 1.132,
      "step": 1260
    },
    {
      "epoch": 1.29532614278377,
      "grad_norm": 2.893172264099121,
      "learning_rate": 3.1861438903692425e-06,
      "loss": 1.379,
      "step": 1261
    },
    {
      "epoch": 1.2963533641499743,
      "grad_norm": 2.9168691635131836,
      "learning_rate": 3.184240578606776e-06,
      "loss": 1.4124,
      "step": 1262
    },
    {
      "epoch": 1.2973805855161786,
      "grad_norm": 2.7513487339019775,
      "learning_rate": 3.182337266844309e-06,
      "loss": 1.4343,
      "step": 1263
    },
    {
      "epoch": 1.2984078068823832,
      "grad_norm": 2.8485500812530518,
      "learning_rate": 3.1804339550818426e-06,
      "loss": 1.467,
      "step": 1264
    },
    {
      "epoch": 1.2994350282485876,
      "grad_norm": 2.6620397567749023,
      "learning_rate": 3.178530643319376e-06,
      "loss": 1.4191,
      "step": 1265
    },
    {
      "epoch": 1.300462249614792,
      "grad_norm": 2.659393072128296,
      "learning_rate": 3.176627331556909e-06,
      "loss": 1.3934,
      "step": 1266
    },
    {
      "epoch": 1.3014894709809965,
      "grad_norm": 2.8579442501068115,
      "learning_rate": 3.1747240197944422e-06,
      "loss": 1.4399,
      "step": 1267
    },
    {
      "epoch": 1.3025166923472007,
      "grad_norm": 2.7417430877685547,
      "learning_rate": 3.172820708031976e-06,
      "loss": 1.3956,
      "step": 1268
    },
    {
      "epoch": 1.3035439137134053,
      "grad_norm": 3.1082868576049805,
      "learning_rate": 3.1709173962695094e-06,
      "loss": 1.4768,
      "step": 1269
    },
    {
      "epoch": 1.3045711350796096,
      "grad_norm": 3.099637985229492,
      "learning_rate": 3.1690140845070427e-06,
      "loss": 1.4679,
      "step": 1270
    },
    {
      "epoch": 1.3045711350796096,
      "eval_loss": 1.7878395318984985,
      "eval_runtime": 37.0795,
      "eval_samples_per_second": 26.969,
      "eval_steps_per_second": 1.133,
      "step": 1270
    },
    {
      "epoch": 1.305598356445814,
      "grad_norm": 2.8106865882873535,
      "learning_rate": 3.167110772744576e-06,
      "loss": 1.4257,
      "step": 1271
    },
    {
      "epoch": 1.3066255778120186,
      "grad_norm": 2.7419369220733643,
      "learning_rate": 3.1652074609821095e-06,
      "loss": 1.4197,
      "step": 1272
    },
    {
      "epoch": 1.307652799178223,
      "grad_norm": 2.7077760696411133,
      "learning_rate": 3.1633041492196424e-06,
      "loss": 1.4188,
      "step": 1273
    },
    {
      "epoch": 1.3086800205444273,
      "grad_norm": 2.6611413955688477,
      "learning_rate": 3.1614008374571758e-06,
      "loss": 1.4456,
      "step": 1274
    },
    {
      "epoch": 1.3097072419106317,
      "grad_norm": 2.8461341857910156,
      "learning_rate": 3.159497525694709e-06,
      "loss": 1.4544,
      "step": 1275
    },
    {
      "epoch": 1.310734463276836,
      "grad_norm": 2.7263684272766113,
      "learning_rate": 3.1575942139322425e-06,
      "loss": 1.4177,
      "step": 1276
    },
    {
      "epoch": 1.3117616846430407,
      "grad_norm": 2.8571529388427734,
      "learning_rate": 3.155690902169776e-06,
      "loss": 1.4473,
      "step": 1277
    },
    {
      "epoch": 1.312788906009245,
      "grad_norm": 2.914841651916504,
      "learning_rate": 3.1537875904073088e-06,
      "loss": 1.4117,
      "step": 1278
    },
    {
      "epoch": 1.3138161273754494,
      "grad_norm": 2.7753429412841797,
      "learning_rate": 3.151884278644842e-06,
      "loss": 1.4359,
      "step": 1279
    },
    {
      "epoch": 1.3148433487416538,
      "grad_norm": 2.624159097671509,
      "learning_rate": 3.1499809668823755e-06,
      "loss": 1.4072,
      "step": 1280
    },
    {
      "epoch": 1.3148433487416538,
      "eval_loss": 1.7820065021514893,
      "eval_runtime": 37.0679,
      "eval_samples_per_second": 26.978,
      "eval_steps_per_second": 1.133,
      "step": 1280
    },
    {
      "epoch": 1.3158705701078581,
      "grad_norm": 2.742490291595459,
      "learning_rate": 3.148077655119909e-06,
      "loss": 1.4681,
      "step": 1281
    },
    {
      "epoch": 1.3168977914740627,
      "grad_norm": 2.6784660816192627,
      "learning_rate": 3.1461743433574422e-06,
      "loss": 1.406,
      "step": 1282
    },
    {
      "epoch": 1.317925012840267,
      "grad_norm": 2.746730089187622,
      "learning_rate": 3.144271031594975e-06,
      "loss": 1.4538,
      "step": 1283
    },
    {
      "epoch": 1.3189522342064715,
      "grad_norm": 2.746652603149414,
      "learning_rate": 3.1423677198325085e-06,
      "loss": 1.4574,
      "step": 1284
    },
    {
      "epoch": 1.319979455572676,
      "grad_norm": 2.842672109603882,
      "learning_rate": 3.140464408070042e-06,
      "loss": 1.414,
      "step": 1285
    },
    {
      "epoch": 1.3210066769388804,
      "grad_norm": 2.7332921028137207,
      "learning_rate": 3.1385610963075753e-06,
      "loss": 1.4115,
      "step": 1286
    },
    {
      "epoch": 1.3220338983050848,
      "grad_norm": 2.6771159172058105,
      "learning_rate": 3.1366577845451086e-06,
      "loss": 1.4023,
      "step": 1287
    },
    {
      "epoch": 1.3230611196712891,
      "grad_norm": 2.689605951309204,
      "learning_rate": 3.134754472782642e-06,
      "loss": 1.3897,
      "step": 1288
    },
    {
      "epoch": 1.3240883410374935,
      "grad_norm": 2.7937700748443604,
      "learning_rate": 3.1328511610201758e-06,
      "loss": 1.3783,
      "step": 1289
    },
    {
      "epoch": 1.325115562403698,
      "grad_norm": 2.643263816833496,
      "learning_rate": 3.1309478492577087e-06,
      "loss": 1.378,
      "step": 1290
    },
    {
      "epoch": 1.325115562403698,
      "eval_loss": 1.7850252389907837,
      "eval_runtime": 37.0816,
      "eval_samples_per_second": 26.968,
      "eval_steps_per_second": 1.133,
      "step": 1290
    },
    {
      "epoch": 1.3261427837699025,
      "grad_norm": 2.7702224254608154,
      "learning_rate": 3.129044537495242e-06,
      "loss": 1.4236,
      "step": 1291
    },
    {
      "epoch": 1.3271700051361068,
      "grad_norm": 2.783445358276367,
      "learning_rate": 3.1271412257327754e-06,
      "loss": 1.4454,
      "step": 1292
    },
    {
      "epoch": 1.3281972265023112,
      "grad_norm": 2.556509494781494,
      "learning_rate": 3.125237913970309e-06,
      "loss": 1.3998,
      "step": 1293
    },
    {
      "epoch": 1.3292244478685156,
      "grad_norm": 2.706526041030884,
      "learning_rate": 3.123334602207842e-06,
      "loss": 1.3781,
      "step": 1294
    },
    {
      "epoch": 1.3302516692347202,
      "grad_norm": 2.9328322410583496,
      "learning_rate": 3.121431290445375e-06,
      "loss": 1.4747,
      "step": 1295
    },
    {
      "epoch": 1.3312788906009245,
      "grad_norm": 3.0525310039520264,
      "learning_rate": 3.1195279786829085e-06,
      "loss": 1.5197,
      "step": 1296
    },
    {
      "epoch": 1.332306111967129,
      "grad_norm": 2.715700149536133,
      "learning_rate": 3.117624666920442e-06,
      "loss": 1.4059,
      "step": 1297
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 2.7344141006469727,
      "learning_rate": 3.115721355157975e-06,
      "loss": 1.4101,
      "step": 1298
    },
    {
      "epoch": 1.3343605546995376,
      "grad_norm": 2.746990203857422,
      "learning_rate": 3.1138180433955085e-06,
      "loss": 1.3625,
      "step": 1299
    },
    {
      "epoch": 1.3353877760657422,
      "grad_norm": 2.817866086959839,
      "learning_rate": 3.1119147316330415e-06,
      "loss": 1.4414,
      "step": 1300
    },
    {
      "epoch": 1.3353877760657422,
      "eval_loss": 1.7872337102890015,
      "eval_runtime": 37.2584,
      "eval_samples_per_second": 26.84,
      "eval_steps_per_second": 1.127,
      "step": 1300
    },
    {
      "epoch": 1.3364149974319466,
      "grad_norm": 2.814664125442505,
      "learning_rate": 3.110011419870575e-06,
      "loss": 1.4768,
      "step": 1301
    },
    {
      "epoch": 1.337442218798151,
      "grad_norm": 2.7719714641571045,
      "learning_rate": 3.1081081081081082e-06,
      "loss": 1.4128,
      "step": 1302
    },
    {
      "epoch": 1.3384694401643555,
      "grad_norm": 2.8300364017486572,
      "learning_rate": 3.1062047963456416e-06,
      "loss": 1.4419,
      "step": 1303
    },
    {
      "epoch": 1.33949666153056,
      "grad_norm": 2.943674087524414,
      "learning_rate": 3.104301484583175e-06,
      "loss": 1.4841,
      "step": 1304
    },
    {
      "epoch": 1.3405238828967643,
      "grad_norm": 2.7583727836608887,
      "learning_rate": 3.1023981728207083e-06,
      "loss": 1.4674,
      "step": 1305
    },
    {
      "epoch": 1.3415511042629686,
      "grad_norm": 2.716261386871338,
      "learning_rate": 3.1004948610582412e-06,
      "loss": 1.4292,
      "step": 1306
    },
    {
      "epoch": 1.342578325629173,
      "grad_norm": 2.8233978748321533,
      "learning_rate": 3.0985915492957746e-06,
      "loss": 1.4208,
      "step": 1307
    },
    {
      "epoch": 1.3436055469953776,
      "grad_norm": 2.829420804977417,
      "learning_rate": 3.096688237533308e-06,
      "loss": 1.3828,
      "step": 1308
    },
    {
      "epoch": 1.344632768361582,
      "grad_norm": 2.791999101638794,
      "learning_rate": 3.0947849257708418e-06,
      "loss": 1.4751,
      "step": 1309
    },
    {
      "epoch": 1.3456599897277863,
      "grad_norm": 2.691257953643799,
      "learning_rate": 3.092881614008375e-06,
      "loss": 1.423,
      "step": 1310
    },
    {
      "epoch": 1.3456599897277863,
      "eval_loss": 1.7816330194473267,
      "eval_runtime": 36.9065,
      "eval_samples_per_second": 27.095,
      "eval_steps_per_second": 1.138,
      "step": 1310
    },
    {
      "epoch": 1.3466872110939907,
      "grad_norm": 2.8363473415374756,
      "learning_rate": 3.0909783022459085e-06,
      "loss": 1.4771,
      "step": 1311
    },
    {
      "epoch": 1.347714432460195,
      "grad_norm": 2.720055103302002,
      "learning_rate": 3.0890749904834414e-06,
      "loss": 1.4126,
      "step": 1312
    },
    {
      "epoch": 1.3487416538263997,
      "grad_norm": 2.8794236183166504,
      "learning_rate": 3.0871716787209748e-06,
      "loss": 1.4649,
      "step": 1313
    },
    {
      "epoch": 1.349768875192604,
      "grad_norm": 2.760241746902466,
      "learning_rate": 3.085268366958508e-06,
      "loss": 1.4427,
      "step": 1314
    },
    {
      "epoch": 1.3507960965588084,
      "grad_norm": 3.031611680984497,
      "learning_rate": 3.0833650551960415e-06,
      "loss": 1.4476,
      "step": 1315
    },
    {
      "epoch": 1.3518233179250128,
      "grad_norm": 2.7548346519470215,
      "learning_rate": 3.081461743433575e-06,
      "loss": 1.4036,
      "step": 1316
    },
    {
      "epoch": 1.3528505392912171,
      "grad_norm": 2.7737832069396973,
      "learning_rate": 3.079558431671108e-06,
      "loss": 1.4209,
      "step": 1317
    },
    {
      "epoch": 1.3538777606574217,
      "grad_norm": 2.580578565597534,
      "learning_rate": 3.077655119908641e-06,
      "loss": 1.3899,
      "step": 1318
    },
    {
      "epoch": 1.354904982023626,
      "grad_norm": 2.7918219566345215,
      "learning_rate": 3.0757518081461745e-06,
      "loss": 1.4205,
      "step": 1319
    },
    {
      "epoch": 1.3559322033898304,
      "grad_norm": 2.807173490524292,
      "learning_rate": 3.073848496383708e-06,
      "loss": 1.4371,
      "step": 1320
    },
    {
      "epoch": 1.3559322033898304,
      "eval_loss": 1.7834100723266602,
      "eval_runtime": 37.0369,
      "eval_samples_per_second": 27.0,
      "eval_steps_per_second": 1.134,
      "step": 1320
    },
    {
      "epoch": 1.356959424756035,
      "grad_norm": 2.856569290161133,
      "learning_rate": 3.0719451846212413e-06,
      "loss": 1.4054,
      "step": 1321
    },
    {
      "epoch": 1.3579866461222394,
      "grad_norm": 2.850403308868408,
      "learning_rate": 3.0700418728587746e-06,
      "loss": 1.4248,
      "step": 1322
    },
    {
      "epoch": 1.3590138674884438,
      "grad_norm": 2.8618571758270264,
      "learning_rate": 3.0681385610963076e-06,
      "loss": 1.4066,
      "step": 1323
    },
    {
      "epoch": 1.3600410888546481,
      "grad_norm": 3.0298640727996826,
      "learning_rate": 3.066235249333841e-06,
      "loss": 1.4479,
      "step": 1324
    },
    {
      "epoch": 1.3610683102208525,
      "grad_norm": 2.9725687503814697,
      "learning_rate": 3.0643319375713743e-06,
      "loss": 1.4837,
      "step": 1325
    },
    {
      "epoch": 1.362095531587057,
      "grad_norm": 2.8397116661071777,
      "learning_rate": 3.0624286258089076e-06,
      "loss": 1.4248,
      "step": 1326
    },
    {
      "epoch": 1.3631227529532615,
      "grad_norm": 2.7257447242736816,
      "learning_rate": 3.060525314046441e-06,
      "loss": 1.4047,
      "step": 1327
    },
    {
      "epoch": 1.3641499743194658,
      "grad_norm": 2.832534074783325,
      "learning_rate": 3.058622002283974e-06,
      "loss": 1.4491,
      "step": 1328
    },
    {
      "epoch": 1.3651771956856702,
      "grad_norm": 2.6238863468170166,
      "learning_rate": 3.0567186905215077e-06,
      "loss": 1.457,
      "step": 1329
    },
    {
      "epoch": 1.3662044170518746,
      "grad_norm": 2.7993323802948,
      "learning_rate": 3.054815378759041e-06,
      "loss": 1.458,
      "step": 1330
    },
    {
      "epoch": 1.3662044170518746,
      "eval_loss": 1.7857885360717773,
      "eval_runtime": 37.0668,
      "eval_samples_per_second": 26.978,
      "eval_steps_per_second": 1.133,
      "step": 1330
    },
    {
      "epoch": 1.3672316384180792,
      "grad_norm": 2.798299789428711,
      "learning_rate": 3.0529120669965745e-06,
      "loss": 1.4539,
      "step": 1331
    },
    {
      "epoch": 1.3682588597842835,
      "grad_norm": 2.7995524406433105,
      "learning_rate": 3.051008755234108e-06,
      "loss": 1.3929,
      "step": 1332
    },
    {
      "epoch": 1.3692860811504879,
      "grad_norm": 3.026519298553467,
      "learning_rate": 3.049105443471641e-06,
      "loss": 1.4,
      "step": 1333
    },
    {
      "epoch": 1.3703133025166925,
      "grad_norm": 3.040426731109619,
      "learning_rate": 3.0472021317091745e-06,
      "loss": 1.5383,
      "step": 1334
    },
    {
      "epoch": 1.3713405238828968,
      "grad_norm": 2.8738269805908203,
      "learning_rate": 3.0452988199467075e-06,
      "loss": 1.4581,
      "step": 1335
    },
    {
      "epoch": 1.3723677452491012,
      "grad_norm": 2.895745038986206,
      "learning_rate": 3.043395508184241e-06,
      "loss": 1.4508,
      "step": 1336
    },
    {
      "epoch": 1.3733949666153056,
      "grad_norm": 2.872739315032959,
      "learning_rate": 3.041492196421774e-06,
      "loss": 1.413,
      "step": 1337
    },
    {
      "epoch": 1.37442218798151,
      "grad_norm": 2.717970132827759,
      "learning_rate": 3.0395888846593076e-06,
      "loss": 1.4433,
      "step": 1338
    },
    {
      "epoch": 1.3754494093477145,
      "grad_norm": 2.799251079559326,
      "learning_rate": 3.037685572896841e-06,
      "loss": 1.4237,
      "step": 1339
    },
    {
      "epoch": 1.376476630713919,
      "grad_norm": 2.8966126441955566,
      "learning_rate": 3.035782261134374e-06,
      "loss": 1.4386,
      "step": 1340
    },
    {
      "epoch": 1.376476630713919,
      "eval_loss": 1.795745611190796,
      "eval_runtime": 37.0421,
      "eval_samples_per_second": 26.996,
      "eval_steps_per_second": 1.134,
      "step": 1340
    },
    {
      "epoch": 1.3775038520801233,
      "grad_norm": 3.0871410369873047,
      "learning_rate": 3.0338789493719072e-06,
      "loss": 1.4416,
      "step": 1341
    },
    {
      "epoch": 1.3785310734463276,
      "grad_norm": 2.829460382461548,
      "learning_rate": 3.0319756376094406e-06,
      "loss": 1.4361,
      "step": 1342
    },
    {
      "epoch": 1.379558294812532,
      "grad_norm": 3.0431859493255615,
      "learning_rate": 3.030072325846974e-06,
      "loss": 1.4431,
      "step": 1343
    },
    {
      "epoch": 1.3805855161787366,
      "grad_norm": 2.966308116912842,
      "learning_rate": 3.0281690140845073e-06,
      "loss": 1.4022,
      "step": 1344
    },
    {
      "epoch": 1.381612737544941,
      "grad_norm": 2.976332664489746,
      "learning_rate": 3.0262657023220403e-06,
      "loss": 1.4481,
      "step": 1345
    },
    {
      "epoch": 1.3826399589111453,
      "grad_norm": 2.716144561767578,
      "learning_rate": 3.0243623905595736e-06,
      "loss": 1.4437,
      "step": 1346
    },
    {
      "epoch": 1.3836671802773497,
      "grad_norm": 2.913194417953491,
      "learning_rate": 3.022459078797107e-06,
      "loss": 1.4414,
      "step": 1347
    },
    {
      "epoch": 1.384694401643554,
      "grad_norm": 2.9889321327209473,
      "learning_rate": 3.0205557670346404e-06,
      "loss": 1.4333,
      "step": 1348
    },
    {
      "epoch": 1.3857216230097587,
      "grad_norm": 2.657104730606079,
      "learning_rate": 3.018652455272174e-06,
      "loss": 1.3999,
      "step": 1349
    },
    {
      "epoch": 1.386748844375963,
      "grad_norm": 2.988452196121216,
      "learning_rate": 3.0167491435097075e-06,
      "loss": 1.5192,
      "step": 1350
    },
    {
      "epoch": 1.386748844375963,
      "eval_loss": 1.7963600158691406,
      "eval_runtime": 37.062,
      "eval_samples_per_second": 26.982,
      "eval_steps_per_second": 1.133,
      "step": 1350
    },
    {
      "epoch": 1.3877760657421674,
      "grad_norm": 2.9976272583007812,
      "learning_rate": 3.014845831747241e-06,
      "loss": 1.4288,
      "step": 1351
    },
    {
      "epoch": 1.388803287108372,
      "grad_norm": 2.9255707263946533,
      "learning_rate": 3.012942519984774e-06,
      "loss": 1.4047,
      "step": 1352
    },
    {
      "epoch": 1.3898305084745763,
      "grad_norm": 2.995906352996826,
      "learning_rate": 3.011039208222307e-06,
      "loss": 1.4151,
