{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.027221366204417,
  "eval_steps": 10,
  "global_step": 1000,
  "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": 2.7873,
      "step": 1
    },
    {
      "epoch": 0.002054442732408834,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7904,
      "step": 2
    },
    {
      "epoch": 0.0030816640986132513,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8348,
      "step": 3
    },
    {
      "epoch": 0.004108885464817668,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8357,
      "step": 4
    },
    {
      "epoch": 0.005136106831022085,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8153,
      "step": 5
    },
    {
      "epoch": 0.0061633281972265025,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.825,
      "step": 6
    },
    {
      "epoch": 0.007190549563430919,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8002,
      "step": 7
    },
    {
      "epoch": 0.008217770929635337,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8413,
      "step": 8
    },
    {
      "epoch": 0.009244992295839754,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8157,
      "step": 9
    },
    {
      "epoch": 0.01027221366204417,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8249,
      "step": 10
    },
    {
      "epoch": 0.01027221366204417,
      "eval_loss": 2.8518412113189697,
      "eval_runtime": 43.212,
      "eval_samples_per_second": 23.142,
      "eval_steps_per_second": 0.972,
      "step": 10
    },
    {
      "epoch": 0.011299435028248588,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7829,
      "step": 11
    },
    {
      "epoch": 0.012326656394453005,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7819,
      "step": 12
    },
    {
      "epoch": 0.01335387776065742,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8355,
      "step": 13
    },
    {
      "epoch": 0.014381099126861838,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8138,
      "step": 14
    },
    {
      "epoch": 0.015408320493066256,
      "grad_norm": 5.162191390991211,
      "learning_rate": 1.7123287671232876e-08,
      "loss": 2.8088,
      "step": 15
    },
    {
      "epoch": 0.016435541859270673,
      "grad_norm": 5.5954084396362305,
      "learning_rate": 3.424657534246575e-08,
      "loss": 2.7883,
      "step": 16
    },
    {
      "epoch": 0.01746276322547509,
      "grad_norm": 5.542337417602539,
      "learning_rate": 5.136986301369863e-08,
      "loss": 2.7625,
      "step": 17
    },
    {
      "epoch": 0.01848998459167951,
      "grad_norm": 5.300518989562988,
      "learning_rate": 6.84931506849315e-08,
      "loss": 2.84,
      "step": 18
    },
    {
      "epoch": 0.019517205957883924,
      "grad_norm": 5.6646881103515625,
      "learning_rate": 8.561643835616439e-08,
      "loss": 2.761,
      "step": 19
    },
    {
      "epoch": 0.02054442732408834,
      "grad_norm": 5.375557899475098,
      "learning_rate": 1.0273972602739726e-07,
      "loss": 2.8215,
      "step": 20
    },
    {
      "epoch": 0.02054442732408834,
      "eval_loss": 2.8507144451141357,
      "eval_runtime": 45.1133,
      "eval_samples_per_second": 22.166,
      "eval_steps_per_second": 0.931,
      "step": 20
    },
    {
      "epoch": 0.02157164869029276,
      "grad_norm": 5.1731977462768555,
      "learning_rate": 1.1986301369863014e-07,
      "loss": 2.8348,
      "step": 21
    },
    {
      "epoch": 0.022598870056497175,
      "grad_norm": 5.370945930480957,
      "learning_rate": 1.36986301369863e-07,
      "loss": 2.822,
      "step": 22
    },
    {
      "epoch": 0.02362609142270159,
      "grad_norm": 5.550389289855957,
      "learning_rate": 1.541095890410959e-07,
      "loss": 2.7904,
      "step": 23
    },
    {
      "epoch": 0.02465331278890601,
      "grad_norm": 5.339347839355469,
      "learning_rate": 1.7123287671232878e-07,
      "loss": 2.8066,
      "step": 24
    },
    {
      "epoch": 0.025680534155110426,
      "grad_norm": 5.2319722175598145,
      "learning_rate": 1.8835616438356165e-07,
      "loss": 2.7976,
      "step": 25
    },
    {
      "epoch": 0.02670775552131484,
      "grad_norm": 5.207149982452393,
      "learning_rate": 2.0547945205479452e-07,
      "loss": 2.8162,
      "step": 26
    },
    {
      "epoch": 0.02773497688751926,