      "step": 1353
    },
    {
      "epoch": 1.3908577298407807,
      "grad_norm": 2.745427370071411,
      "learning_rate": 3.0091358964598405e-06,
      "loss": 1.3784,
      "step": 1354
    },
    {
      "epoch": 1.391884951206985,
      "grad_norm": 2.8206543922424316,
      "learning_rate": 3.007232584697374e-06,
      "loss": 1.4307,
      "step": 1355
    },
    {
      "epoch": 1.3929121725731894,
      "grad_norm": 2.7975900173187256,
      "learning_rate": 3.0053292729349073e-06,
      "loss": 1.4363,
      "step": 1356
    },
    {
      "epoch": 1.393939393939394,
      "grad_norm": 2.8240561485290527,
      "learning_rate": 3.00342596117244e-06,
      "loss": 1.4284,
      "step": 1357
    },
    {
      "epoch": 1.3949666153055984,
      "grad_norm": 2.8556740283966064,
      "learning_rate": 3.0015226494099736e-06,
      "loss": 1.4252,
      "step": 1358
    },
    {
      "epoch": 1.3959938366718028,
      "grad_norm": 2.828031539916992,
      "learning_rate": 2.999619337647507e-06,
      "loss": 1.4068,
      "step": 1359
    },
    {
      "epoch": 1.3970210580380071,
      "grad_norm": 2.8133797645568848,
      "learning_rate": 2.9977160258850403e-06,
      "loss": 1.4114,
      "step": 1360
    },
    {
      "epoch": 1.3970210580380071,
      "eval_loss": 1.7951425313949585,
      "eval_runtime": 37.0796,
      "eval_samples_per_second": 26.969,
      "eval_steps_per_second": 1.133,
      "step": 1360
    },
    {
      "epoch": 1.3980482794042115,
      "grad_norm": 2.8920187950134277,
      "learning_rate": 2.9958127141225736e-06,
      "loss": 1.3862,
      "step": 1361
    },
    {
      "epoch": 1.399075500770416,
      "grad_norm": 2.936598777770996,
      "learning_rate": 2.9939094023601066e-06,
      "loss": 1.4565,
      "step": 1362
    },
    {
      "epoch": 1.4001027221366205,
      "grad_norm": 2.8578219413757324,
      "learning_rate": 2.99200609059764e-06,
      "loss": 1.4291,
      "step": 1363
    },
    {
      "epoch": 1.4011299435028248,
      "grad_norm": 2.8153412342071533,
      "learning_rate": 2.9901027788351733e-06,
      "loss": 1.3765,
      "step": 1364
    },
    {
      "epoch": 1.4021571648690292,
      "grad_norm": 2.8321216106414795,
      "learning_rate": 2.9881994670727067e-06,
      "loss": 1.3914,
      "step": 1365
    },
    {
      "epoch": 1.4031843862352336,
      "grad_norm": 2.7363576889038086,
      "learning_rate": 2.98629615531024e-06,
      "loss": 1.4091,
      "step": 1366
    },
    {
      "epoch": 1.4042116076014381,
      "grad_norm": 2.8216500282287598,
      "learning_rate": 2.984392843547773e-06,
      "loss": 1.398,
      "step": 1367
    },
    {
      "epoch": 1.4052388289676425,
      "grad_norm": 2.8419880867004395,
      "learning_rate": 2.9824895317853063e-06,
      "loss": 1.423,
      "step": 1368
    },
    {
      "epoch": 1.4062660503338469,
      "grad_norm": 2.7820470333099365,
      "learning_rate": 2.98058622002284e-06,
      "loss": 1.4205,
      "step": 1369
    },
    {
      "epoch": 1.4072932717000515,
      "grad_norm": 2.8953442573547363,
      "learning_rate": 2.9786829082603735e-06,
      "loss": 1.4667,
      "step": 1370
    },
    {
      "epoch": 1.4072932717000515,
      "eval_loss": 1.7979086637496948,
      "eval_runtime": 37.0854,
      "eval_samples_per_second": 26.965,
      "eval_steps_per_second": 1.133,
      "step": 1370
    },
    {
      "epoch": 1.4083204930662558,
      "grad_norm": 2.8208730220794678,
      "learning_rate": 2.976779596497907e-06,
      "loss": 1.4541,
      "step": 1371
    },
    {
      "epoch": 1.4093477144324602,
      "grad_norm": 2.7347970008850098,
      "learning_rate": 2.97487628473544e-06,
      "loss": 1.4673,
      "step": 1372
    },
    {
      "epoch": 1.4103749357986646,
      "grad_norm": 2.8528873920440674,
      "learning_rate": 2.9729729729729736e-06,
      "loss": 1.4267,
      "step": 1373
    },
    {
      "epoch": 1.411402157164869,
      "grad_norm": 2.8298377990722656,
      "learning_rate": 2.9710696612105065e-06,
      "loss": 1.4713,
      "step": 1374
    },
    {
      "epoch": 1.4124293785310735,
      "grad_norm": 2.847799301147461,
      "learning_rate": 2.96916634944804e-06,
      "loss": 1.4141,
      "step": 1375
    },
    {
      "epoch": 1.413456599897278,
      "grad_norm": 2.8330819606781006,
      "learning_rate": 2.9672630376855732e-06,
      "loss": 1.4706,
      "step": 1376
    },
    {
      "epoch": 1.4144838212634823,
      "grad_norm": 2.9275906085968018,
      "learning_rate": 2.9653597259231066e-06,
      "loss": 1.4371,
      "step": 1377
    },
    {
      "epoch": 1.4155110426296866,
      "grad_norm": 2.7579028606414795,
      "learning_rate": 2.96345641416064e-06,
      "loss": 1.3921,
      "step": 1378
    },
    {
      "epoch": 1.416538263995891,
      "grad_norm": 2.783045530319214,
      "learning_rate": 2.961553102398173e-06,
      "loss": 1.3295,
      "step": 1379
    },
    {
      "epoch": 1.4175654853620956,
      "grad_norm": 2.946359157562256,
      "learning_rate": 2.9596497906357063e-06,
      "loss": 1.4655,
      "step": 1380
    },
    {
      "epoch": 1.4175654853620956,
      "eval_loss": 1.7879260778427124,
      "eval_runtime": 37.1105,
      "eval_samples_per_second": 26.947,
      "eval_steps_per_second": 1.132,
      "step": 1380
    },
    {
      "epoch": 1.4185927067283,
      "grad_norm": 3.1604318618774414,
      "learning_rate": 2.9577464788732396e-06,
      "loss": 1.4746,
      "step": 1381
    },
    {
      "epoch": 1.4196199280945043,
      "grad_norm": 2.602548122406006,
      "learning_rate": 2.955843167110773e-06,
      "loss": 1.4024,
      "step": 1382
    },
    {
      "epoch": 1.420647149460709,
      "grad_norm": 2.774137258529663,
      "learning_rate": 2.9539398553483064e-06,
      "loss": 1.4257,
      "step": 1383
    },
    {
      "epoch": 1.4216743708269133,
      "grad_norm": 2.898256540298462,
      "learning_rate": 2.9520365435858393e-06,
      "loss": 1.4254,
      "step": 1384
    },
    {
      "epoch": 1.4227015921931176,
      "grad_norm": 2.8015098571777344,
      "learning_rate": 2.9501332318233727e-06,
      "loss": 1.495,
      "step": 1385
    },
    {
      "epoch": 1.423728813559322,
      "grad_norm": 2.7862651348114014,
      "learning_rate": 2.948229920060906e-06,
      "loss": 1.4355,
      "step": 1386
    },
    {
      "epoch": 1.4247560349255264,
      "grad_norm": 2.7947680950164795,
      "learning_rate": 2.9463266082984394e-06,
      "loss": 1.4126,
      "step": 1387
    },
    {
      "epoch": 1.425783256291731,
      "grad_norm": 2.8239989280700684,
      "learning_rate": 2.9444232965359727e-06,
      "loss": 1.417,
      "step": 1388
    },
    {
      "epoch": 1.4268104776579353,
      "grad_norm": 2.845521926879883,
      "learning_rate": 2.9425199847735057e-06,
      "loss": 1.4406,
      "step": 1389
    },
    {
      "epoch": 1.4278376990241397,
      "grad_norm": 2.9024460315704346,
      "learning_rate": 2.94061667301104e-06,
      "loss": 1.4083,
      "step": 1390
    },
    {
      "epoch": 1.4278376990241397,
      "eval_loss": 1.7976808547973633,
      "eval_runtime": 37.16,
      "eval_samples_per_second": 26.911,
      "eval_steps_per_second": 1.13,
      "step": 1390
    },
    {
      "epoch": 1.428864920390344,
      "grad_norm": 2.7083208560943604,
      "learning_rate": 2.938713361248573e-06,
      "loss": 1.3428,
      "step": 1391
    },
    {
      "epoch": 1.4298921417565484,
      "grad_norm": 2.7808196544647217,
      "learning_rate": 2.936810049486106e-06,
      "loss": 1.4262,
      "step": 1392
    },
    {
      "epoch": 1.430919363122753,
      "grad_norm": 2.8307440280914307,
      "learning_rate": 2.9349067377236396e-06,
      "loss": 1.4328,
      "step": 1393
    },
    {
      "epoch": 1.4319465844889574,
      "grad_norm": 2.8189167976379395,
      "learning_rate": 2.933003425961173e-06,
      "loss": 1.3878,
      "step": 1394
    },
    {
      "epoch": 1.4329738058551618,
      "grad_norm": 2.7747530937194824,
      "learning_rate": 2.9311001141987063e-06,
      "loss": 1.404,
      "step": 1395
    },
    {
      "epoch": 1.4340010272213661,
      "grad_norm": 2.747140884399414,
      "learning_rate": 2.9291968024362392e-06,
      "loss": 1.3954,
      "step": 1396
    },
    {
      "epoch": 1.4350282485875705,
      "grad_norm": 2.665836811065674,
      "learning_rate": 2.9272934906737726e-06,
      "loss": 1.3868,
      "step": 1397
    },
    {
      "epoch": 1.436055469953775,
      "grad_norm": 2.6918723583221436,
      "learning_rate": 2.925390178911306e-06,
      "loss": 1.3959,
      "step": 1398
    },
    {
      "epoch": 1.4370826913199795,
      "grad_norm": 2.8271594047546387,
      "learning_rate": 2.9234868671488393e-06,
      "loss": 1.4114,
      "step": 1399
    },
    {
      "epoch": 1.4381099126861838,
      "grad_norm": 3.037952184677124,
      "learning_rate": 2.9215835553863727e-06,
      "loss": 1.4267,
      "step": 1400
    },
    {
      "epoch": 1.4381099126861838,
      "eval_loss": 1.7927908897399902,
      "eval_runtime": 37.3671,
      "eval_samples_per_second": 26.761,
      "eval_steps_per_second": 1.124,
      "step": 1400
    },
    {
      "epoch": 1.4391371340523884,
      "grad_norm": 3.0185632705688477,
      "learning_rate": 2.9196802436239056e-06,
      "loss": 1.4211,
      "step": 1401
    },
    {
      "epoch": 1.4401643554185928,
      "grad_norm": 2.678468942642212,
      "learning_rate": 2.917776931861439e-06,
      "loss": 1.398,
      "step": 1402
    },
    {
      "epoch": 1.4411915767847971,
      "grad_norm": 2.8228728771209717,
      "learning_rate": 2.9158736200989723e-06,
      "loss": 1.3946,
      "step": 1403
    },
    {
      "epoch": 1.4422187981510015,
      "grad_norm": 2.6711578369140625,
      "learning_rate": 2.9139703083365057e-06,
      "loss": 1.3985,
      "step": 1404
    },
    {
      "epoch": 1.4432460195172059,
      "grad_norm": 2.7891721725463867,
      "learning_rate": 2.912066996574039e-06,
      "loss": 1.4232,
      "step": 1405
    },
    {
      "epoch": 1.4442732408834105,
      "grad_norm": 2.833613395690918,
      "learning_rate": 2.910163684811572e-06,
      "loss": 1.3629,
      "step": 1406
    },
    {
      "epoch": 1.4453004622496148,
      "grad_norm": 3.0415570735931396,
      "learning_rate": 2.9082603730491054e-06,
      "loss": 1.4666,
      "step": 1407
    },
    {
      "epoch": 1.4463276836158192,
      "grad_norm": 2.831063747406006,
      "learning_rate": 2.9063570612866387e-06,
      "loss": 1.4477,
      "step": 1408
    },
    {
      "epoch": 1.4473549049820236,
      "grad_norm": 2.8253402709960938,
      "learning_rate": 2.904453749524172e-06,
      "loss": 1.3729,
      "step": 1409
    },
    {
      "epoch": 1.448382126348228,
      "grad_norm": 2.9134843349456787,
      "learning_rate": 2.902550437761706e-06,
      "loss": 1.3958,
      "step": 1410
    },
    {
      "epoch": 1.448382126348228,
      "eval_loss": 1.7833805084228516,
      "eval_runtime": 36.9111,
      "eval_samples_per_second": 27.092,
      "eval_steps_per_second": 1.138,
      "step": 1410
    },
    {
      "epoch": 1.4494093477144325,
      "grad_norm": 2.962355852127075,
      "learning_rate": 2.9006471259992392e-06,
      "loss": 1.4143,
      "step": 1411
    },
    {
      "epoch": 1.450436569080637,
      "grad_norm": 2.738837242126465,
      "learning_rate": 2.8987438142367726e-06,
      "loss": 1.4051,
      "step": 1412
    },
    {
      "epoch": 1.4514637904468413,
      "grad_norm": 2.766331195831299,
      "learning_rate": 2.8968405024743055e-06,
      "loss": 1.4628,
      "step": 1413
    },
    {
      "epoch": 1.4524910118130456,
      "grad_norm": 2.6052863597869873,
      "learning_rate": 2.894937190711839e-06,
      "loss": 1.3662,
      "step": 1414
    },
    {
      "epoch": 1.45351823317925,
      "grad_norm": 2.8454713821411133,
      "learning_rate": 2.8930338789493723e-06,
      "loss": 1.4395,
      "step": 1415
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 2.683232545852661,
      "learning_rate": 2.8911305671869056e-06,
      "loss": 1.4243,
      "step": 1416
    },
    {
      "epoch": 1.455572675911659,
      "grad_norm": 2.798797845840454,
      "learning_rate": 2.889227255424439e-06,
      "loss": 1.3627,
      "step": 1417
    },
    {
      "epoch": 1.4565998972778633,
      "grad_norm": 2.821502923965454,
      "learning_rate": 2.887323943661972e-06,
      "loss": 1.4109,
      "step": 1418
    },
    {
      "epoch": 1.457627118644068,
      "grad_norm": 2.8054518699645996,
      "learning_rate": 2.8854206318995053e-06,
      "loss": 1.3906,
      "step": 1419
    },
    {
      "epoch": 1.4586543400102723,
      "grad_norm": 2.802000045776367,
      "learning_rate": 2.8835173201370386e-06,
      "loss": 1.4024,
      "step": 1420
    },
    {
      "epoch": 1.4586543400102723,
      "eval_loss": 1.7884918451309204,
      "eval_runtime": 37.0267,
      "eval_samples_per_second": 27.008,
      "eval_steps_per_second": 1.134,
      "step": 1420
    },
    {
      "epoch": 1.4596815613764766,
      "grad_norm": 3.097747564315796,
      "learning_rate": 2.881614008374572e-06,
      "loss": 1.4695,
      "step": 1421
    },
    {
      "epoch": 1.460708782742681,
      "grad_norm": 3.01665997505188,
      "learning_rate": 2.8797106966121054e-06,
      "loss": 1.4321,
      "step": 1422
    },
    {
      "epoch": 1.4617360041088854,
      "grad_norm": 2.9308114051818848,
      "learning_rate": 2.8778073848496383e-06,
      "loss": 1.4084,
      "step": 1423
    },
    {
      "epoch": 1.46276322547509,
      "grad_norm": 2.8440439701080322,
      "learning_rate": 2.8759040730871717e-06,
      "loss": 1.4473,
      "step": 1424
    },
    {
      "epoch": 1.4637904468412943,
      "grad_norm": 2.6690547466278076,
      "learning_rate": 2.874000761324705e-06,
      "loss": 1.407,
      "step": 1425
    },
    {
      "epoch": 1.4648176682074987,
      "grad_norm": 2.7961244583129883,
      "learning_rate": 2.8720974495622384e-06,
      "loss": 1.4414,
      "step": 1426
    },
    {
      "epoch": 1.465844889573703,
      "grad_norm": 2.891948938369751,
      "learning_rate": 2.8701941377997718e-06,
      "loss": 1.4185,
      "step": 1427
    },
    {
      "epoch": 1.4668721109399074,
      "grad_norm": 2.790811538696289,
      "learning_rate": 2.8682908260373047e-06,
      "loss": 1.4528,
      "step": 1428
    },
    {
      "epoch": 1.467899332306112,
      "grad_norm": 2.9391868114471436,
      "learning_rate": 2.866387514274838e-06,
      "loss": 1.4291,
      "step": 1429
    },
    {
      "epoch": 1.4689265536723164,
      "grad_norm": 2.9476356506347656,
      "learning_rate": 2.864484202512372e-06,
      "loss": 1.4373,
      "step": 1430
    },
    {
      "epoch": 1.4689265536723164,
      "eval_loss": 1.7926347255706787,
      "eval_runtime": 37.0308,
      "eval_samples_per_second": 27.005,
      "eval_steps_per_second": 1.134,
      "step": 1430
    },
    {
      "epoch": 1.4699537750385208,
      "grad_norm": 2.907172203063965,
      "learning_rate": 2.8625808907499052e-06,
      "loss": 1.3983,
      "step": 1431
    },
    {
      "epoch": 1.4709809964047253,
      "grad_norm": 2.8310680389404297,
      "learning_rate": 2.8606775789874386e-06,
      "loss": 1.4281,
      "step": 1432
    },
    {
      "epoch": 1.4720082177709297,
      "grad_norm": 2.8614001274108887,
      "learning_rate": 2.858774267224972e-06,
      "loss": 1.3975,
      "step": 1433
    },
    {
      "epoch": 1.473035439137134,
      "grad_norm": 3.1007702350616455,
      "learning_rate": 2.8568709554625053e-06,
      "loss": 1.4252,
      "step": 1434
    },
    {
      "epoch": 1.4740626605033385,
      "grad_norm": 2.879520893096924,
      "learning_rate": 2.8549676437000382e-06,
      "loss": 1.468,
      "step": 1435
    },
    {
      "epoch": 1.4750898818695428,
      "grad_norm": 2.9735469818115234,
      "learning_rate": 2.8530643319375716e-06,
      "loss": 1.4341,
      "step": 1436
    },
    {
      "epoch": 1.4761171032357474,
      "grad_norm": 3.0768795013427734,
      "learning_rate": 2.851161020175105e-06,
      "loss": 1.4122,
      "step": 1437
    },
    {
      "epoch": 1.4771443246019518,
      "grad_norm": 2.930915355682373,
      "learning_rate": 2.8492577084126383e-06,
      "loss": 1.3904,
      "step": 1438
    },
    {
      "epoch": 1.4781715459681561,
      "grad_norm": 2.659090280532837,
      "learning_rate": 2.8473543966501717e-06,
      "loss": 1.4261,
      "step": 1439
    },
    {
      "epoch": 1.4791987673343605,
      "grad_norm": 3.020371913909912,
      "learning_rate": 2.8454510848877046e-06,
      "loss": 1.4636,
      "step": 1440
    },
    {
      "epoch": 1.4791987673343605,
      "eval_loss": 1.8039511442184448,
      "eval_runtime": 37.0373,
      "eval_samples_per_second": 27.0,
      "eval_steps_per_second": 1.134,
      "step": 1440
    },
    {
      "epoch": 1.4802259887005649,
      "grad_norm": 3.131986141204834,
      "learning_rate": 2.843547773125238e-06,
      "loss": 1.4178,
      "step": 1441
    },
    {
      "epoch": 1.4812532100667695,
      "grad_norm": 2.8436167240142822,
      "learning_rate": 2.8416444613627714e-06,
      "loss": 1.4149,
      "step": 1442
    },
    {
      "epoch": 1.4822804314329738,
      "grad_norm": 2.8628804683685303,
      "learning_rate": 2.8397411496003047e-06,
      "loss": 1.3751,
      "step": 1443
    },
    {
      "epoch": 1.4833076527991782,
      "grad_norm": 2.917836904525757,
      "learning_rate": 2.837837837837838e-06,
      "loss": 1.4191,
      "step": 1444
    },
    {
      "epoch": 1.4843348741653826,
      "grad_norm": 2.951035737991333,
      "learning_rate": 2.835934526075371e-06,
      "loss": 1.4352,
      "step": 1445
    },
    {
      "epoch": 1.485362095531587,
      "grad_norm": 2.7296388149261475,
      "learning_rate": 2.8340312143129044e-06,
      "loss": 1.3856,
      "step": 1446
    },
    {
      "epoch": 1.4863893168977915,
      "grad_norm": 2.7064192295074463,
      "learning_rate": 2.8321279025504377e-06,
      "loss": 1.3782,
      "step": 1447
    },
    {
      "epoch": 1.487416538263996,
      "grad_norm": 2.9402806758880615,
      "learning_rate": 2.830224590787971e-06,
      "loss": 1.4829,
      "step": 1448
    },
    {
      "epoch": 1.4884437596302003,
      "grad_norm": 2.9216394424438477,
      "learning_rate": 2.8283212790255045e-06,
      "loss": 1.4173,
      "step": 1449
    },
    {
      "epoch": 1.4894709809964048,
      "grad_norm": 2.7382750511169434,
      "learning_rate": 2.8264179672630383e-06,
      "loss": 1.4431,
      "step": 1450
    },
    {
      "epoch": 1.4894709809964048,
      "eval_loss": 1.798187255859375,
      "eval_runtime": 37.0401,
      "eval_samples_per_second": 26.998,
      "eval_steps_per_second": 1.134,
      "step": 1450
    },
    {
      "epoch": 1.4904982023626092,
      "grad_norm": 2.781414270401001,
      "learning_rate": 2.8245146555005716e-06,
      "loss": 1.4096,
      "step": 1451
    },
    {
      "epoch": 1.4915254237288136,
      "grad_norm": 2.8284265995025635,
      "learning_rate": 2.8226113437381046e-06,
      "loss": 1.3561,
      "step": 1452
    },
    {
      "epoch": 1.492552645095018,
      "grad_norm": 2.6621487140655518,
      "learning_rate": 2.820708031975638e-06,
      "loss": 1.4232,
      "step": 1453
    },
    {
      "epoch": 1.4935798664612223,
      "grad_norm": 2.9116251468658447,
      "learning_rate": 2.8188047202131713e-06,
      "loss": 1.4431,
      "step": 1454
    },
    {
      "epoch": 1.494607087827427,
      "grad_norm": 2.749554395675659,
      "learning_rate": 2.8169014084507046e-06,
      "loss": 1.4603,
      "step": 1455
    },
    {
      "epoch": 1.4956343091936313,
      "grad_norm": 2.699585199356079,
      "learning_rate": 2.814998096688238e-06,
      "loss": 1.414,
      "step": 1456
    },
    {
      "epoch": 1.4966615305598356,
      "grad_norm": 2.6940858364105225,
      "learning_rate": 2.813094784925771e-06,
      "loss": 1.3673,
      "step": 1457
    },
    {
      "epoch": 1.49768875192604,
      "grad_norm": 2.859602689743042,
      "learning_rate": 2.8111914731633043e-06,
      "loss": 1.4664,
      "step": 1458
    },
    {
      "epoch": 1.4987159732922444,
      "grad_norm": 2.8793201446533203,
      "learning_rate": 2.8092881614008377e-06,
      "loss": 1.4168,
      "step": 1459
    },
    {
      "epoch": 1.499743194658449,
      "grad_norm": 2.8155760765075684,
      "learning_rate": 2.807384849638371e-06,
      "loss": 1.3917,
      "step": 1460
    },
    {
      "epoch": 1.499743194658449,
      "eval_loss": 1.780331015586853,
      "eval_runtime": 37.0489,
      "eval_samples_per_second": 26.991,
      "eval_steps_per_second": 1.134,
      "step": 1460
    },
    {
      "epoch": 1.5007704160246533,
      "grad_norm": 2.7490365505218506,
      "learning_rate": 2.8054815378759044e-06,
      "loss": 1.4326,
      "step": 1461
    },
    {
      "epoch": 1.5017976373908577,
      "grad_norm": 2.8913350105285645,
      "learning_rate": 2.8035782261134378e-06,
      "loss": 1.403,
      "step": 1462
    },
    {
      "epoch": 1.5028248587570623,
      "grad_norm": 2.83650279045105,
      "learning_rate": 2.8016749143509707e-06,
      "loss": 1.4107,
      "step": 1463
    },
    {
      "epoch": 1.5038520801232664,
      "grad_norm": 2.838649272918701,
      "learning_rate": 2.799771602588504e-06,
      "loss": 1.4131,
      "step": 1464
    },
    {
      "epoch": 1.504879301489471,
      "grad_norm": 2.8424947261810303,
      "learning_rate": 2.7978682908260374e-06,
      "loss": 1.4322,
      "step": 1465
    },
    {
      "epoch": 1.5059065228556754,
      "grad_norm": 2.8207895755767822,
      "learning_rate": 2.795964979063571e-06,
      "loss": 1.4464,
      "step": 1466
    },
    {
      "epoch": 1.5069337442218798,
      "grad_norm": 2.897221088409424,
      "learning_rate": 2.794061667301104e-06,
      "loss": 1.4237,
      "step": 1467
    },
    {
      "epoch": 1.5079609655880843,
      "grad_norm": 2.766166925430298,
      "learning_rate": 2.792158355538637e-06,
      "loss": 1.4375,
      "step": 1468
    },
    {
      "epoch": 1.5089881869542885,
      "grad_norm": 2.8115463256835938,
      "learning_rate": 2.7902550437761705e-06,
      "loss": 1.4265,
      "step": 1469
    },
    {
      "epoch": 1.510015408320493,
      "grad_norm": 2.826871156692505,
      "learning_rate": 2.7883517320137042e-06,
      "loss": 1.3933,
      "step": 1470
    },
    {
      "epoch": 1.510015408320493,
      "eval_loss": 1.780141830444336,
      "eval_runtime": 37.0572,
      "eval_samples_per_second": 26.985,
      "eval_steps_per_second": 1.133,
      "step": 1470
    },
    {
      "epoch": 1.5110426296866974,
      "grad_norm": 2.737398147583008,
      "learning_rate": 2.7864484202512376e-06,
      "loss": 1.3898,
      "step": 1471
    },
    {
      "epoch": 1.5120698510529018,
      "grad_norm": 2.9325804710388184,
      "learning_rate": 2.784545108488771e-06,
      "loss": 1.4139,
      "step": 1472
    },
    {
      "epoch": 1.5130970724191064,
      "grad_norm": 2.7400333881378174,
      "learning_rate": 2.7826417967263043e-06,
      "loss": 1.4122,
      "step": 1473
    },
    {
      "epoch": 1.5141242937853108,
      "grad_norm": 2.835869550704956,
      "learning_rate": 2.7807384849638373e-06,
      "loss": 1.4634,
      "step": 1474
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 2.7512247562408447,
      "learning_rate": 2.7788351732013706e-06,
      "loss": 1.409,
      "step": 1475
    },
    {
      "epoch": 1.5161787365177197,
      "grad_norm": 2.745054244995117,
      "learning_rate": 2.776931861438904e-06,
      "loss": 1.4004,
      "step": 1476
    },
    {
      "epoch": 1.5172059578839239,
      "grad_norm": 2.7766499519348145,
      "learning_rate": 2.7750285496764374e-06,
      "loss": 1.4197,
      "step": 1477
    },
    {
      "epoch": 1.5182331792501285,
      "grad_norm": 2.793663263320923,
      "learning_rate": 2.7731252379139707e-06,
      "loss": 1.4205,
      "step": 1478
    },
    {
      "epoch": 1.5192604006163328,
      "grad_norm": 2.674372434616089,
      "learning_rate": 2.771221926151504e-06,
      "loss": 1.3648,
      "step": 1479
    },
    {
      "epoch": 1.5202876219825372,
      "grad_norm": 2.7583112716674805,
      "learning_rate": 2.769318614389037e-06,
      "loss": 1.4296,
      "step": 1480
    },
    {
      "epoch": 1.5202876219825372,
      "eval_loss": 1.7793656587600708,
      "eval_runtime": 37.05,
      "eval_samples_per_second": 26.991,
      "eval_steps_per_second": 1.134,
      "step": 1480
    },
    {
      "epoch": 1.5213148433487418,
      "grad_norm": 2.776137351989746,
      "learning_rate": 2.7674153026265704e-06,
      "loss": 1.4267,
      "step": 1481
    },
    {
      "epoch": 1.522342064714946,
      "grad_norm": 2.8750247955322266,
      "learning_rate": 2.7655119908641037e-06,
      "loss": 1.4311,
      "step": 1482
    },
    {
      "epoch": 1.5233692860811505,
      "grad_norm": 2.740621566772461,
      "learning_rate": 2.763608679101637e-06,
      "loss": 1.4198,
      "step": 1483
    },
    {
      "epoch": 1.5243965074473549,
      "grad_norm": 2.875997304916382,
      "learning_rate": 2.7617053673391705e-06,
      "loss": 1.4557,
      "step": 1484
    },
    {
      "epoch": 1.5254237288135593,
      "grad_norm": 2.8406331539154053,
      "learning_rate": 2.7598020555767034e-06,
      "loss": 1.4504,
      "step": 1485
    },
    {
      "epoch": 1.5264509501797638,
      "grad_norm": 2.6013126373291016,
      "learning_rate": 2.7578987438142368e-06,
      "loss": 1.3829,
      "step": 1486
    },
    {
      "epoch": 1.5274781715459682,
      "grad_norm": 2.7469065189361572,
      "learning_rate": 2.75599543205177e-06,
      "loss": 1.4314,
      "step": 1487
    },
    {
      "epoch": 1.5285053929121726,
      "grad_norm": 2.8129167556762695,
      "learning_rate": 2.7540921202893035e-06,
      "loss": 1.4491,
      "step": 1488
    },
    {
      "epoch": 1.529532614278377,
      "grad_norm": 2.9677329063415527,
      "learning_rate": 2.752188808526837e-06,
      "loss": 1.4292,
      "step": 1489
    },
    {
      "epoch": 1.5305598356445813,
      "grad_norm": 2.7910637855529785,
      "learning_rate": 2.75028549676437e-06,
      "loss": 1.4157,
      "step": 1490
    },
    {
      "epoch": 1.5305598356445813,
      "eval_loss": 1.7833755016326904,
      "eval_runtime": 37.1104,
      "eval_samples_per_second": 26.947,
      "eval_steps_per_second": 1.132,
      "step": 1490
    },
    {
      "epoch": 1.531587057010786,
      "grad_norm": 2.7772984504699707,
      "learning_rate": 2.7483821850019036e-06,
      "loss": 1.4125,
      "step": 1491
    },
    {
      "epoch": 1.5326142783769903,
      "grad_norm": 2.747514247894287,
      "learning_rate": 2.746478873239437e-06,
      "loss": 1.4057,
      "step": 1492
    },
    {
      "epoch": 1.5336414997431946,
      "grad_norm": 2.7529642581939697,
      "learning_rate": 2.7445755614769703e-06,
      "loss": 1.4066,
      "step": 1493
    },
    {
      "epoch": 1.5346687211093992,
      "grad_norm": 2.8375084400177,
      "learning_rate": 2.7426722497145037e-06,
      "loss": 1.3883,
      "step": 1494
    },
    {
      "epoch": 1.5356959424756034,
      "grad_norm": 2.7483835220336914,
      "learning_rate": 2.740768937952037e-06,
      "loss": 1.3915,
      "step": 1495
    },
    {
      "epoch": 1.536723163841808,
      "grad_norm": 2.837123394012451,
      "learning_rate": 2.7388656261895704e-06,
      "loss": 1.4082,
      "step": 1496
    },
    {
      "epoch": 1.5377503852080123,
      "grad_norm": 2.909313440322876,
      "learning_rate": 2.7369623144271033e-06,
      "loss": 1.4815,
      "step": 1497
    },
    {
      "epoch": 1.5387776065742167,
      "grad_norm": 2.931723117828369,
      "learning_rate": 2.7350590026646367e-06,
      "loss": 1.4099,
      "step": 1498
    },
    {
      "epoch": 1.5398048279404213,
      "grad_norm": 2.9692134857177734,
      "learning_rate": 2.73315569090217e-06,
      "loss": 1.3637,
      "step": 1499
    },
    {
      "epoch": 1.5408320493066254,
      "grad_norm": 2.974923610687256,
      "learning_rate": 2.7312523791397034e-06,
      "loss": 1.4505,
      "step": 1500
    },
    {
      "epoch": 1.5408320493066254,
      "eval_loss": 1.7900456190109253,
      "eval_runtime": 37.3025,
      "eval_samples_per_second": 26.808,
      "eval_steps_per_second": 1.126,
      "step": 1500
    },
    {
      "epoch": 1.54185927067283,
      "grad_norm": 2.9976389408111572,
      "learning_rate": 2.7293490673772368e-06,
      "loss": 1.4726,
      "step": 1501
    },
    {
      "epoch": 1.5428864920390344,
      "grad_norm": 2.8478479385375977,
      "learning_rate": 2.7274457556147697e-06,
      "loss": 1.4304,
      "step": 1502
    },
    {
      "epoch": 1.5439137134052388,
      "grad_norm": 2.9100635051727295,
      "learning_rate": 2.725542443852303e-06,
      "loss": 1.4107,
      "step": 1503
    },
    {
      "epoch": 1.5449409347714433,
      "grad_norm": 2.903132677078247,
      "learning_rate": 2.7236391320898365e-06,
      "loss": 1.4206,
      "step": 1504
    },
    {
      "epoch": 1.5459681561376477,
      "grad_norm": 2.9177560806274414,
      "learning_rate": 2.72173582032737e-06,
      "loss": 1.3948,
      "step": 1505
    },
    {
      "epoch": 1.546995377503852,
      "grad_norm": 2.7861175537109375,
      "learning_rate": 2.719832508564903e-06,
      "loss": 1.4123,
      "step": 1506
    },
    {
      "epoch": 1.5480225988700564,
      "grad_norm": 2.787735939025879,
      "learning_rate": 2.717929196802436e-06,
      "loss": 1.4199,
      "step": 1507
    },
    {
      "epoch": 1.5490498202362608,
      "grad_norm": 2.889784097671509,
      "learning_rate": 2.7160258850399695e-06,
      "loss": 1.3901,
      "step": 1508
    },
    {
      "epoch": 1.5500770416024654,
      "grad_norm": 2.8383727073669434,
      "learning_rate": 2.714122573277503e-06,
      "loss": 1.4145,
      "step": 1509
    },
    {
      "epoch": 1.5511042629686698,
      "grad_norm": 2.7701327800750732,
      "learning_rate": 2.712219261515036e-06,
      "loss": 1.4213,
      "step": 1510
    },
    {
      "epoch": 1.5511042629686698,
      "eval_loss": 1.7863596677780151,
      "eval_runtime": 36.8927,
      "eval_samples_per_second": 27.106,
      "eval_steps_per_second": 1.138,
      "step": 1510
    },
    {
      "epoch": 1.5521314843348741,
      "grad_norm": 2.7612502574920654,
      "learning_rate": 2.71031594975257e-06,
      "loss": 1.4286,
      "step": 1511
    },
    {
      "epoch": 1.5531587057010787,
      "grad_norm": 2.696126937866211,
      "learning_rate": 2.7084126379901034e-06,
      "loss": 1.4359,
      "step": 1512
    },
    {
      "epoch": 1.5541859270672829,
      "grad_norm": 2.840223550796509,
      "learning_rate": 2.7065093262276367e-06,
      "loss": 1.4424,
      "step": 1513
    },
    {
      "epoch": 1.5552131484334875,
      "grad_norm": 2.7309999465942383,
      "learning_rate": 2.7046060144651697e-06,
      "loss": 1.4567,
      "step": 1514
    },
    {
      "epoch": 1.5562403697996918,
      "grad_norm": 2.8637242317199707,
      "learning_rate": 2.702702702702703e-06,
      "loss": 1.4699,
      "step": 1515
    },
    {
      "epoch": 1.5572675911658962,
      "grad_norm": 2.653404474258423,
      "learning_rate": 2.7007993909402364e-06,
      "loss": 1.3445,
      "step": 1516
    },
    {
      "epoch": 1.5582948125321008,
      "grad_norm": 2.9771206378936768,
      "learning_rate": 2.6988960791777697e-06,
      "loss": 1.4588,
      "step": 1517
    },
    {
      "epoch": 1.559322033898305,
      "grad_norm": 2.888413667678833,
      "learning_rate": 2.696992767415303e-06,
      "loss": 1.425,
      "step": 1518
    },
    {
      "epoch": 1.5603492552645095,
      "grad_norm": 2.8841593265533447,
      "learning_rate": 2.695089455652836e-06,
      "loss": 1.4101,
      "step": 1519
    },
    {
      "epoch": 1.5613764766307139,
      "grad_norm": 2.990192413330078,
      "learning_rate": 2.6931861438903694e-06,
      "loss": 1.3897,
      "step": 1520
    },
    {
      "epoch": 1.5613764766307139,
      "eval_loss": 1.7787587642669678,
      "eval_runtime": 36.9999,
      "eval_samples_per_second": 27.027,
      "eval_steps_per_second": 1.135,
      "step": 1520
    },
    {
      "epoch": 1.5624036979969183,
      "grad_norm": 2.8698887825012207,
      "learning_rate": 2.6912828321279028e-06,
      "loss": 1.4324,
      "step": 1521
    },
    {
      "epoch": 1.5634309193631228,
      "grad_norm": 2.8924384117126465,
      "learning_rate": 2.689379520365436e-06,
      "loss": 1.4073,
      "step": 1522
    },
    {
      "epoch": 1.5644581407293272,
      "grad_norm": 2.92336106300354,
      "learning_rate": 2.6874762086029695e-06,
      "loss": 1.4471,
      "step": 1523
    },
    {
      "epoch": 1.5654853620955316,
      "grad_norm": 2.7212202548980713,
      "learning_rate": 2.6855728968405024e-06,
      "loss": 1.3902,
      "step": 1524
    },
    {
      "epoch": 1.5665125834617362,
      "grad_norm": 2.711582660675049,
      "learning_rate": 2.683669585078036e-06,
      "loss": 1.4294,
      "step": 1525
    },
    {
      "epoch": 1.5675398048279403,
      "grad_norm": 2.6806275844573975,
      "learning_rate": 2.681766273315569e-06,
      "loss": 1.3683,
      "step": 1526
    },
    {
      "epoch": 1.568567026194145,
      "grad_norm": 2.920119524002075,
      "learning_rate": 2.6798629615531025e-06,
      "loss": 1.4194,
      "step": 1527
    },
    {
      "epoch": 1.5695942475603493,
      "grad_norm": 2.7487800121307373,
      "learning_rate": 2.677959649790636e-06,
      "loss": 1.4091,
      "step": 1528
    },
    {
      "epoch": 1.5706214689265536,
      "grad_norm": 2.7557766437530518,
      "learning_rate": 2.676056338028169e-06,
      "loss": 1.4453,
      "step": 1529
    },
    {
      "epoch": 1.5716486902927582,
      "grad_norm": 2.7834057807922363,
      "learning_rate": 2.674153026265702e-06,
      "loss": 1.4381,
      "step": 1530
    },
    {
      "epoch": 1.5716486902927582,
      "eval_loss": 1.7810354232788086,
      "eval_runtime": 37.037,
      "eval_samples_per_second": 27.0,
      "eval_steps_per_second": 1.134,
      "step": 1530
    },
    {
      "epoch": 1.5726759116589624,
      "grad_norm": 3.107299327850342,
      "learning_rate": 2.672249714503236e-06,
      "loss": 1.4475,
      "step": 1531
    },
    {
      "epoch": 1.573703133025167,
      "grad_norm": 2.813086748123169,
      "learning_rate": 2.6703464027407693e-06,
      "loss": 1.4062,
      "step": 1532
    },
    {
      "epoch": 1.5747303543913713,
      "grad_norm": 2.781416416168213,
      "learning_rate": 2.6684430909783027e-06,
      "loss": 1.4268,
      "step": 1533
    },
    {
      "epoch": 1.5757575757575757,
      "grad_norm": 2.681947708129883,
      "learning_rate": 2.666539779215836e-06,
      "loss": 1.4121,
      "step": 1534
    },
    {
      "epoch": 1.5767847971237803,
      "grad_norm": 2.656013011932373,
      "learning_rate": 2.6646364674533694e-06,
      "loss": 1.4003,
      "step": 1535
    },
    {
      "epoch": 1.5778120184899846,
      "grad_norm": 2.5161869525909424,
      "learning_rate": 2.6627331556909024e-06,
      "loss": 1.3127,
      "step": 1536
    },
    {
      "epoch": 1.578839239856189,
      "grad_norm": 2.842507839202881,
      "learning_rate": 2.6608298439284357e-06,
      "loss": 1.4067,
      "step": 1537
    },
    {
      "epoch": 1.5798664612223934,
      "grad_norm": 2.732778787612915,
      "learning_rate": 2.658926532165969e-06,
      "loss": 1.4071,
      "step": 1538
    },
    {
      "epoch": 1.5808936825885977,
      "grad_norm": 2.802293300628662,
      "learning_rate": 2.6570232204035024e-06,
      "loss": 1.3637,
      "step": 1539
    },
    {
      "epoch": 1.5819209039548023,
      "grad_norm": 2.927441358566284,
      "learning_rate": 2.655119908641036e-06,
      "loss": 1.405,
      "step": 1540
    },
    {
      "epoch": 1.5819209039548023,
      "eval_loss": 1.7797114849090576,
      "eval_runtime": 37.0514,
      "eval_samples_per_second": 26.99,
      "eval_steps_per_second": 1.134,
      "step": 1540
    },
    {
      "epoch": 1.5829481253210067,
      "grad_norm": 2.959746837615967,
      "learning_rate": 2.6532165968785687e-06,
      "loss": 1.4123,
      "step": 1541
    },
    {
      "epoch": 1.583975346687211,
      "grad_norm": 2.679551124572754,
      "learning_rate": 2.651313285116102e-06,
      "loss": 1.3742,
      "step": 1542
    },
    {
      "epoch": 1.5850025680534157,
      "grad_norm": 2.7431442737579346,
      "learning_rate": 2.6494099733536355e-06,
      "loss": 1.3825,
      "step": 1543
    },
    {
      "epoch": 1.5860297894196198,
      "grad_norm": 3.3630471229553223,
      "learning_rate": 2.647506661591169e-06,
      "loss": 1.5599,
      "step": 1544
    },
    {
      "epoch": 1.5870570107858244,
      "grad_norm": 2.8524038791656494,
      "learning_rate": 2.645603349828702e-06,
      "loss": 1.4293,
      "step": 1545
    },
    {
      "epoch": 1.5880842321520288,
      "grad_norm": 2.871917486190796,
      "learning_rate": 2.643700038066235e-06,
      "loss": 1.3716,
      "step": 1546
    },
    {
      "epoch": 1.5891114535182331,
      "grad_norm": 2.7151851654052734,
      "learning_rate": 2.6417967263037685e-06,
      "loss": 1.43,
      "step": 1547
    },
    {
      "epoch": 1.5901386748844377,
      "grad_norm": 2.7912635803222656,
      "learning_rate": 2.639893414541302e-06,
      "loss": 1.4401,
      "step": 1548
    },
    {
      "epoch": 1.5911658962506419,
      "grad_norm": 2.9613499641418457,
      "learning_rate": 2.6379901027788352e-06,
      "loss": 1.4691,
      "step": 1549
    },
    {
      "epoch": 1.5921931176168465,
      "grad_norm": 2.8831710815429688,
      "learning_rate": 2.6360867910163686e-06,
      "loss": 1.3975,
      "step": 1550
    },
    {
      "epoch": 1.5921931176168465,
      "eval_loss": 1.7762643098831177,
      "eval_runtime": 37.0485,
      "eval_samples_per_second": 26.992,
      "eval_steps_per_second": 1.134,
      "step": 1550
    },
    {
      "epoch": 1.5932203389830508,
      "grad_norm": 2.6928372383117676,
      "learning_rate": 2.6341834792539024e-06,
      "loss": 1.4179,
      "step": 1551
    },
    {
      "epoch": 1.5942475603492552,
      "grad_norm": 2.8088696002960205,
      "learning_rate": 2.6322801674914357e-06,
      "loss": 1.417,
      "step": 1552
    },
    {
      "epoch": 1.5952747817154598,
      "grad_norm": 2.926135778427124,
      "learning_rate": 2.6303768557289687e-06,
      "loss": 1.4145,
      "step": 1553
    },
    {
      "epoch": 1.5963020030816641,
      "grad_norm": 2.7283592224121094,
      "learning_rate": 2.628473543966502e-06,
      "loss": 1.3714,
      "step": 1554
    },
    {
      "epoch": 1.5973292244478685,
      "grad_norm": 2.7970361709594727,
      "learning_rate": 2.6265702322040354e-06,
      "loss": 1.4393,
      "step": 1555
    },
    {
      "epoch": 1.5983564458140729,
      "grad_norm": 2.717223644256592,
      "learning_rate": 2.6246669204415688e-06,
      "loss": 1.3857,
      "step": 1556
    },
    {
      "epoch": 1.5993836671802772,
      "grad_norm": 2.817638635635376,
      "learning_rate": 2.622763608679102e-06,
      "loss": 1.3863,
      "step": 1557
    },
    {
      "epoch": 1.6004108885464818,
      "grad_norm": 2.7856762409210205,
      "learning_rate": 2.620860296916635e-06,
      "loss": 1.4052,
      "step": 1558
    },
    {
      "epoch": 1.6014381099126862,
      "grad_norm": 2.9281742572784424,
      "learning_rate": 2.6189569851541684e-06,
      "loss": 1.3417,
      "step": 1559
    },
    {
      "epoch": 1.6024653312788906,
      "grad_norm": 2.8528225421905518,
      "learning_rate": 2.617053673391702e-06,
      "loss": 1.4486,
      "step": 1560
    },
    {
      "epoch": 1.6024653312788906,
      "eval_loss": 1.771835207939148,
      "eval_runtime": 37.0601,
      "eval_samples_per_second": 26.983,
      "eval_steps_per_second": 1.133,
      "step": 1560
    },
    {
      "epoch": 1.6034925526450952,
      "grad_norm": 3.0976126194000244,
      "learning_rate": 2.615150361629235e-06,
      "loss": 1.4722,
      "step": 1561
    },
    {
      "epoch": 1.6045197740112993,
      "grad_norm": 2.9993839263916016,
      "learning_rate": 2.6132470498667685e-06,
      "loss": 1.4265,
      "step": 1562
    },
    {
      "epoch": 1.605546995377504,
      "grad_norm": 2.896975517272949,
      "learning_rate": 2.6113437381043015e-06,
      "loss": 1.4035,
      "step": 1563
    },
    {
      "epoch": 1.6065742167437083,
      "grad_norm": 2.9382545948028564,
      "learning_rate": 2.609440426341835e-06,
      "loss": 1.3934,
      "step": 1564
    },
    {
      "epoch": 1.6076014381099126,
      "grad_norm": 2.950702667236328,
      "learning_rate": 2.607537114579368e-06,
      "loss": 1.4272,
      "step": 1565
    },
    {
      "epoch": 1.6086286594761172,
      "grad_norm": 2.888334274291992,
      "learning_rate": 2.6056338028169015e-06,
      "loss": 1.4088,
      "step": 1566
    },
    {
      "epoch": 1.6096558808423214,
      "grad_norm": 2.6972687244415283,
      "learning_rate": 2.603730491054435e-06,
      "loss": 1.391,
      "step": 1567
    },
    {
      "epoch": 1.610683102208526,
      "grad_norm": 2.834188461303711,
      "learning_rate": 2.601827179291968e-06,
      "loss": 1.4241,
      "step": 1568
    },
    {
      "epoch": 1.6117103235747303,
      "grad_norm": 2.894808769226074,
      "learning_rate": 2.599923867529501e-06,
      "loss": 1.4297,
      "step": 1569
    },
    {
      "epoch": 1.6127375449409347,
      "grad_norm": 2.9836509227752686,
      "learning_rate": 2.5980205557670346e-06,
      "loss": 1.4086,
      "step": 1570
    },
    {
      "epoch": 1.6127375449409347,
      "eval_loss": 1.7795302867889404,
      "eval_runtime": 37.0629,
      "eval_samples_per_second": 26.981,
      "eval_steps_per_second": 1.133,
      "step": 1570
    },
    {
      "epoch": 1.6137647663071393,
      "grad_norm": 2.8716723918914795,
      "learning_rate": 2.5961172440045684e-06,
      "loss": 1.4425,
      "step": 1571
    },
    {
      "epoch": 1.6147919876733436,
      "grad_norm": 2.8877105712890625,
      "learning_rate": 2.5942139322421017e-06,
      "loss": 1.4557,
      "step": 1572
    },
    {
      "epoch": 1.615819209039548,
      "grad_norm": 2.8737246990203857,
      "learning_rate": 2.592310620479635e-06,
      "loss": 1.4015,
      "step": 1573
    },
    {
      "epoch": 1.6168464304057526,
      "grad_norm": 2.7048048973083496,
      "learning_rate": 2.5904073087171684e-06,
      "loss": 1.361,
      "step": 1574
    },
    {
      "epoch": 1.6178736517719567,
      "grad_norm": 2.79799222946167,
      "learning_rate": 2.5885039969547014e-06,
      "loss": 1.4325,
      "step": 1575
    },
    {
      "epoch": 1.6189008731381613,
      "grad_norm": 2.788114309310913,
      "learning_rate": 2.5866006851922347e-06,
      "loss": 1.378,
      "step": 1576
    },
    {
      "epoch": 1.6199280945043657,
      "grad_norm": 2.7423717975616455,
      "learning_rate": 2.584697373429768e-06,
      "loss": 1.3792,
      "step": 1577
    },
    {
      "epoch": 1.62095531587057,
      "grad_norm": 2.777557849884033,
      "learning_rate": 2.5827940616673015e-06,
      "loss": 1.407,
      "step": 1578
    },
    {
      "epoch": 1.6219825372367747,
      "grad_norm": 3.0097639560699463,
      "learning_rate": 2.580890749904835e-06,
      "loss": 1.4509,
      "step": 1579
    },
    {
      "epoch": 1.6230097586029788,
      "grad_norm": 2.74116849899292,
      "learning_rate": 2.5789874381423678e-06,
      "loss": 1.3975,
      "step": 1580
    },
    {
      "epoch": 1.6230097586029788,
      "eval_loss": 1.7757364511489868,
      "eval_runtime": 37.0779,
      "eval_samples_per_second": 26.97,
      "eval_steps_per_second": 1.133,
      "step": 1580
    },
    {
      "epoch": 1.6240369799691834,
      "grad_norm": 2.670633316040039,
      "learning_rate": 2.577084126379901e-06,
      "loss": 1.401,
      "step": 1581
    },
    {
      "epoch": 1.6250642013353878,
      "grad_norm": 2.7129952907562256,
      "learning_rate": 2.5751808146174345e-06,
      "loss": 1.3663,
      "step": 1582
    },
    {
      "epoch": 1.6260914227015921,
      "grad_norm": 2.839162826538086,
      "learning_rate": 2.573277502854968e-06,
      "loss": 1.4189,
      "step": 1583
    },
    {
      "epoch": 1.6271186440677967,
      "grad_norm": 2.7310187816619873,
      "learning_rate": 2.5713741910925012e-06,
      "loss": 1.3899,
      "step": 1584
    },
    {
      "epoch": 1.6281458654340009,
      "grad_norm": 2.6793200969696045,
      "learning_rate": 2.569470879330034e-06,
      "loss": 1.3926,
      "step": 1585
    },
    {
      "epoch": 1.6291730868002054,
      "grad_norm": 2.946446418762207,
      "learning_rate": 2.5675675675675675e-06,
      "loss": 1.4539,
      "step": 1586
    },
    {
      "epoch": 1.6302003081664098,
      "grad_norm": 2.683105945587158,
      "learning_rate": 2.565664255805101e-06,
      "loss": 1.399,
      "step": 1587
    },
    {
      "epoch": 1.6312275295326142,
      "grad_norm": 2.8176918029785156,
      "learning_rate": 2.5637609440426343e-06,
      "loss": 1.427,
      "step": 1588
    },
    {
      "epoch": 1.6322547508988188,
      "grad_norm": 2.646280288696289,
      "learning_rate": 2.5618576322801676e-06,
      "loss": 1.3738,
      "step": 1589
    },
    {
      "epoch": 1.6332819722650231,
      "grad_norm": 2.9067888259887695,
      "learning_rate": 2.5599543205177006e-06,
      "loss": 1.4666,
      "step": 1590
    },
    {
      "epoch": 1.6332819722650231,
      "eval_loss": 1.7719416618347168,
      "eval_runtime": 37.1503,
      "eval_samples_per_second": 26.918,
      "eval_steps_per_second": 1.131,
      "step": 1590
    },
    {
      "epoch": 1.6343091936312275,
      "grad_norm": 2.797149658203125,
      "learning_rate": 2.558051008755234e-06,
      "loss": 1.4162,
      "step": 1591
    },
    {
      "epoch": 1.635336414997432,
      "grad_norm": 2.8294360637664795,
      "learning_rate": 2.5561476969927677e-06,
      "loss": 1.4658,
      "step": 1592
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 2.738529682159424,
      "learning_rate": 2.554244385230301e-06,
      "loss": 1.3883,
      "step": 1593
    },
    {
      "epoch": 1.6373908577298408,
      "grad_norm": 2.841029167175293,
      "learning_rate": 2.5523410734678344e-06,
      "loss": 1.4181,
      "step": 1594
    },
    {
      "epoch": 1.6384180790960452,
      "grad_norm": 2.9257984161376953,
      "learning_rate": 2.550437761705368e-06,
      "loss": 1.3491,
      "step": 1595
    },
    {
      "epoch": 1.6394453004622496,
      "grad_norm": 2.846874713897705,
      "learning_rate": 2.548534449942901e-06,
      "loss": 1.3832,
      "step": 1596
    },
    {
      "epoch": 1.6404725218284542,
      "grad_norm": 2.9274935722351074,
      "learning_rate": 2.546631138180434e-06,
      "loss": 1.4215,
      "step": 1597
    },
    {
      "epoch": 1.6414997431946583,
      "grad_norm": 2.9953150749206543,
      "learning_rate": 2.5447278264179675e-06,
      "loss": 1.4098,
      "step": 1598
    },
    {
      "epoch": 1.642526964560863,
      "grad_norm": 2.7812561988830566,
      "learning_rate": 2.542824514655501e-06,
      "loss": 1.3806,
      "step": 1599
    },
    {
      "epoch": 1.6435541859270673,
      "grad_norm": 2.9627180099487305,
      "learning_rate": 2.540921202893034e-06,
      "loss": 1.4623,
      "step": 1600
    },
    {
      "epoch": 1.6435541859270673,
      "eval_loss": 1.7626967430114746,
      "eval_runtime": 37.2795,
      "eval_samples_per_second": 26.824,
      "eval_steps_per_second": 1.127,
      "step": 1600
    },
    {
      "epoch": 1.6445814072932716,
      "grad_norm": 2.8668668270111084,
      "learning_rate": 2.5390178911305675e-06,
      "loss": 1.4446,
      "step": 1601
    },
    {
      "epoch": 1.6456086286594762,
      "grad_norm": 2.7962896823883057,
      "learning_rate": 2.5371145793681005e-06,
      "loss": 1.4155,
      "step": 1602
    },
    {
      "epoch": 1.6466358500256806,
      "grad_norm": 2.76855731010437,
      "learning_rate": 2.535211267605634e-06,
      "loss": 1.4096,
      "step": 1603
    },
    {
      "epoch": 1.647663071391885,
      "grad_norm": 2.7888379096984863,
      "learning_rate": 2.533307955843167e-06,
      "loss": 1.3544,
      "step": 1604
    },
    {
      "epoch": 1.6486902927580893,
      "grad_norm": 2.9665539264678955,
      "learning_rate": 2.5314046440807006e-06,
      "loss": 1.4157,
      "step": 1605
    },
    {
      "epoch": 1.6497175141242937,
      "grad_norm": 2.9601891040802,
      "learning_rate": 2.529501332318234e-06,
      "loss": 1.3963,
      "step": 1606
    },
    {
      "epoch": 1.6507447354904983,
      "grad_norm": 2.897718667984009,
      "learning_rate": 2.5275980205557673e-06,
      "loss": 1.4498,
      "step": 1607
    },
    {
      "epoch": 1.6517719568567026,
      "grad_norm": 2.9813809394836426,
      "learning_rate": 2.5256947087933002e-06,
      "loss": 1.4791,
      "step": 1608
    },
    {
      "epoch": 1.652799178222907,
      "grad_norm": 3.025689125061035,
      "learning_rate": 2.5237913970308336e-06,
      "loss": 1.4616,
      "step": 1609
    },
    {
      "epoch": 1.6538263995891116,
      "grad_norm": 2.6942813396453857,
      "learning_rate": 2.521888085268367e-06,
      "loss": 1.4112,
      "step": 1610
    },
    {
      "epoch": 1.6538263995891116,
      "eval_loss": 1.775415301322937,
      "eval_runtime": 36.8669,
      "eval_samples_per_second": 27.125,
      "eval_steps_per_second": 1.139,
      "step": 1610
    },
    {
      "epoch": 1.6548536209553157,
      "grad_norm": 2.8190956115722656,
      "learning_rate": 2.5199847735059003e-06,
      "loss": 1.4559,
      "step": 1611
    },
    {
      "epoch": 1.6558808423215203,
      "grad_norm": 2.7085635662078857,
      "learning_rate": 2.518081461743434e-06,
      "loss": 1.3905,
      "step": 1612
    },
    {
      "epoch": 1.6569080636877247,
      "grad_norm": 2.856039524078369,
      "learning_rate": 2.5161781499809675e-06,
      "loss": 1.4223,
      "step": 1613
    },
    {
      "epoch": 1.657935285053929,
      "grad_norm": 2.703293800354004,
      "learning_rate": 2.5142748382185004e-06,
      "loss": 1.3809,
      "step": 1614
    },
    {
      "epoch": 1.6589625064201337,
      "grad_norm": 2.7698261737823486,
      "learning_rate": 2.5123715264560338e-06,
      "loss": 1.3732,
      "step": 1615
    },
    {
      "epoch": 1.6599897277863378,
      "grad_norm": 2.8664398193359375,
      "learning_rate": 2.510468214693567e-06,
      "loss": 1.3994,
      "step": 1616
    },
    {
      "epoch": 1.6610169491525424,
      "grad_norm": 2.843365430831909,
      "learning_rate": 2.5085649029311005e-06,
      "loss": 1.3466,
      "step": 1617
    },
    {
      "epoch": 1.6620441705187468,
      "grad_norm": 2.7871944904327393,
      "learning_rate": 2.506661591168634e-06,
      "loss": 1.3096,
      "step": 1618
    },
    {
      "epoch": 1.6630713918849511,
      "grad_norm": 2.9175424575805664,
      "learning_rate": 2.504758279406167e-06,
      "loss": 1.3931,
      "step": 1619
    },
    {
      "epoch": 1.6640986132511557,
      "grad_norm": 2.926497220993042,
      "learning_rate": 2.5028549676437e-06,
      "loss": 1.4342,
      "step": 1620
    },
    {
      "epoch": 1.6640986132511557,
      "eval_loss": 1.773324966430664,
      "eval_runtime": 36.9745,
      "eval_samples_per_second": 27.046,
      "eval_steps_per_second": 1.136,
      "step": 1620
    },
    {
      "epoch": 1.66512583461736,
      "grad_norm": 2.7389914989471436,
      "learning_rate": 2.5009516558812335e-06,
      "loss": 1.3808,
      "step": 1621
    },
    {
      "epoch": 1.6661530559835644,
      "grad_norm": 2.7649383544921875,
      "learning_rate": 2.499048344118767e-06,
      "loss": 1.4307,
      "step": 1622
    },
    {
      "epoch": 1.667180277349769,
      "grad_norm": 2.8508212566375732,
      "learning_rate": 2.4971450323563002e-06,
      "loss": 1.4699,
      "step": 1623
    },
    {
      "epoch": 1.6682074987159732,
      "grad_norm": 2.8107385635375977,
      "learning_rate": 2.4952417205938336e-06,
      "loss": 1.3802,
      "step": 1624
    },
    {
      "epoch": 1.6692347200821778,
      "grad_norm": 2.820322036743164,
      "learning_rate": 2.4933384088313666e-06,
      "loss": 1.3896,
      "step": 1625
    },
    {
      "epoch": 1.6702619414483821,
      "grad_norm": 2.7912018299102783,
      "learning_rate": 2.4914350970689e-06,
      "loss": 1.3782,
      "step": 1626
    },
    {
      "epoch": 1.6712891628145865,
      "grad_norm": 2.854904890060425,
      "learning_rate": 2.4895317853064337e-06,
      "loss": 1.386,
      "step": 1627
    },
    {
      "epoch": 1.672316384180791,
      "grad_norm": 3.2605831623077393,
      "learning_rate": 2.4876284735439666e-06,
      "loss": 1.446,
      "step": 1628
    },
    {
      "epoch": 1.6733436055469952,
      "grad_norm": 3.029327630996704,
      "learning_rate": 2.4857251617815e-06,
      "loss": 1.414,
      "step": 1629
    },
    {
      "epoch": 1.6743708269131998,
      "grad_norm": 2.9044439792633057,
      "learning_rate": 2.4838218500190334e-06,
      "loss": 1.4219,
      "step": 1630
    },
    {
      "epoch": 1.6743708269131998,
      "eval_loss": 1.778739094734192,
      "eval_runtime": 36.9917,
      "eval_samples_per_second": 27.033,
      "eval_steps_per_second": 1.135,
      "step": 1630
    },
    {
      "epoch": 1.6753980482794042,
      "grad_norm": 2.6776554584503174,
      "learning_rate": 2.4819185382565667e-06,
      "loss": 1.4001,
      "step": 1631
    },
    {
      "epoch": 1.6764252696456086,
      "grad_norm": 2.841038227081299,
      "learning_rate": 2.4800152264941e-06,
      "loss": 1.406,
      "step": 1632
    },
    {
      "epoch": 1.6774524910118132,
      "grad_norm": 2.848771095275879,
      "learning_rate": 2.478111914731633e-06,
      "loss": 1.3663,
      "step": 1633
    },
    {
      "epoch": 1.6784797123780173,
      "grad_norm": 2.8403055667877197,
      "learning_rate": 2.4762086029691664e-06,
      "loss": 1.4065,
      "step": 1634
    },
    {
      "epoch": 1.6795069337442219,
      "grad_norm": 3.0035014152526855,
      "learning_rate": 2.4743052912066998e-06,
      "loss": 1.5197,
      "step": 1635
    },
    {
      "epoch": 1.6805341551104263,
      "grad_norm": 2.764261245727539,
      "learning_rate": 2.472401979444233e-06,
      "loss": 1.37,
      "step": 1636
    },
    {
      "epoch": 1.6815613764766306,
      "grad_norm": 3.0496931076049805,
      "learning_rate": 2.4704986676817665e-06,
      "loss": 1.4657,
      "step": 1637
    },
    {
      "epoch": 1.6825885978428352,
      "grad_norm": 2.8577051162719727,
      "learning_rate": 2.4685953559193e-06,
      "loss": 1.4019,
      "step": 1638
    },
    {
      "epoch": 1.6836158192090396,
      "grad_norm": 2.8699984550476074,
      "learning_rate": 2.466692044156833e-06,
      "loss": 1.4508,
      "step": 1639
    },
    {
      "epoch": 1.684643040575244,
      "grad_norm": 2.8038368225097656,
      "learning_rate": 2.4647887323943666e-06,
      "loss": 1.3859,
      "step": 1640
    },
    {
      "epoch": 1.684643040575244,
      "eval_loss": 1.7762134075164795,
      "eval_runtime": 37.0006,
      "eval_samples_per_second": 27.027,
      "eval_steps_per_second": 1.135,
      "step": 1640
    },
    {
      "epoch": 1.6856702619414485,
      "grad_norm": 2.745168447494507,
      "learning_rate": 2.4628854206319e-06,
      "loss": 1.3581,
      "step": 1641
    },
    {
      "epoch": 1.6866974833076527,
      "grad_norm": 2.8323488235473633,
      "learning_rate": 2.460982108869433e-06,
      "loss": 1.3661,
      "step": 1642
    },
    {
      "epoch": 1.6877247046738573,
      "grad_norm": 2.8715322017669678,
      "learning_rate": 2.4590787971069662e-06,
      "loss": 1.4519,
      "step": 1643
    },
    {
      "epoch": 1.6887519260400616,
      "grad_norm": 2.752286672592163,
      "learning_rate": 2.4571754853444996e-06,
      "loss": 1.3886,
      "step": 1644
    },
    {
      "epoch": 1.689779147406266,
      "grad_norm": 2.7300102710723877,
      "learning_rate": 2.455272173582033e-06,
      "loss": 1.3767,
      "step": 1645
    },
    {
      "epoch": 1.6908063687724706,
      "grad_norm": 2.803774356842041,
      "learning_rate": 2.4533688618195663e-06,
      "loss": 1.39,
      "step": 1646
    },
    {
      "epoch": 1.6918335901386747,
      "grad_norm": 2.6127994060516357,
      "learning_rate": 2.4514655500570997e-06,
      "loss": 1.3186,
      "step": 1647
    },
    {
      "epoch": 1.6928608115048793,
      "grad_norm": 2.718994140625,
      "learning_rate": 2.449562238294633e-06,
      "loss": 1.3693,
      "step": 1648
    },
    {
      "epoch": 1.6938880328710837,
      "grad_norm": 2.635831117630005,
      "learning_rate": 2.4476589265321664e-06,
      "loss": 1.3342,
      "step": 1649
    },
    {
      "epoch": 1.694915254237288,
      "grad_norm": 3.0353753566741943,
      "learning_rate": 2.4457556147696993e-06,
      "loss": 1.4433,
      "step": 1650
    },
    {
      "epoch": 1.694915254237288,
      "eval_loss": 1.7774006128311157,
      "eval_runtime": 37.0101,
      "eval_samples_per_second": 27.02,
      "eval_steps_per_second": 1.135,
      "step": 1650
    },
    {
      "epoch": 1.6959424756034926,
      "grad_norm": 2.909377098083496,
      "learning_rate": 2.4438523030072327e-06,
      "loss": 1.4173,
      "step": 1651
    },
    {
      "epoch": 1.696969696969697,
      "grad_norm": 2.9672727584838867,
      "learning_rate": 2.441948991244766e-06,
      "loss": 1.4274,
      "step": 1652
    },
    {
      "epoch": 1.6979969183359014,
      "grad_norm": 2.822268486022949,
      "learning_rate": 2.4400456794822994e-06,
      "loss": 1.4012,
      "step": 1653
    },
    {
      "epoch": 1.6990241397021058,
      "grad_norm": 2.8190255165100098,
      "learning_rate": 2.438142367719833e-06,
      "loss": 1.4251,
      "step": 1654
    },
    {
      "epoch": 1.7000513610683101,
      "grad_norm": 2.8970980644226074,
      "learning_rate": 2.4362390559573657e-06,
      "loss": 1.3961,
      "step": 1655
    },
    {
      "epoch": 1.7010785824345147,
      "grad_norm": 2.9281842708587646,
      "learning_rate": 2.434335744194899e-06,
      "loss": 1.3903,
      "step": 1656
    },
    {
      "epoch": 1.702105803800719,
      "grad_norm": 3.0265564918518066,
      "learning_rate": 2.432432432432433e-06,
      "loss": 1.4319,
      "step": 1657
    },
    {
      "epoch": 1.7031330251669234,
      "grad_norm": 2.785891056060791,
      "learning_rate": 2.4305291206699662e-06,
      "loss": 1.3561,
      "step": 1658
    },
    {
      "epoch": 1.704160246533128,
      "grad_norm": 2.8406574726104736,
      "learning_rate": 2.428625808907499e-06,
      "loss": 1.4198,
      "step": 1659
    },
    {
      "epoch": 1.7051874678993322,
      "grad_norm": 2.856170415878296,
      "learning_rate": 2.4267224971450325e-06,
      "loss": 1.446,
      "step": 1660
    },
    {
      "epoch": 1.7051874678993322,
      "eval_loss": 1.768333077430725,
      "eval_runtime": 37.0244,
      "eval_samples_per_second": 27.009,
      "eval_steps_per_second": 1.134,
      "step": 1660
    },
    {
      "epoch": 1.7062146892655368,
      "grad_norm": 2.7454917430877686,
      "learning_rate": 2.424819185382566e-06,
      "loss": 1.4046,
      "step": 1661
    },
    {
      "epoch": 1.7072419106317411,
      "grad_norm": 2.9186973571777344,
      "learning_rate": 2.4229158736200993e-06,
      "loss": 1.3616,
      "step": 1662
    },
    {
      "epoch": 1.7082691319979455,
      "grad_norm": 2.859410047531128,
      "learning_rate": 2.4210125618576326e-06,
      "loss": 1.3594,
      "step": 1663
    },
    {
      "epoch": 1.70929635336415,
      "grad_norm": 2.8369641304016113,
      "learning_rate": 2.4191092500951656e-06,
      "loss": 1.3924,
      "step": 1664
    },
    {
      "epoch": 1.7103235747303542,
      "grad_norm": 2.8959717750549316,
      "learning_rate": 2.417205938332699e-06,
      "loss": 1.4301,
      "step": 1665
    },
    {
      "epoch": 1.7113507960965588,
      "grad_norm": 2.8268096446990967,
      "learning_rate": 2.4153026265702323e-06,
      "loss": 1.3397,
      "step": 1666
    },
    {
      "epoch": 1.7123780174627632,
      "grad_norm": 2.717118740081787,
      "learning_rate": 2.4133993148077657e-06,
      "loss": 1.3706,
      "step": 1667
    },
    {
      "epoch": 1.7134052388289676,
      "grad_norm": 2.865064859390259,
      "learning_rate": 2.411496003045299e-06,
      "loss": 1.4356,
      "step": 1668
    },
    {
      "epoch": 1.7144324601951721,
      "grad_norm": 2.8731207847595215,
      "learning_rate": 2.4095926912828324e-06,
      "loss": 1.3786,
      "step": 1669
    },
    {
      "epoch": 1.7154596815613765,
      "grad_norm": 2.7409722805023193,
      "learning_rate": 2.4076893795203658e-06,
      "loss": 1.3995,
      "step": 1670
    },
    {
      "epoch": 1.7154596815613765,
      "eval_loss": 1.7796638011932373,
      "eval_runtime": 37.0606,
      "eval_samples_per_second": 26.983,
      "eval_steps_per_second": 1.133,
      "step": 1670
    },
    {
      "epoch": 1.7164869029275809,
      "grad_norm": 2.7279207706451416,
      "learning_rate": 2.405786067757899e-06,
      "loss": 1.3809,
      "step": 1671
    },
    {
      "epoch": 1.7175141242937855,
      "grad_norm": 3.0839486122131348,
      "learning_rate": 2.403882755995432e-06,
      "loss": 1.472,
      "step": 1672
    },
    {
      "epoch": 1.7185413456599896,
      "grad_norm": 2.810347557067871,
      "learning_rate": 2.4019794442329654e-06,
      "loss": 1.3999,
      "step": 1673
    },
    {
      "epoch": 1.7195685670261942,
      "grad_norm": 2.843372106552124,
      "learning_rate": 2.4000761324704988e-06,
      "loss": 1.3667,
      "step": 1674
    },
    {
      "epoch": 1.7205957883923986,
      "grad_norm": 2.8315467834472656,
      "learning_rate": 2.398172820708032e-06,
      "loss": 1.4193,
      "step": 1675
    },
    {
      "epoch": 1.721623009758603,
      "grad_norm": 2.647193431854248,
      "learning_rate": 2.3962695089455655e-06,
      "loss": 1.3826,
      "step": 1676
    },
    {
      "epoch": 1.7226502311248075,
      "grad_norm": 2.8349292278289795,
      "learning_rate": 2.3943661971830984e-06,
      "loss": 1.3776,
      "step": 1677
    },
    {
      "epoch": 1.7236774524910117,
      "grad_norm": 2.723031520843506,
      "learning_rate": 2.3924628854206322e-06,
      "loss": 1.3851,
      "step": 1678
    },
    {
      "epoch": 1.7247046738572163,
      "grad_norm": 2.7333242893218994,
      "learning_rate": 2.3905595736581656e-06,
      "loss": 1.3876,
      "step": 1679
    },
    {
      "epoch": 1.7257318952234206,
      "grad_norm": 2.844578266143799,
      "learning_rate": 2.388656261895699e-06,
      "loss": 1.4353,
      "step": 1680
    },
    {
      "epoch": 1.7257318952234206,
      "eval_loss": 1.7787370681762695,
      "eval_runtime": 37.0536,
      "eval_samples_per_second": 26.988,
      "eval_steps_per_second": 1.133,
      "step": 1680
    },
    {
      "epoch": 1.726759116589625,
      "grad_norm": 2.9545679092407227,
      "learning_rate": 2.386752950133232e-06,
      "loss": 1.4284,
      "step": 1681
    },
    {
      "epoch": 1.7277863379558296,
      "grad_norm": 2.9255502223968506,
      "learning_rate": 2.3848496383707653e-06,
      "loss": 1.4102,
      "step": 1682
    },
    {
      "epoch": 1.7288135593220337,
      "grad_norm": 3.032148838043213,
      "learning_rate": 2.3829463266082986e-06,
      "loss": 1.4044,
      "step": 1683
    },
    {
      "epoch": 1.7298407806882383,
      "grad_norm": 2.960099935531616,
      "learning_rate": 2.381043014845832e-06,
      "loss": 1.4274,
      "step": 1684
    },
    {
      "epoch": 1.7308680020544427,
      "grad_norm": 2.8444836139678955,
      "learning_rate": 2.3791397030833653e-06,
      "loss": 1.4059,
      "step": 1685
    },
    {
      "epoch": 1.731895223420647,
      "grad_norm": 2.7982702255249023,
      "learning_rate": 2.3772363913208983e-06,
      "loss": 1.3865,
      "step": 1686
    },
    {
      "epoch": 1.7329224447868516,
      "grad_norm": 2.762106418609619,
      "learning_rate": 2.3753330795584316e-06,
      "loss": 1.3728,
      "step": 1687
    },
    {
      "epoch": 1.733949666153056,
      "grad_norm": 2.8059279918670654,
      "learning_rate": 2.3734297677959654e-06,
      "loss": 1.3924,
      "step": 1688
    },
    {
      "epoch": 1.7349768875192604,
      "grad_norm": 2.913541316986084,
      "learning_rate": 2.3715264560334984e-06,
      "loss": 1.4347,
      "step": 1689
    },
    {
      "epoch": 1.736004108885465,
      "grad_norm": 2.7966978549957275,
      "learning_rate": 2.3696231442710317e-06,
      "loss": 1.4246,
      "step": 1690
    },
    {
      "epoch": 1.736004108885465,
      "eval_loss": 1.766000747680664,
      "eval_runtime": 37.1692,
      "eval_samples_per_second": 26.904,
      "eval_steps_per_second": 1.13,
      "step": 1690
    },
    {
      "epoch": 1.7370313302516691,
      "grad_norm": 2.795525550842285,
      "learning_rate": 2.367719832508565e-06,
      "loss": 1.4169,
      "step": 1691
    },
    {
      "epoch": 1.7380585516178737,
      "grad_norm": 2.8152341842651367,
      "learning_rate": 2.3658165207460985e-06,
      "loss": 1.438,
      "step": 1692
    },
    {
      "epoch": 1.739085772984078,
      "grad_norm": 2.6966261863708496,
      "learning_rate": 2.363913208983632e-06,
      "loss": 1.4164,
      "step": 1693
    },
    {
      "epoch": 1.7401129943502824,
      "grad_norm": 2.875034809112549,
      "learning_rate": 2.362009897221165e-06,
      "loss": 1.4325,
      "step": 1694
    },
    {
      "epoch": 1.741140215716487,
      "grad_norm": 2.9886977672576904,
      "learning_rate": 2.360106585458698e-06,
      "loss": 1.4277,
      "step": 1695
    },
    {
      "epoch": 1.7421674370826912,
      "grad_norm": 2.735360860824585,
      "learning_rate": 2.3582032736962315e-06,
      "loss": 1.3688,
      "step": 1696
    },
    {
      "epoch": 1.7431946584488958,
      "grad_norm": 2.823995351791382,
      "learning_rate": 2.356299961933765e-06,
      "loss": 1.4062,
      "step": 1697
    },
    {
      "epoch": 1.7442218798151001,
      "grad_norm": 2.7787744998931885,
      "learning_rate": 2.354396650171298e-06,
      "loss": 1.3782,
      "step": 1698
    },
    {
      "epoch": 1.7452491011813045,
      "grad_norm": 2.745983600616455,
      "learning_rate": 2.3524933384088316e-06,
      "loss": 1.3682,
      "step": 1699
    },
    {
      "epoch": 1.746276322547509,
      "grad_norm": 2.853743553161621,
      "learning_rate": 2.350590026646365e-06,
      "loss": 1.4544,
      "step": 1700
    },
    {
      "epoch": 1.746276322547509,
      "eval_loss": 1.7642710208892822,
      "eval_runtime": 37.3227,
      "eval_samples_per_second": 26.793,
      "eval_steps_per_second": 1.125,
      "step": 1700
    },
    {
      "epoch": 1.7473035439137135,
      "grad_norm": 2.8187296390533447,
      "learning_rate": 2.3486867148838983e-06,
      "loss": 1.3874,
      "step": 1701
    },
    {
      "epoch": 1.7483307652799178,
      "grad_norm": 2.766657829284668,
      "learning_rate": 2.3467834031214317e-06,
      "loss": 1.37,
      "step": 1702
    },
    {
      "epoch": 1.7493579866461222,
      "grad_norm": 2.796694755554199,
      "learning_rate": 2.3448800913589646e-06,
      "loss": 1.3909,
      "step": 1703
    },
    {
      "epoch": 1.7503852080123266,
      "grad_norm": 2.8409624099731445,
      "learning_rate": 2.342976779596498e-06,
      "loss": 1.4091,
      "step": 1704
    },
    {
      "epoch": 1.7514124293785311,
      "grad_norm": 2.7160699367523193,
      "learning_rate": 2.3410734678340313e-06,
      "loss": 1.3771,
      "step": 1705
    },
    {
      "epoch": 1.7524396507447355,
      "grad_norm": 2.7986226081848145,
      "learning_rate": 2.3391701560715647e-06,
      "loss": 1.4323,
      "step": 1706
    },
    {
      "epoch": 1.7534668721109399,
      "grad_norm": 2.814997911453247,
      "learning_rate": 2.337266844309098e-06,
      "loss": 1.3903,
      "step": 1707
    },
    {
      "epoch": 1.7544940934771445,
      "grad_norm": 2.7355244159698486,
      "learning_rate": 2.3353635325466314e-06,
      "loss": 1.3955,
      "step": 1708
    },
    {
      "epoch": 1.7555213148433486,
      "grad_norm": 2.8889238834381104,
      "learning_rate": 2.3334602207841648e-06,
      "loss": 1.4037,
      "step": 1709
    },
    {
      "epoch": 1.7565485362095532,
      "grad_norm": 2.9253830909729004,
      "learning_rate": 2.331556909021698e-06,
      "loss": 1.3982,
      "step": 1710
    },
    {
      "epoch": 1.7565485362095532,
      "eval_loss": 1.7730076313018799,
      "eval_runtime": 36.931,
      "eval_samples_per_second": 27.078,
      "eval_steps_per_second": 1.137,
      "step": 1710
    },
    {
      "epoch": 1.7575757575757576,
      "grad_norm": 2.8409945964813232,
      "learning_rate": 2.3296535972592315e-06,
      "loss": 1.4313,
      "step": 1711
    },
    {
      "epoch": 1.758602978941962,
      "grad_norm": 2.7974541187286377,
      "learning_rate": 2.3277502854967644e-06,
      "loss": 1.375,
      "step": 1712
    },
    {
      "epoch": 1.7596302003081665,
      "grad_norm": 2.828486204147339,
      "learning_rate": 2.325846973734298e-06,
      "loss": 1.4406,
      "step": 1713
    },
    {
      "epoch": 1.7606574216743707,
      "grad_norm": 2.8244118690490723,
      "learning_rate": 2.323943661971831e-06,
      "loss": 1.3771,
      "step": 1714
    },
    {
      "epoch": 1.7616846430405753,
      "grad_norm": 2.7830047607421875,
      "learning_rate": 2.3220403502093645e-06,
      "loss": 1.4408,
      "step": 1715
    },
    {
      "epoch": 1.7627118644067796,
      "grad_norm": 2.889033079147339,
      "learning_rate": 2.320137038446898e-06,
      "loss": 1.438,
      "step": 1716
    },
    {
      "epoch": 1.763739085772984,
      "grad_norm": 2.9460082054138184,
      "learning_rate": 2.318233726684431e-06,
      "loss": 1.4424,
      "step": 1717
    },
    {
      "epoch": 1.7647663071391886,
      "grad_norm": 2.8492050170898438,
      "learning_rate": 2.3163304149219646e-06,
      "loss": 1.4354,
      "step": 1718
    },
    {
      "epoch": 1.765793528505393,
      "grad_norm": 2.8115553855895996,
      "learning_rate": 2.314427103159498e-06,
      "loss": 1.4064,
      "step": 1719
    },
    {
      "epoch": 1.7668207498715973,
      "grad_norm": 2.8483548164367676,
      "learning_rate": 2.312523791397031e-06,
      "loss": 1.3915,
      "step": 1720
    },
    {
      "epoch": 1.7668207498715973,
      "eval_loss": 1.770641803741455,
      "eval_runtime": 37.0317,
      "eval_samples_per_second": 27.004,
      "eval_steps_per_second": 1.134,
      "step": 1720
    },
    {
      "epoch": 1.767847971237802,
      "grad_norm": 2.67576003074646,
      "learning_rate": 2.3106204796345643e-06,
      "loss": 1.3665,
      "step": 1721
    },
    {
      "epoch": 1.768875192604006,
      "grad_norm": 2.815186023712158,
      "learning_rate": 2.3087171678720976e-06,
      "loss": 1.3975,
      "step": 1722
    },
    {
      "epoch": 1.7699024139702106,
      "grad_norm": 2.9318761825561523,
      "learning_rate": 2.306813856109631e-06,
      "loss": 1.4506,
      "step": 1723
    },
    {
      "epoch": 1.770929635336415,
      "grad_norm": 2.7512404918670654,
      "learning_rate": 2.3049105443471644e-06,
      "loss": 1.4047,
      "step": 1724
    },
    {
      "epoch": 1.7719568567026194,
      "grad_norm": 2.8040406703948975,
      "learning_rate": 2.3030072325846973e-06,
      "loss": 1.3813,
      "step": 1725
    },
    {
      "epoch": 1.772984078068824,
      "grad_norm": 2.7659668922424316,
      "learning_rate": 2.3011039208222307e-06,
      "loss": 1.373,
      "step": 1726
    },
    {
      "epoch": 1.774011299435028,
      "grad_norm": 2.7986247539520264,
      "learning_rate": 2.299200609059764e-06,
      "loss": 1.3846,
      "step": 1727
    },
    {
      "epoch": 1.7750385208012327,
      "grad_norm": 2.8309032917022705,
      "learning_rate": 2.297297297297298e-06,
      "loss": 1.3854,
      "step": 1728
    },
    {
      "epoch": 1.776065742167437,
      "grad_norm": 2.7988357543945312,
      "learning_rate": 2.2953939855348308e-06,
      "loss": 1.3544,
      "step": 1729
    },
    {
      "epoch": 1.7770929635336414,
      "grad_norm": 2.8682641983032227,
      "learning_rate": 2.293490673772364e-06,
      "loss": 1.4388,
      "step": 1730
    },
    {
      "epoch": 1.7770929635336414,
      "eval_loss": 1.7706880569458008,
      "eval_runtime": 37.0355,
      "eval_samples_per_second": 27.001,
      "eval_steps_per_second": 1.134,
      "step": 1730
    },
    {
      "epoch": 1.778120184899846,
      "grad_norm": 2.8318183422088623,
      "learning_rate": 2.2915873620098975e-06,
      "loss": 1.4257,
      "step": 1731
    },
    {
      "epoch": 1.7791474062660502,
      "grad_norm": 2.8488996028900146,
      "learning_rate": 2.289684050247431e-06,
      "loss": 1.4152,
      "step": 1732
    },
    {
      "epoch": 1.7801746276322548,
      "grad_norm": 2.7117509841918945,
      "learning_rate": 2.287780738484964e-06,
      "loss": 1.3849,
      "step": 1733
    },
    {
      "epoch": 1.7812018489984591,
      "grad_norm": 2.7878878116607666,
      "learning_rate": 2.285877426722497e-06,
      "loss": 1.4071,
      "step": 1734
    },
    {
      "epoch": 1.7822290703646635,
      "grad_norm": 2.876696825027466,
      "learning_rate": 2.2839741149600305e-06,
      "loss": 1.4393,
      "step": 1735
    },
    {
      "epoch": 1.783256291730868,
      "grad_norm": 2.8824517726898193,
      "learning_rate": 2.282070803197564e-06,
      "loss": 1.4225,
      "step": 1736
    },
    {
      "epoch": 1.7842835130970724,
      "grad_norm": 2.8785459995269775,
      "learning_rate": 2.2801674914350972e-06,
      "loss": 1.4093,
      "step": 1737
    },
    {
      "epoch": 1.7853107344632768,
      "grad_norm": 2.7482051849365234,
      "learning_rate": 2.2782641796726306e-06,
      "loss": 1.4004,
      "step": 1738
    },
    {
      "epoch": 1.7863379558294814,
      "grad_norm": 2.812790632247925,
      "learning_rate": 2.276360867910164e-06,
      "loss": 1.4284,
      "step": 1739
    },
    {
      "epoch": 1.7873651771956856,
      "grad_norm": 2.8116066455841064,
      "learning_rate": 2.2744575561476973e-06,
      "loss": 1.3601,
      "step": 1740
    },
    {
      "epoch": 1.7873651771956856,
      "eval_loss": 1.780893325805664,
      "eval_runtime": 37.0422,
      "eval_samples_per_second": 26.996,
      "eval_steps_per_second": 1.134,
      "step": 1740
    },
    {
      "epoch": 1.7883923985618901,
      "grad_norm": 2.7929999828338623,
      "learning_rate": 2.2725542443852307e-06,
      "loss": 1.4034,
      "step": 1741
    },
    {
      "epoch": 1.7894196199280945,
      "grad_norm": 2.8496506214141846,
      "learning_rate": 2.2706509326227636e-06,
      "loss": 1.4414,
      "step": 1742
    },
    {
      "epoch": 1.7904468412942989,
      "grad_norm": 2.7631287574768066,
      "learning_rate": 2.268747620860297e-06,
      "loss": 1.4117,
      "step": 1743
    },
    {
      "epoch": 1.7914740626605035,
      "grad_norm": 2.731952428817749,
      "learning_rate": 2.2668443090978303e-06,
      "loss": 1.4064,
      "step": 1744
    },
    {
      "epoch": 1.7925012840267076,
      "grad_norm": 2.7945852279663086,
      "learning_rate": 2.2649409973353637e-06,
      "loss": 1.4329,
      "step": 1745
    },
    {
      "epoch": 1.7935285053929122,
      "grad_norm": 2.7181437015533447,
      "learning_rate": 2.263037685572897e-06,
      "loss": 1.3821,
      "step": 1746
    },
    {
      "epoch": 1.7945557267591166,
      "grad_norm": 2.792771816253662,
      "learning_rate": 2.26113437381043e-06,
      "loss": 1.3789,
      "step": 1747
    },
    {
      "epoch": 1.795582948125321,
      "grad_norm": 2.9297096729278564,
      "learning_rate": 2.259231062047964e-06,
      "loss": 1.4049,
      "step": 1748
    },
    {
      "epoch": 1.7966101694915255,
      "grad_norm": 2.9424288272857666,
      "learning_rate": 2.257327750285497e-06,
      "loss": 1.4275,
      "step": 1749
    },
    {
      "epoch": 1.7976373908577299,
      "grad_norm": 2.880979537963867,
      "learning_rate": 2.2554244385230305e-06,
      "loss": 1.3911,
      "step": 1750
    },
    {
      "epoch": 1.7976373908577299,
      "eval_loss": 1.7886379957199097,
      "eval_runtime": 37.0413,
      "eval_samples_per_second": 26.997,
      "eval_steps_per_second": 1.134,
      "step": 1750
    },
    {
      "epoch": 1.7986646122239343,
      "grad_norm": 2.9970686435699463,
      "learning_rate": 2.2535211267605635e-06,
      "loss": 1.3976,
      "step": 1751
    },
    {
      "epoch": 1.7996918335901386,
      "grad_norm": 3.0055971145629883,
      "learning_rate": 2.251617814998097e-06,
      "loss": 1.4029,
      "step": 1752
    },
    {
      "epoch": 1.800719054956343,
      "grad_norm": 2.8585352897644043,
      "learning_rate": 2.24971450323563e-06,
      "loss": 1.3279,
      "step": 1753
    },
    {
      "epoch": 1.8017462763225476,
      "grad_norm": 2.8452632427215576,
      "learning_rate": 2.2478111914731636e-06,
      "loss": 1.4141,
      "step": 1754
    },
    {
      "epoch": 1.802773497688752,
      "grad_norm": 2.642672538757324,
      "learning_rate": 2.245907879710697e-06,
      "loss": 1.2864,
      "step": 1755
    },
    {
      "epoch": 1.8038007190549563,
      "grad_norm": 2.8887314796447754,
      "learning_rate": 2.24400456794823e-06,
      "loss": 1.4301,
      "step": 1756
    },
    {
      "epoch": 1.804827940421161,
      "grad_norm": 2.735727071762085,
      "learning_rate": 2.2421012561857632e-06,
      "loss": 1.3908,
      "step": 1757
    },
    {
      "epoch": 1.805855161787365,
      "grad_norm": 2.861366033554077,
      "learning_rate": 2.240197944423297e-06,
      "loss": 1.3757,
      "step": 1758
    },
    {
      "epoch": 1.8068823831535696,
      "grad_norm": 2.7908987998962402,
      "learning_rate": 2.23829463266083e-06,
      "loss": 1.3758,
      "step": 1759
    },
    {
      "epoch": 1.807909604519774,
      "grad_norm": 2.602125644683838,
      "learning_rate": 2.2363913208983633e-06,
      "loss": 1.369,
      "step": 1760
    },
    {
      "epoch": 1.807909604519774,
      "eval_loss": 1.7980055809020996,
      "eval_runtime": 37.0615,
      "eval_samples_per_second": 26.982,
      "eval_steps_per_second": 1.133,
      "step": 1760
    },
    {
      "epoch": 1.8089368258859784,
      "grad_norm": 2.693953275680542,
      "learning_rate": 2.2344880091358967e-06,
      "loss": 1.4095,
      "step": 1761
    },
    {
      "epoch": 1.809964047252183,
      "grad_norm": 2.772446632385254,
      "learning_rate": 2.23258469737343e-06,
      "loss": 1.386,
      "step": 1762
    },
    {
      "epoch": 1.810991268618387,
      "grad_norm": 2.7845568656921387,
      "learning_rate": 2.2306813856109634e-06,
      "loss": 1.4293,
      "step": 1763
    },
    {
      "epoch": 1.8120184899845917,
      "grad_norm": 2.786252498626709,
      "learning_rate": 2.2287780738484963e-06,
      "loss": 1.3805,
      "step": 1764
    },
    {
      "epoch": 1.813045711350796,
      "grad_norm": 2.859053611755371,
      "learning_rate": 2.2268747620860297e-06,
      "loss": 1.4585,
      "step": 1765
    },
    {
      "epoch": 1.8140729327170004,
      "grad_norm": 2.8749146461486816,
      "learning_rate": 2.224971450323563e-06,
      "loss": 1.393,
      "step": 1766
    },
    {
      "epoch": 1.815100154083205,
      "grad_norm": 2.740787982940674,
      "learning_rate": 2.2230681385610964e-06,
      "loss": 1.3309,
      "step": 1767
    },
    {
      "epoch": 1.8161273754494094,
      "grad_norm": 2.834662914276123,
      "learning_rate": 2.2211648267986298e-06,
      "loss": 1.3961,
      "step": 1768
    },
    {
      "epoch": 1.8171545968156138,
      "grad_norm": 2.8033571243286133,
      "learning_rate": 2.219261515036163e-06,
      "loss": 1.3747,
      "step": 1769
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.9068171977996826,
      "learning_rate": 2.2173582032736965e-06,
      "loss": 1.4091,
      "step": 1770
    },
    {
      "epoch": 1.8181818181818183,
      "eval_loss": 1.7923928499221802,
      "eval_runtime": 37.067,
      "eval_samples_per_second": 26.978,
      "eval_steps_per_second": 1.133,
      "step": 1770
    },
    {
      "epoch": 1.8192090395480225,
      "grad_norm": 2.8439955711364746,
      "learning_rate": 2.21545489151123e-06,
      "loss": 1.4281,
      "step": 1771
    },
    {
      "epoch": 1.820236260914227,
      "grad_norm": 2.861231565475464,
      "learning_rate": 2.2135515797487632e-06,
      "loss": 1.3887,
      "step": 1772
    },
    {
      "epoch": 1.8212634822804314,
      "grad_norm": 2.986788034439087,
      "learning_rate": 2.211648267986296e-06,
      "loss": 1.395,
      "step": 1773
    },
    {
      "epoch": 1.8222907036466358,
      "grad_norm": 2.755415678024292,
      "learning_rate": 2.2097449562238295e-06,
      "loss": 1.3885,
      "step": 1774
    },
    {
      "epoch": 1.8233179250128404,
      "grad_norm": 2.7685859203338623,
      "learning_rate": 2.207841644461363e-06,
      "loss": 1.3861,
      "step": 1775
    },
    {
      "epoch": 1.8243451463790445,
      "grad_norm": 2.970322608947754,
      "learning_rate": 2.2059383326988963e-06,
      "loss": 1.4436,
      "step": 1776
    },
    {
      "epoch": 1.8253723677452491,
      "grad_norm": 2.997922658920288,
      "learning_rate": 2.2040350209364296e-06,
      "loss": 1.3978,
      "step": 1777
    },
    {
      "epoch": 1.8263995891114535,
      "grad_norm": 2.734039545059204,
      "learning_rate": 2.2021317091739626e-06,
      "loss": 1.3623,
      "step": 1778
    },
    {
      "epoch": 1.8274268104776579,
      "grad_norm": 2.8555188179016113,
      "learning_rate": 2.2002283974114963e-06,
      "loss": 1.3998,
      "step": 1779
    },
    {
      "epoch": 1.8284540318438625,
      "grad_norm": 2.8464975357055664,
      "learning_rate": 2.1983250856490297e-06,
      "loss": 1.3346,
      "step": 1780
    },
    {
      "epoch": 1.8284540318438625,
      "eval_loss": 1.7876062393188477,
      "eval_runtime": 37.0771,
      "eval_samples_per_second": 26.971,
      "eval_steps_per_second": 1.133,
      "step": 1780
    },
    {
      "epoch": 1.8294812532100666,
      "grad_norm": 3.012516975402832,
      "learning_rate": 2.1964217738865626e-06,
      "loss": 1.3803,
      "step": 1781
    },
    {
      "epoch": 1.8305084745762712,
      "grad_norm": 2.941331148147583,
      "learning_rate": 2.194518462124096e-06,
      "loss": 1.4033,
      "step": 1782
    },
    {
      "epoch": 1.8315356959424756,
      "grad_norm": 3.090533494949341,
      "learning_rate": 2.1926151503616294e-06,
      "loss": 1.443,
      "step": 1783
    },
    {
      "epoch": 1.83256291730868,
      "grad_norm": 2.890427827835083,
      "learning_rate": 2.1907118385991627e-06,
      "loss": 1.4089,
      "step": 1784
    },
    {
      "epoch": 1.8335901386748845,
      "grad_norm": 2.8417627811431885,
      "learning_rate": 2.188808526836696e-06,
      "loss": 1.3722,
      "step": 1785
    },
    {
      "epoch": 1.8346173600410889,
      "grad_norm": 2.7787647247314453,
      "learning_rate": 2.1869052150742295e-06,
      "loss": 1.4011,
      "step": 1786
    },
    {
      "epoch": 1.8356445814072933,
      "grad_norm": 2.7441141605377197,
      "learning_rate": 2.1850019033117624e-06,
      "loss": 1.3262,
      "step": 1787
    },
    {
      "epoch": 1.8366718027734978,
      "grad_norm": 2.873112678527832,
      "learning_rate": 2.1830985915492958e-06,
      "loss": 1.3896,
      "step": 1788
    },
    {
      "epoch": 1.837699024139702,
      "grad_norm": 2.8022804260253906,
      "learning_rate": 2.1811952797868295e-06,
      "loss": 1.4035,
      "step": 1789
    },
    {
      "epoch": 1.8387262455059066,
      "grad_norm": 2.8220114707946777,
      "learning_rate": 2.1792919680243625e-06,
      "loss": 1.3748,
      "step": 1790
    },
    {
      "epoch": 1.8387262455059066,
      "eval_loss": 1.7839077711105347,
      "eval_runtime": 37.0886,
      "eval_samples_per_second": 26.962,
      "eval_steps_per_second": 1.132,
      "step": 1790
    },
    {
      "epoch": 1.839753466872111,
      "grad_norm": 2.769643783569336,
      "learning_rate": 2.177388656261896e-06,
      "loss": 1.4084,
      "step": 1791
    },
    {
      "epoch": 1.8407806882383153,
      "grad_norm": 2.8555970191955566,
      "learning_rate": 2.1754853444994292e-06,
      "loss": 1.4772,
      "step": 1792
    },
    {
      "epoch": 1.84180790960452,
      "grad_norm": 2.777965545654297,
      "learning_rate": 2.1735820327369626e-06,
      "loss": 1.3024,
      "step": 1793
    },
    {
      "epoch": 1.842835130970724,
      "grad_norm": 2.8747634887695312,
      "learning_rate": 2.171678720974496e-06,
      "loss": 1.4195,
      "step": 1794
    },
    {
      "epoch": 1.8438623523369286,
      "grad_norm": 2.8822247982025146,
      "learning_rate": 2.169775409212029e-06,
      "loss": 1.4151,
      "step": 1795
    },
    {
      "epoch": 1.844889573703133,
      "grad_norm": 2.8203372955322266,
      "learning_rate": 2.1678720974495622e-06,
      "loss": 1.4223,
      "step": 1796
    },
    {
      "epoch": 1.8459167950693374,
      "grad_norm": 2.632070779800415,
      "learning_rate": 2.1659687856870956e-06,
      "loss": 1.4141,
      "step": 1797
    },
    {
      "epoch": 1.846944016435542,
      "grad_norm": 2.746530771255493,
      "learning_rate": 2.164065473924629e-06,
      "loss": 1.4351,
      "step": 1798
    },
    {
      "epoch": 1.8479712378017463,
      "grad_norm": 2.7693254947662354,
      "learning_rate": 2.1621621621621623e-06,
      "loss": 1.3588,
      "step": 1799
    },
    {
      "epoch": 1.8489984591679507,
      "grad_norm": 2.7944774627685547,
      "learning_rate": 2.1602588503996957e-06,
      "loss": 1.4181,
      "step": 1800
    },
    {
      "epoch": 1.8489984591679507,
      "eval_loss": 1.780879259109497,
      "eval_runtime": 37.2375,
      "eval_samples_per_second": 26.855,
      "eval_steps_per_second": 1.128,
      "step": 1800
    },
    {
      "epoch": 1.850025680534155,
      "grad_norm": 2.827951669692993,
      "learning_rate": 2.158355538637229e-06,
      "loss": 1.3857,
      "step": 1801
    },
    {
      "epoch": 1.8510529019003594,
      "grad_norm": 2.7634339332580566,
      "learning_rate": 2.1564522268747624e-06,
      "loss": 1.387,
      "step": 1802
    },
    {
      "epoch": 1.852080123266564,
      "grad_norm": 3.014620542526245,
      "learning_rate": 2.1545489151122958e-06,
      "loss": 1.4595,
      "step": 1803
    },
    {
      "epoch": 1.8531073446327684,
      "grad_norm": 2.7546815872192383,
      "learning_rate": 2.1526456033498287e-06,
      "loss": 1.4098,
      "step": 1804
    },
    {
      "epoch": 1.8541345659989728,
      "grad_norm": 2.7237050533294678,
      "learning_rate": 2.150742291587362e-06,
      "loss": 1.4099,
      "step": 1805
    },
    {
      "epoch": 1.8551617873651773,
      "grad_norm": 2.8354947566986084,
      "learning_rate": 2.1488389798248954e-06,
      "loss": 1.395,
      "step": 1806
    },
    {
      "epoch": 1.8561890087313815,
      "grad_norm": 2.863032102584839,
      "learning_rate": 2.146935668062429e-06,
      "loss": 1.4275,
      "step": 1807
    },
    {
      "epoch": 1.857216230097586,
      "grad_norm": 2.6963112354278564,
      "learning_rate": 2.145032356299962e-06,
      "loss": 1.3567,
      "step": 1808
    },
    {
      "epoch": 1.8582434514637904,
      "grad_norm": 2.7628366947174072,
      "learning_rate": 2.1431290445374955e-06,
      "loss": 1.4036,
      "step": 1809
    },
    {
      "epoch": 1.8592706728299948,
      "grad_norm": 2.790919065475464,
      "learning_rate": 2.141225732775029e-06,
      "loss": 1.3679,
      "step": 1810
    },
    {
      "epoch": 1.8592706728299948,
      "eval_loss": 1.7804644107818604,
      "eval_runtime": 36.9549,
      "eval_samples_per_second": 27.06,
      "eval_steps_per_second": 1.137,
      "step": 1810
    },
    {
      "epoch": 1.8602978941961994,
      "grad_norm": 2.913525104522705,
      "learning_rate": 2.1393224210125623e-06,
      "loss": 1.4015,
      "step": 1811
    },
    {
      "epoch": 1.8613251155624035,
      "grad_norm": 2.8357040882110596,
      "learning_rate": 2.137419109250095e-06,
      "loss": 1.3912,
      "step": 1812
    },
    {
      "epoch": 1.8623523369286081,
      "grad_norm": 2.989529848098755,
      "learning_rate": 2.1355157974876286e-06,
      "loss": 1.4397,
      "step": 1813
    },
    {
      "epoch": 1.8633795582948125,
      "grad_norm": 2.8843417167663574,
      "learning_rate": 2.133612485725162e-06,
      "loss": 1.4343,
      "step": 1814
    },
    {
      "epoch": 1.8644067796610169,
      "grad_norm": 2.882571220397949,
      "learning_rate": 2.1317091739626953e-06,
      "loss": 1.342,
      "step": 1815
    },
    {
      "epoch": 1.8654340010272215,
      "grad_norm": 2.8986611366271973,
      "learning_rate": 2.1298058622002286e-06,
      "loss": 1.3687,
      "step": 1816
    },
    {
      "epoch": 1.8664612223934258,
      "grad_norm": 2.8779523372650146,
      "learning_rate": 2.1279025504377616e-06,
      "loss": 1.4005,
      "step": 1817
    },
    {
      "epoch": 1.8674884437596302,
      "grad_norm": 2.7829904556274414,
      "learning_rate": 2.125999238675295e-06,
      "loss": 1.3843,
      "step": 1818
    },
    {
      "epoch": 1.8685156651258348,
      "grad_norm": 3.0068447589874268,
      "learning_rate": 2.1240959269128287e-06,
      "loss": 1.4347,
      "step": 1819
    },
    {
      "epoch": 1.869542886492039,
      "grad_norm": 2.945225715637207,
      "learning_rate": 2.122192615150362e-06,
      "loss": 1.3934,
      "step": 1820
    },
    {
      "epoch": 1.869542886492039,
      "eval_loss": 1.7796295881271362,
      "eval_runtime": 37.0423,
      "eval_samples_per_second": 26.996,
      "eval_steps_per_second": 1.134,
      "step": 1820
    },
    {
      "epoch": 1.8705701078582435,
      "grad_norm": 2.79140305519104,
      "learning_rate": 2.120289303387895e-06,
      "loss": 1.3387,
      "step": 1821
    },
    {
      "epoch": 1.8715973292244479,
      "grad_norm": 2.897294282913208,
      "learning_rate": 2.1183859916254284e-06,
      "loss": 1.3905,
      "step": 1822
    },
    {
      "epoch": 1.8726245505906522,
      "grad_norm": 2.8181400299072266,
      "learning_rate": 2.1164826798629618e-06,
      "loss": 1.4219,
      "step": 1823
    },
    {
      "epoch": 1.8736517719568568,
      "grad_norm": 2.695333480834961,
      "learning_rate": 2.114579368100495e-06,
      "loss": 1.38,
      "step": 1824
    },
    {
      "epoch": 1.874678993323061,
      "grad_norm": 2.7668285369873047,
      "learning_rate": 2.1126760563380285e-06,
      "loss": 1.3613,
      "step": 1825
    },
    {
      "epoch": 1.8757062146892656,
      "grad_norm": 2.723405599594116,
      "learning_rate": 2.1107727445755614e-06,
      "loss": 1.3344,
      "step": 1826
    },
    {
      "epoch": 1.87673343605547,
      "grad_norm": 3.01149320602417,
      "learning_rate": 2.1088694328130948e-06,
      "loss": 1.4672,
      "step": 1827
    },
    {
      "epoch": 1.8777606574216743,
      "grad_norm": 2.7828991413116455,
      "learning_rate": 2.106966121050628e-06,
      "loss": 1.3671,
      "step": 1828
    },
    {
      "epoch": 1.878787878787879,
      "grad_norm": 2.792518138885498,
      "learning_rate": 2.1050628092881615e-06,
      "loss": 1.4182,
      "step": 1829
    },
    {
      "epoch": 1.879815100154083,
      "grad_norm": 2.935760736465454,
      "learning_rate": 2.103159497525695e-06,
      "loss": 1.4623,
      "step": 1830
    },
    {
      "epoch": 1.879815100154083,
      "eval_loss": 1.7818807363510132,
      "eval_runtime": 37.0809,
      "eval_samples_per_second": 26.968,
      "eval_steps_per_second": 1.133,
      "step": 1830
    },
    {
      "epoch": 1.8808423215202876,
      "grad_norm": 2.849071741104126,
      "learning_rate": 2.1012561857632282e-06,
      "loss": 1.3777,
      "step": 1831
    },
    {
      "epoch": 1.881869542886492,
      "grad_norm": 2.978045701980591,
      "learning_rate": 2.0993528740007616e-06,
      "loss": 1.4325,
      "step": 1832
    },
    {
      "epoch": 1.8828967642526964,
      "grad_norm": 2.8763439655303955,
      "learning_rate": 2.097449562238295e-06,
      "loss": 1.4417,
      "step": 1833
    },
    {
      "epoch": 1.883923985618901,
      "grad_norm": 2.891221523284912,
      "learning_rate": 2.095546250475828e-06,
      "loss": 1.3881,
      "step": 1834
    },
    {
      "epoch": 1.8849512069851053,
      "grad_norm": 2.9743025302886963,
      "learning_rate": 2.0936429387133613e-06,
      "loss": 1.3956,
      "step": 1835
    },
    {
      "epoch": 1.8859784283513097,
      "grad_norm": 2.7557995319366455,
      "learning_rate": 2.0917396269508946e-06,
      "loss": 1.4511,
      "step": 1836
    },
    {
      "epoch": 1.8870056497175143,
      "grad_norm": 2.878021240234375,
      "learning_rate": 2.089836315188428e-06,
      "loss": 1.4206,
      "step": 1837
    },
    {
      "epoch": 1.8880328710837184,
      "grad_norm": 2.9022324085235596,
      "learning_rate": 2.0879330034259614e-06,
      "loss": 1.4256,
      "step": 1838
    },
    {
      "epoch": 1.889060092449923,
      "grad_norm": 2.8805174827575684,
      "learning_rate": 2.0860296916634947e-06,
      "loss": 1.3752,
      "step": 1839
    },
    {
      "epoch": 1.8900873138161274,
      "grad_norm": 2.887437105178833,
      "learning_rate": 2.084126379901028e-06,
      "loss": 1.4636,
      "step": 1840
    },
    {
      "epoch": 1.8900873138161274,
      "eval_loss": 1.775463581085205,
      "eval_runtime": 37.0928,
      "eval_samples_per_second": 26.959,
      "eval_steps_per_second": 1.132,
      "step": 1840
    },
    {
      "epoch": 1.8911145351823317,
      "grad_norm": 2.8477346897125244,
      "learning_rate": 2.0822230681385614e-06,
      "loss": 1.3653,
      "step": 1841
    },
    {
      "epoch": 1.8921417565485363,
      "grad_norm": 2.939516067504883,
      "learning_rate": 2.080319756376095e-06,
      "loss": 1.4216,
      "step": 1842
    },
    {
      "epoch": 1.8931689779147405,
      "grad_norm": 2.869184970855713,
      "learning_rate": 2.0784164446136277e-06,
      "loss": 1.4138,
      "step": 1843
    },
    {
      "epoch": 1.894196199280945,
      "grad_norm": 2.7639729976654053,
      "learning_rate": 2.076513132851161e-06,
      "loss": 1.3837,
      "step": 1844
    },
    {
      "epoch": 1.8952234206471494,
      "grad_norm": 2.9656383991241455,
      "learning_rate": 2.0746098210886945e-06,
      "loss": 1.4168,
      "step": 1845
    },
    {
      "epoch": 1.8962506420133538,
      "grad_norm": 2.80155873298645,
      "learning_rate": 2.072706509326228e-06,
      "loss": 1.3879,
      "step": 1846
    },
    {
      "epoch": 1.8972778633795584,
      "grad_norm": 2.742859363555908,
      "learning_rate": 2.070803197563761e-06,
      "loss": 1.397,
      "step": 1847
    },
    {
      "epoch": 1.8983050847457628,
      "grad_norm": 2.790229082107544,
      "learning_rate": 2.068899885801294e-06,
      "loss": 1.3894,
      "step": 1848
    },
    {
      "epoch": 1.8993323061119671,
      "grad_norm": 2.87921142578125,
      "learning_rate": 2.066996574038828e-06,
      "loss": 1.4122,
      "step": 1849
    },
    {
      "epoch": 1.9003595274781715,
      "grad_norm": 2.937082052230835,
      "learning_rate": 2.0650932622763613e-06,
      "loss": 1.435,
      "step": 1850
    },
    {
      "epoch": 1.9003595274781715,
      "eval_loss": 1.7634403705596924,
      "eval_runtime": 37.0377,
      "eval_samples_per_second": 27.0,
      "eval_steps_per_second": 1.134,
      "step": 1850
    },
    {
      "epoch": 1.9013867488443759,
      "grad_norm": 2.830686569213867,
      "learning_rate": 2.0631899505138942e-06,
      "loss": 1.3248,
      "step": 1851
    },
    {
      "epoch": 1.9024139702105805,
      "grad_norm": 2.878131151199341,
      "learning_rate": 2.0612866387514276e-06,
      "loss": 1.3778,
      "step": 1852
    },
    {
      "epoch": 1.9034411915767848,
      "grad_norm": 2.894076108932495,
      "learning_rate": 2.059383326988961e-06,
      "loss": 1.4073,
      "step": 1853
    },
    {
      "epoch": 1.9044684129429892,
      "grad_norm": 2.816612958908081,
      "learning_rate": 2.0574800152264943e-06,
      "loss": 1.3873,
      "step": 1854
    },
    {
      "epoch": 1.9054956343091938,
      "grad_norm": 2.905581474304199,
      "learning_rate": 2.0555767034640277e-06,
      "loss": 1.3644,
      "step": 1855
    },
    {
      "epoch": 1.906522855675398,
      "grad_norm": 2.7126173973083496,
      "learning_rate": 2.053673391701561e-06,
      "loss": 1.3402,
      "step": 1856
    },
    {
      "epoch": 1.9075500770416025,
      "grad_norm": 2.8092000484466553,
      "learning_rate": 2.051770079939094e-06,
      "loss": 1.4095,
      "step": 1857
    },
    {
      "epoch": 1.9085772984078069,
      "grad_norm": 2.7571353912353516,
      "learning_rate": 2.0498667681766273e-06,
      "loss": 1.3657,
      "step": 1858
    },
    {
      "epoch": 1.9096045197740112,
      "grad_norm": 2.8373515605926514,
      "learning_rate": 2.047963456414161e-06,
      "loss": 1.3643,
      "step": 1859
    },
    {
      "epoch": 1.9106317411402158,
      "grad_norm": 2.825209379196167,
      "learning_rate": 2.046060144651694e-06,
      "loss": 1.3859,
      "step": 1860
    },
    {
      "epoch": 1.9106317411402158,
      "eval_loss": 1.7664082050323486,
      "eval_runtime": 37.1245,
      "eval_samples_per_second": 26.936,
      "eval_steps_per_second": 1.131,
      "step": 1860
    },
    {
      "epoch": 1.91165896250642,
      "grad_norm": 2.7947168350219727,
      "learning_rate": 2.0441568328892274e-06,
      "loss": 1.4002,
      "step": 1861
    },
    {
      "epoch": 1.9126861838726246,
      "grad_norm": 2.8314290046691895,
      "learning_rate": 2.0422535211267608e-06,
      "loss": 1.4017,
      "step": 1862
    },
    {
      "epoch": 1.913713405238829,
      "grad_norm": 2.7938215732574463,
      "learning_rate": 2.040350209364294e-06,
      "loss": 1.379,
      "step": 1863
    },
    {
      "epoch": 1.9147406266050333,
      "grad_norm": 2.7426047325134277,
      "learning_rate": 2.0384468976018275e-06,
      "loss": 1.3615,
      "step": 1864
    },
    {
      "epoch": 1.915767847971238,
      "grad_norm": 2.916954278945923,
      "learning_rate": 2.0365435858393604e-06,
      "loss": 1.3756,
      "step": 1865
    },
    {
      "epoch": 1.9167950693374423,
      "grad_norm": 2.7564570903778076,
      "learning_rate": 2.034640274076894e-06,
      "loss": 1.3716,
      "step": 1866
    },
    {
      "epoch": 1.9178222907036466,
      "grad_norm": 2.911621332168579,
      "learning_rate": 2.032736962314427e-06,
      "loss": 1.4378,
      "step": 1867
    },
    {
      "epoch": 1.9188495120698512,
      "grad_norm": 2.805112361907959,
      "learning_rate": 2.0308336505519605e-06,
      "loss": 1.3823,
      "step": 1868
    },
    {
      "epoch": 1.9198767334360554,
      "grad_norm": 3.079866647720337,
      "learning_rate": 2.028930338789494e-06,
      "loss": 1.4388,
      "step": 1869
    },
    {
      "epoch": 1.92090395480226,
      "grad_norm": 2.9153823852539062,
      "learning_rate": 2.0270270270270273e-06,
      "loss": 1.359,
      "step": 1870
    },
    {
      "epoch": 1.92090395480226,
      "eval_loss": 1.7608962059020996,
      "eval_runtime": 37.1265,
      "eval_samples_per_second": 26.935,
      "eval_steps_per_second": 1.131,
      "step": 1870
    },
    {
      "epoch": 1.9219311761684643,
      "grad_norm": 2.8992526531219482,
      "learning_rate": 2.0251237152645606e-06,
      "loss": 1.3814,
      "step": 1871
    },
    {
      "epoch": 1.9229583975346687,
      "grad_norm": 2.8838071823120117,
      "learning_rate": 2.023220403502094e-06,
      "loss": 1.3642,
      "step": 1872
    },
    {
      "epoch": 1.9239856189008733,
      "grad_norm": 2.8734290599823,
      "learning_rate": 2.0213170917396273e-06,
      "loss": 1.3366,
      "step": 1873
    },
    {
      "epoch": 1.9250128402670774,
      "grad_norm": 2.929142951965332,
      "learning_rate": 2.0194137799771603e-06,
      "loss": 1.3725,
      "step": 1874
    },
    {
      "epoch": 1.926040061633282,
      "grad_norm": 2.984711170196533,
      "learning_rate": 2.0175104682146937e-06,
      "loss": 1.3825,
      "step": 1875
    },
    {
      "epoch": 1.9270672829994864,
      "grad_norm": 2.916740894317627,
      "learning_rate": 2.015607156452227e-06,
      "loss": 1.3261,
      "step": 1876
    },
    {
      "epoch": 1.9280945043656907,
      "grad_norm": 2.8688554763793945,
      "learning_rate": 2.0137038446897604e-06,
      "loss": 1.365,
      "step": 1877
    },
    {
      "epoch": 1.9291217257318953,
      "grad_norm": 2.932525396347046,
      "learning_rate": 2.0118005329272937e-06,
      "loss": 1.3654,
      "step": 1878
    },
    {
      "epoch": 1.9301489470980995,
      "grad_norm": 2.8874945640563965,
      "learning_rate": 2.0098972211648267e-06,
      "loss": 1.3745,
      "step": 1879
    },
    {
      "epoch": 1.931176168464304,
      "grad_norm": 2.7608230113983154,
      "learning_rate": 2.0079939094023605e-06,
      "loss": 1.3718,
      "step": 1880
    },
    {
      "epoch": 1.931176168464304,
      "eval_loss": 1.7662792205810547,
      "eval_runtime": 37.1426,
      "eval_samples_per_second": 26.923,
      "eval_steps_per_second": 1.131,
      "step": 1880
    },
    {
      "epoch": 1.9322033898305084,
      "grad_norm": 2.7790932655334473,
      "learning_rate": 2.006090597639894e-06,
      "loss": 1.4227,
      "step": 1881
    },
    {
      "epoch": 1.9332306111967128,
      "grad_norm": 2.9143898487091064,
      "learning_rate": 2.0041872858774268e-06,
      "loss": 1.4396,
      "step": 1882
    },
    {
      "epoch": 1.9342578325629174,
      "grad_norm": 2.744253158569336,
      "learning_rate": 2.00228397411496e-06,
      "loss": 1.4123,
      "step": 1883
    },
    {
      "epoch": 1.9352850539291218,
      "grad_norm": 2.7352776527404785,
      "learning_rate": 2.0003806623524935e-06,
      "loss": 1.3936,
      "step": 1884
    },
    {
      "epoch": 1.9363122752953261,
      "grad_norm": 2.7074668407440186,
      "learning_rate": 1.998477350590027e-06,
      "loss": 1.3201,
      "step": 1885
    },
    {
      "epoch": 1.9373394966615307,
      "grad_norm": 2.848219871520996,
      "learning_rate": 1.9965740388275602e-06,
      "loss": 1.4048,
      "step": 1886
    },
    {
      "epoch": 1.9383667180277349,
      "grad_norm": 2.7833354473114014,
      "learning_rate": 1.994670727065093e-06,
      "loss": 1.4007,
      "step": 1887
    },
    {
      "epoch": 1.9393939393939394,
      "grad_norm": 2.83203387260437,
      "learning_rate": 1.9927674153026265e-06,
      "loss": 1.3471,
      "step": 1888
    },
    {
      "epoch": 1.9404211607601438,
      "grad_norm": 2.99426007270813,
      "learning_rate": 1.99086410354016e-06,
      "loss": 1.407,
      "step": 1889
    },
    {
      "epoch": 1.9414483821263482,
      "grad_norm": 2.8964831829071045,
      "learning_rate": 1.9889607917776937e-06,
      "loss": 1.3784,
      "step": 1890
    },
    {
      "epoch": 1.9414483821263482,
      "eval_loss": 1.7859200239181519,
      "eval_runtime": 37.16,
      "eval_samples_per_second": 26.911,
      "eval_steps_per_second": 1.13,
      "step": 1890
    },
    {
      "epoch": 1.9424756034925528,
      "grad_norm": 2.938148021697998,
      "learning_rate": 1.9870574800152266e-06,
      "loss": 1.3932,
      "step": 1891
    },
    {
      "epoch": 1.943502824858757,
      "grad_norm": 2.9096012115478516,
      "learning_rate": 1.98515416825276e-06,
      "loss": 1.4051,
      "step": 1892
    },
    {
      "epoch": 1.9445300462249615,
      "grad_norm": 2.8964526653289795,
      "learning_rate": 1.9832508564902933e-06,
      "loss": 1.3961,
      "step": 1893
    },
    {
      "epoch": 1.9455572675911659,
      "grad_norm": 2.681792974472046,
      "learning_rate": 1.9813475447278267e-06,
      "loss": 1.3175,
      "step": 1894
    },
    {
      "epoch": 1.9465844889573702,
      "grad_norm": 2.8340206146240234,
      "learning_rate": 1.97944423296536e-06,
      "loss": 1.3955,
      "step": 1895
    },
    {
      "epoch": 1.9476117103235748,
      "grad_norm": 2.9196391105651855,
      "learning_rate": 1.977540921202893e-06,
      "loss": 1.3833,
      "step": 1896
    },
    {
      "epoch": 1.9486389316897792,
      "grad_norm": 2.7774484157562256,
      "learning_rate": 1.9756376094404264e-06,
      "loss": 1.3645,
      "step": 1897
    },
    {
      "epoch": 1.9496661530559836,
      "grad_norm": 2.7222559452056885,
      "learning_rate": 1.9737342976779597e-06,
      "loss": 1.342,
      "step": 1898
    },
    {
      "epoch": 1.950693374422188,
      "grad_norm": 2.816084623336792,
      "learning_rate": 1.971830985915493e-06,
      "loss": 1.4255,
      "step": 1899
    },
    {
      "epoch": 1.9517205957883923,
      "grad_norm": 2.7530157566070557,
      "learning_rate": 1.9699276741530264e-06,
      "loss": 1.3792,
      "step": 1900
    },
    {
      "epoch": 1.9517205957883923,
      "eval_loss": 1.7724789381027222,
      "eval_runtime": 37.3499,
      "eval_samples_per_second": 26.774,
      "eval_steps_per_second": 1.125,
      "step": 1900
    },
    {
      "epoch": 1.9527478171545969,
      "grad_norm": 2.7674543857574463,
      "learning_rate": 1.96802436239056e-06,
      "loss": 1.3888,
      "step": 1901
    },
    {
      "epoch": 1.9537750385208013,
      "grad_norm": 2.9137115478515625,
      "learning_rate": 1.966121050628093e-06,
      "loss": 1.3656,
      "step": 1902
    },
    {
      "epoch": 1.9548022598870056,
      "grad_norm": 2.8127288818359375,
      "learning_rate": 1.9642177388656265e-06,
      "loss": 1.4109,
      "step": 1903
    },
    {
      "epoch": 1.9558294812532102,
      "grad_norm": 2.906191110610962,
      "learning_rate": 1.9623144271031595e-06,
      "loss": 1.4258,
      "step": 1904
    },
    {
      "epoch": 1.9568567026194144,
      "grad_norm": 2.9624578952789307,
      "learning_rate": 1.960411115340693e-06,
      "loss": 1.4328,
      "step": 1905
    },
    {
      "epoch": 1.957883923985619,
      "grad_norm": 2.8785276412963867,
      "learning_rate": 1.958507803578226e-06,
      "loss": 1.3912,
      "step": 1906
    },
    {
      "epoch": 1.9589111453518233,
      "grad_norm": 2.848287343978882,
      "learning_rate": 1.9566044918157596e-06,
      "loss": 1.4052,
      "step": 1907
    },
    {
      "epoch": 1.9599383667180277,
      "grad_norm": 2.766186237335205,
      "learning_rate": 1.954701180053293e-06,
      "loss": 1.4064,
      "step": 1908
    },
    {
      "epoch": 1.9609655880842323,
      "grad_norm": 2.8182034492492676,
      "learning_rate": 1.952797868290826e-06,
      "loss": 1.3724,
      "step": 1909
    },
    {
      "epoch": 1.9619928094504364,
      "grad_norm": 2.826488494873047,
      "learning_rate": 1.9508945565283596e-06,
      "loss": 1.421,
      "step": 1910
    },
    {
      "epoch": 1.9619928094504364,
      "eval_loss": 1.780763864517212,
      "eval_runtime": 36.9249,
      "eval_samples_per_second": 27.082,
      "eval_steps_per_second": 1.137,
      "step": 1910
    },
    {
      "epoch": 1.963020030816641,
      "grad_norm": 2.876213312149048,
      "learning_rate": 1.948991244765893e-06,
      "loss": 1.3885,
      "step": 1911
    },
    {
      "epoch": 1.9640472521828454,
      "grad_norm": 2.897611141204834,
      "learning_rate": 1.9470879330034264e-06,
      "loss": 1.4126,
      "step": 1912
    },
    {
      "epoch": 1.9650744735490497,
      "grad_norm": 2.806605100631714,
      "learning_rate": 1.9451846212409593e-06,
      "loss": 1.3506,
      "step": 1913
    },
    {
      "epoch": 1.9661016949152543,
      "grad_norm": 2.879645824432373,
      "learning_rate": 1.9432813094784927e-06,
      "loss": 1.409,
      "step": 1914
    },
    {
      "epoch": 1.9671289162814587,
      "grad_norm": 2.871046304702759,
      "learning_rate": 1.941377997716026e-06,
      "loss": 1.3417,
      "step": 1915
    },
    {
      "epoch": 1.968156137647663,
      "grad_norm": 2.8778395652770996,
      "learning_rate": 1.9394746859535594e-06,
      "loss": 1.3779,
      "step": 1916
    },
    {
      "epoch": 1.9691833590138677,
      "grad_norm": 2.8736093044281006,
      "learning_rate": 1.9375713741910928e-06,
      "loss": 1.3355,
      "step": 1917
    },
    {
      "epoch": 1.9702105803800718,
      "grad_norm": 2.822600841522217,
      "learning_rate": 1.9356680624286257e-06,
      "loss": 1.3779,
      "step": 1918
    },
    {
      "epoch": 1.9712378017462764,
      "grad_norm": 2.779498338699341,
      "learning_rate": 1.933764750666159e-06,
      "loss": 1.3328,
      "step": 1919
    },
    {
      "epoch": 1.9722650231124808,
      "grad_norm": 2.9568161964416504,
      "learning_rate": 1.931861438903693e-06,
      "loss": 1.3902,
      "step": 1920
    },
    {
      "epoch": 1.9722650231124808,
      "eval_loss": 1.769534707069397,
      "eval_runtime": 37.0243,
      "eval_samples_per_second": 27.009,
      "eval_steps_per_second": 1.134,
      "step": 1920
    },
    {
      "epoch": 1.9732922444786851,
      "grad_norm": 2.7917582988739014,
      "learning_rate": 1.929958127141226e-06,
      "loss": 1.3517,
      "step": 1921
    },
    {
      "epoch": 1.9743194658448897,
      "grad_norm": 2.8624138832092285,
      "learning_rate": 1.928054815378759e-06,
      "loss": 1.3765,
      "step": 1922
    },
    {
      "epoch": 1.9753466872110939,
      "grad_norm": 2.874619960784912,
      "learning_rate": 1.9261515036162925e-06,
      "loss": 1.4548,
      "step": 1923
    },
    {
      "epoch": 1.9763739085772984,
      "grad_norm": 2.921558141708374,
      "learning_rate": 1.924248191853826e-06,
      "loss": 1.4122,
      "step": 1924
    },
    {
      "epoch": 1.9774011299435028,
      "grad_norm": 2.853748321533203,
      "learning_rate": 1.9223448800913592e-06,
      "loss": 1.3909,
      "step": 1925
    },
    {
      "epoch": 1.9784283513097072,
      "grad_norm": 2.828735828399658,
      "learning_rate": 1.9204415683288926e-06,
      "loss": 1.3748,
      "step": 1926
    },
    {
      "epoch": 1.9794555726759118,
      "grad_norm": 2.8109824657440186,
      "learning_rate": 1.9185382565664255e-06,
      "loss": 1.4051,
      "step": 1927
    },
    {
      "epoch": 1.980482794042116,
      "grad_norm": 3.035722494125366,
      "learning_rate": 1.916634944803959e-06,
      "loss": 1.42,
      "step": 1928
    },
    {
      "epoch": 1.9815100154083205,
      "grad_norm": 2.9289391040802,
      "learning_rate": 1.9147316330414923e-06,
      "loss": 1.3965,
      "step": 1929
    },
    {
      "epoch": 1.9825372367745249,
      "grad_norm": 2.853398561477661,
      "learning_rate": 1.9128283212790256e-06,
      "loss": 1.3869,
      "step": 1930
    },
    {
      "epoch": 1.9825372367745249,
      "eval_loss": 1.7659740447998047,
      "eval_runtime": 37.0479,
      "eval_samples_per_second": 26.992,
      "eval_steps_per_second": 1.134,
      "step": 1930
    },
    {
      "epoch": 1.9835644581407292,
      "grad_norm": 2.9305245876312256,
      "learning_rate": 1.910925009516559e-06,
      "loss": 1.4265,
      "step": 1931
    },
    {
      "epoch": 1.9845916795069338,
      "grad_norm": 2.8339171409606934,
      "learning_rate": 1.9090216977540924e-06,
      "loss": 1.3934,
      "step": 1932
    },
    {
      "epoch": 1.9856189008731382,
      "grad_norm": 2.929807424545288,
      "learning_rate": 1.9071183859916257e-06,
      "loss": 1.4339,
      "step": 1933
    },
    {
      "epoch": 1.9866461222393426,
      "grad_norm": 2.9070687294006348,
      "learning_rate": 1.9052150742291589e-06,
      "loss": 1.4004,
      "step": 1934
    },
    {
      "epoch": 1.9876733436055471,
      "grad_norm": 2.9756429195404053,
      "learning_rate": 1.9033117624666922e-06,
      "loss": 1.404,
      "step": 1935
    },
    {
      "epoch": 1.9887005649717513,
      "grad_norm": 2.831841468811035,
      "learning_rate": 1.9014084507042254e-06,
      "loss": 1.3668,
      "step": 1936
    },
    {
      "epoch": 1.9897277863379559,
      "grad_norm": 2.8649723529815674,
      "learning_rate": 1.8995051389417587e-06,
      "loss": 1.3625,
      "step": 1937
    },
    {
      "epoch": 1.9907550077041603,
      "grad_norm": 2.7570254802703857,
      "learning_rate": 1.8976018271792921e-06,
      "loss": 1.3427,
      "step": 1938
    },
    {
      "epoch": 1.9917822290703646,
      "grad_norm": 2.9655392169952393,
      "learning_rate": 1.8956985154168253e-06,
      "loss": 1.4039,
      "step": 1939
    },
    {
      "epoch": 1.9928094504365692,
      "grad_norm": 2.840308427810669,
      "learning_rate": 1.8937952036543588e-06,
      "loss": 1.3678,
      "step": 1940
    },
    {
      "epoch": 1.9928094504365692,
      "eval_loss": 1.7668676376342773,
      "eval_runtime": 37.0279,
      "eval_samples_per_second": 27.007,
      "eval_steps_per_second": 1.134,
      "step": 1940
    },
    {
      "epoch": 1.9938366718027734,
      "grad_norm": 2.7608296871185303,
      "learning_rate": 1.8918918918918922e-06,
      "loss": 1.3353,
      "step": 1941
    },
    {
      "epoch": 1.994863893168978,
      "grad_norm": 2.861088275909424,
      "learning_rate": 1.8899885801294253e-06,
      "loss": 1.3866,
      "step": 1942
    },
    {
      "epoch": 1.9958911145351823,
      "grad_norm": 2.8978993892669678,
      "learning_rate": 1.8880852683669587e-06,
      "loss": 1.4054,
      "step": 1943
    },
    {
      "epoch": 1.9969183359013867,
      "grad_norm": 2.850844144821167,
      "learning_rate": 1.886181956604492e-06,
      "loss": 1.3692,
      "step": 1944
    },
    {
      "epoch": 1.9979455572675913,
      "grad_norm": 2.8984055519104004,
      "learning_rate": 1.8842786448420252e-06,
      "loss": 1.3681,
      "step": 1945
    },
    {
      "epoch": 1.9989727786337956,
      "grad_norm": 2.8746683597564697,
      "learning_rate": 1.8823753330795586e-06,
      "loss": 1.4047,
      "step": 1946
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.7244315147399902,
      "learning_rate": 1.8804720213170917e-06,
      "loss": 1.3003,
      "step": 1947
    },
    {
      "epoch": 2.0010272213662046,
      "grad_norm": 3.10217022895813,
      "learning_rate": 1.878568709554625e-06,
      "loss": 1.3375,
      "step": 1948
    },
    {
      "epoch": 2.0020544427324087,
      "grad_norm": 3.1777870655059814,
      "learning_rate": 1.8766653977921585e-06,
      "loss": 1.3775,
      "step": 1949
    },
    {
      "epoch": 2.0030816640986133,
      "grad_norm": 3.008085250854492,
      "learning_rate": 1.874762086029692e-06,
      "loss": 1.3435,
      "step": 1950
    },
    {
      "epoch": 2.0030816640986133,
      "eval_loss": 1.7830486297607422,
      "eval_runtime": 37.0192,
      "eval_samples_per_second": 27.013,
      "eval_steps_per_second": 1.135,
      "step": 1950
    },
    {
      "epoch": 2.0041088854648175,
      "grad_norm": 3.228102445602417,
      "learning_rate": 1.8728587742672252e-06,
      "loss": 1.3697,
      "step": 1951
    },
    {
      "epoch": 2.005136106831022,
      "grad_norm": 3.392749071121216,
      "learning_rate": 1.8709554625047585e-06,
      "loss": 1.3813,
      "step": 1952
    },
    {
      "epoch": 2.0061633281972266,
      "grad_norm": 3.264998197555542,
      "learning_rate": 1.8690521507422917e-06,
      "loss": 1.3203,
      "step": 1953
    },
    {
      "epoch": 2.007190549563431,
      "grad_norm": 3.344672679901123,
      "learning_rate": 1.867148838979825e-06,
      "loss": 1.3413,
      "step": 1954
    },
    {
      "epoch": 2.0082177709296354,
      "grad_norm": 3.138148307800293,
      "learning_rate": 1.8652455272173584e-06,
      "loss": 1.3606,
      "step": 1955
    },
    {
      "epoch": 2.00924499229584,
      "grad_norm": 3.1569597721099854,
      "learning_rate": 1.8633422154548916e-06,
      "loss": 1.3846,
      "step": 1956
    },
    {
      "epoch": 2.010272213662044,
      "grad_norm": 3.0843050479888916,
      "learning_rate": 1.861438903692425e-06,
      "loss": 1.282,
      "step": 1957
    },
    {
      "epoch": 2.0112994350282487,
      "grad_norm": 3.3988428115844727,
      "learning_rate": 1.859535591929958e-06,
      "loss": 1.3183,
      "step": 1958
    },
    {
      "epoch": 2.012326656394453,
      "grad_norm": 3.3206543922424316,
      "learning_rate": 1.8576322801674915e-06,
      "loss": 1.3874,
      "step": 1959
    },
    {
      "epoch": 2.0133538777606574,
      "grad_norm": 3.1254804134368896,
      "learning_rate": 1.855728968405025e-06,
      "loss": 1.3335,
      "step": 1960
    },
    {
      "epoch": 2.0133538777606574,
      "eval_loss": 1.780193567276001,
      "eval_runtime": 37.0457,
      "eval_samples_per_second": 26.994,
      "eval_steps_per_second": 1.134,
      "step": 1960
    },
    {
      "epoch": 2.014381099126862,
      "grad_norm": 3.2736592292785645,
      "learning_rate": 1.8538256566425584e-06,
      "loss": 1.3489,
      "step": 1961
    },
    {
      "epoch": 2.015408320493066,
      "grad_norm": 3.1620168685913086,
      "learning_rate": 1.8519223448800915e-06,
      "loss": 1.3898,
      "step": 1962
    },
    {
      "epoch": 2.0164355418592708,
      "grad_norm": 3.3050506114959717,
      "learning_rate": 1.850019033117625e-06,
      "loss": 1.3292,
      "step": 1963
    },
    {
      "epoch": 2.017462763225475,
      "grad_norm": 3.192668914794922,
      "learning_rate": 1.848115721355158e-06,
      "loss": 1.3319,
      "step": 1964
    },
    {
      "epoch": 2.0184899845916795,
      "grad_norm": 3.1518166065216064,
      "learning_rate": 1.8462124095926914e-06,
      "loss": 1.3433,
      "step": 1965
    },
    {
      "epoch": 2.019517205957884,
      "grad_norm": 3.365323305130005,
      "learning_rate": 1.8443090978302248e-06,
      "loss": 1.362,
      "step": 1966
    },
    {
      "epoch": 2.0205444273240882,
      "grad_norm": 3.4271507263183594,
      "learning_rate": 1.842405786067758e-06,
      "loss": 1.3867,
      "step": 1967
    },
    {
      "epoch": 2.021571648690293,
      "grad_norm": 3.167879581451416,
      "learning_rate": 1.8405024743052913e-06,
      "loss": 1.3341,
      "step": 1968
    },
    {
      "epoch": 2.022598870056497,
      "grad_norm": 3.1220595836639404,
      "learning_rate": 1.8385991625428247e-06,
      "loss": 1.379,
      "step": 1969
    },
    {
      "epoch": 2.0236260914227016,
      "grad_norm": 3.03648042678833,
      "learning_rate": 1.8366958507803578e-06,
      "loss": 1.3095,
      "step": 1970
    },
    {
      "epoch": 2.0236260914227016,
      "eval_loss": 1.7782403230667114,
      "eval_runtime": 37.0535,
      "eval_samples_per_second": 26.988,
      "eval_steps_per_second": 1.133,
      "step": 1970
    },
    {
      "epoch": 2.024653312788906,
      "grad_norm": 3.254167318344116,
      "learning_rate": 1.8347925390178914e-06,
      "loss": 1.3625,
      "step": 1971
    },
    {
      "epoch": 2.0256805341551103,
      "grad_norm": 3.0616867542266846,
      "learning_rate": 1.8328892272554247e-06,
      "loss": 1.3059,
      "step": 1972
    },
    {
      "epoch": 2.026707755521315,
      "grad_norm": 3.236483097076416,
      "learning_rate": 1.8309859154929579e-06,
      "loss": 1.3737,
      "step": 1973
    },
    {
      "epoch": 2.0277349768875195,
      "grad_norm": 3.1941466331481934,
      "learning_rate": 1.8290826037304913e-06,
      "loss": 1.35,
      "step": 1974
    },
    {
      "epoch": 2.0287621982537236,
      "grad_norm": 3.0594732761383057,
      "learning_rate": 1.8271792919680246e-06,
      "loss": 1.3261,
      "step": 1975
    },
    {
      "epoch": 2.029789419619928,
      "grad_norm": 3.157635450363159,
      "learning_rate": 1.8252759802055578e-06,
      "loss": 1.3224,
      "step": 1976
    },
    {
      "epoch": 2.0308166409861323,
      "grad_norm": 3.380322217941284,
      "learning_rate": 1.8233726684430911e-06,
      "loss": 1.3742,
      "step": 1977
    },
    {
      "epoch": 2.031843862352337,
      "grad_norm": 3.397022247314453,
      "learning_rate": 1.8214693566806243e-06,
      "loss": 1.324,
      "step": 1978
    },
    {
      "epoch": 2.0328710837185415,
      "grad_norm": 3.187364101409912,
      "learning_rate": 1.8195660449181576e-06,
      "loss": 1.3435,
      "step": 1979
    },
    {
      "epoch": 2.0338983050847457,
      "grad_norm": 3.1330549716949463,
      "learning_rate": 1.817662733155691e-06,
      "loss": 1.3472,
      "step": 1980
    },
    {
      "epoch": 2.0338983050847457,
      "eval_loss": 1.7825566530227661,
      "eval_runtime": 37.0315,
      "eval_samples_per_second": 27.004,
      "eval_steps_per_second": 1.134,
      "step": 1980
    },
    {
      "epoch": 2.0349255264509503,
      "grad_norm": 3.1697194576263428,
      "learning_rate": 1.8157594213932244e-06,
      "loss": 1.347,
      "step": 1981
    },
    {
      "epoch": 2.0359527478171544,
      "grad_norm": 3.1682229042053223,
      "learning_rate": 1.8138561096307577e-06,
      "loss": 1.3588,
      "step": 1982
    },
    {
      "epoch": 2.036979969183359,
      "grad_norm": 3.1456754207611084,
      "learning_rate": 1.811952797868291e-06,
      "loss": 1.2981,
      "step": 1983
    },
    {
      "epoch": 2.0380071905495636,
      "grad_norm": 3.1854794025421143,
      "learning_rate": 1.8100494861058242e-06,
      "loss": 1.3429,
      "step": 1984
    },
    {
      "epoch": 2.0390344119157677,
      "grad_norm": 3.2500100135803223,
      "learning_rate": 1.8081461743433576e-06,
      "loss": 1.3572,
      "step": 1985
    },
    {
      "epoch": 2.0400616332819723,
      "grad_norm": 3.0852582454681396,
      "learning_rate": 1.806242862580891e-06,
      "loss": 1.3081,
      "step": 1986
    },
    {
      "epoch": 2.0410888546481765,
      "grad_norm": 3.3680732250213623,
      "learning_rate": 1.8043395508184241e-06,
      "loss": 1.3584,
      "step": 1987
    },
    {
      "epoch": 2.042116076014381,
      "grad_norm": 3.1346354484558105,
      "learning_rate": 1.8024362390559575e-06,
      "loss": 1.3437,
      "step": 1988
    },
    {
      "epoch": 2.0431432973805856,
      "grad_norm": 3.1831793785095215,
      "learning_rate": 1.8005329272934906e-06,
      "loss": 1.3206,
      "step": 1989
    },
    {
      "epoch": 2.04417051874679,
      "grad_norm": 3.3129804134368896,
      "learning_rate": 1.798629615531024e-06,
      "loss": 1.3992,
      "step": 1990
    },
    {
      "epoch": 2.04417051874679,
      "eval_loss": 1.791827917098999,
      "eval_runtime": 37.1302,
      "eval_samples_per_second": 26.932,
      "eval_steps_per_second": 1.131,
      "step": 1990
    },
    {
      "epoch": 2.0451977401129944,
      "grad_norm": 3.2966983318328857,
      "learning_rate": 1.7967263037685576e-06,
      "loss": 1.3374,
      "step": 1991
    },
    {
      "epoch": 2.046224961479199,
      "grad_norm": 3.2134904861450195,
      "learning_rate": 1.794822992006091e-06,
      "loss": 1.3304,
      "step": 1992
    },
    {
      "epoch": 2.047252182845403,
      "grad_norm": 3.3718373775482178,
      "learning_rate": 1.792919680243624e-06,
      "loss": 1.3898,
      "step": 1993
    },
    {
      "epoch": 2.0482794042116077,
      "grad_norm": 3.1467645168304443,
      "learning_rate": 1.7910163684811574e-06,
      "loss": 1.3089,
      "step": 1994
    },
    {
      "epoch": 2.049306625577812,
      "grad_norm": 3.1504383087158203,
      "learning_rate": 1.7891130567186906e-06,
      "loss": 1.3398,
      "step": 1995
    },
    {
      "epoch": 2.0503338469440164,
      "grad_norm": 3.2479350566864014,
      "learning_rate": 1.787209744956224e-06,
      "loss": 1.3099,
      "step": 1996
    },
    {
      "epoch": 2.051361068310221,
      "grad_norm": 3.311795234680176,
      "learning_rate": 1.7853064331937573e-06,
      "loss": 1.3369,
      "step": 1997
    },
    {
      "epoch": 2.052388289676425,
      "grad_norm": 3.1674633026123047,
      "learning_rate": 1.7834031214312905e-06,
      "loss": 1.3732,
      "step": 1998
    },
    {
      "epoch": 2.0534155110426298,
      "grad_norm": 3.1175575256347656,
      "learning_rate": 1.7814998096688238e-06,
      "loss": 1.3159,
      "step": 1999
    },
    {
      "epoch": 2.054442732408834,
      "grad_norm": 3.216402530670166,
      "learning_rate": 1.779596497906357e-06,
      "loss": 1.3479,
      "step": 2000
    },
    {
      "epoch": 2.054442732408834,
      "eval_loss": 1.7845674753189087,
      "eval_runtime": 37.2755,
      "eval_samples_per_second": 26.827,
      "eval_steps_per_second": 1.127,
      "step": 2000
    }
  ],
  "logging_steps": 1,
  "max_steps": 2919,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 6.23461881872384e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}