      "grad_norm": 4.400204181671143,
      "learning_rate": 2.226027397260274e-07,
      "loss": 2.7808,
      "step": 27
    },
    {
      "epoch": 0.028762198253723677,
      "grad_norm": 4.21675968170166,
      "learning_rate": 2.397260273972603e-07,
      "loss": 2.8044,
      "step": 28
    },
    {
      "epoch": 0.029789419619928096,
      "grad_norm": 4.406878471374512,
      "learning_rate": 2.568493150684932e-07,
      "loss": 2.8162,
      "step": 29
    },
    {
      "epoch": 0.030816640986132512,
      "grad_norm": 4.024835586547852,
      "learning_rate": 2.73972602739726e-07,
      "loss": 2.7994,
      "step": 30
    },
    {
      "epoch": 0.030816640986132512,
      "eval_loss": 2.8124029636383057,
      "eval_runtime": 43.1772,
      "eval_samples_per_second": 23.16,
      "eval_steps_per_second": 0.973,
      "step": 30
    },
    {
      "epoch": 0.03184386235233693,
      "grad_norm": 3.5378901958465576,
      "learning_rate": 2.910958904109589e-07,
      "loss": 2.7637,
      "step": 31
    },
    {
      "epoch": 0.03287108371854135,
      "grad_norm": 3.6289916038513184,
      "learning_rate": 3.082191780821918e-07,
      "loss": 2.8017,
      "step": 32
    },
    {
      "epoch": 0.03389830508474576,
      "grad_norm": 3.6789698600769043,
      "learning_rate": 3.2534246575342466e-07,
      "loss": 2.7723,
      "step": 33
    },
    {
      "epoch": 0.03492552645095018,
      "grad_norm": 3.491389751434326,
      "learning_rate": 3.4246575342465755e-07,
      "loss": 2.7528,
      "step": 34
    },
    {
      "epoch": 0.035952747817154594,
      "grad_norm": 3.383080005645752,
      "learning_rate": 3.595890410958904e-07,
      "loss": 2.7418,
      "step": 35
    },
    {
      "epoch": 0.03697996918335902,
      "grad_norm": 3.4823594093322754,
      "learning_rate": 3.767123287671233e-07,
      "loss": 2.6662,
      "step": 36
    },
    {
      "epoch": 0.03800719054956343,
      "grad_norm": 3.2146108150482178,
      "learning_rate": 3.938356164383562e-07,
      "loss": 2.7659,
      "step": 37
    },
    {
      "epoch": 0.03903441191576785,
      "grad_norm": 2.9549412727355957,
      "learning_rate": 4.1095890410958903e-07,
      "loss": 2.6838,
      "step": 38
    },
    {
      "epoch": 0.040061633281972264,
      "grad_norm": 3.0744080543518066,
      "learning_rate": 4.2808219178082193e-07,
      "loss": 2.6879,
      "step": 39
    },
    {
      "epoch": 0.04108885464817668,
      "grad_norm": 3.19840669631958,
      "learning_rate": 4.452054794520548e-07,
      "loss": 2.7443,
      "step": 40
    },
    {
      "epoch": 0.04108885464817668,
      "eval_loss": 2.7469444274902344,
      "eval_runtime": 46.4676,
      "eval_samples_per_second": 21.52,
      "eval_steps_per_second": 0.904,
      "step": 40
    },
    {
      "epoch": 0.042116076014381096,
      "grad_norm": 3.178406000137329,
      "learning_rate": 4.6232876712328767e-07,
      "loss": 2.7036,
      "step": 41
    },
    {
      "epoch": 0.04314329738058552,
      "grad_norm": 3.2032673358917236,
      "learning_rate": 4.794520547945206e-07,
      "loss": 2.6465,
      "step": 42
    },
    {
      "epoch": 0.044170518746789934,
      "grad_norm": 3.1525962352752686,
      "learning_rate": 4.965753424657534e-07,
      "loss": 2.6996,
      "step": 43
    },
    {
      "epoch": 0.04519774011299435,
      "grad_norm": 2.907524585723877,
      "learning_rate": 5.136986301369864e-07,
      "loss": 2.6581,
      "step": 44
    },
    {
      "epoch": 0.046224961479198766,
      "grad_norm": 2.5885250568389893,
      "learning_rate": 5.308219178082192e-07,
      "loss": 2.6773,
      "step": 45
    },
    {
      "epoch": 0.04725218284540318,
      "grad_norm": 2.569068431854248,
      "learning_rate": 5.47945205479452e-07,
      "loss": 2.6635,
      "step": 46
    },
    {
      "epoch": 0.048279404211607604,
      "grad_norm": 2.5313518047332764,
      "learning_rate": 5.65068493150685e-07,
      "loss": 2.706,
      "step": 47
    },
    {
      "epoch": 0.04930662557781202,
      "grad_norm": 2.615442991256714,
      "learning_rate": 5.821917808219178e-07,
      "loss": 2.638,
      "step": 48
    },
    {
      "epoch": 0.050333846944016436,
      "grad_norm": 2.7420806884765625,
      "learning_rate": 5.993150684931507e-07,
      "loss": 2.6398,
      "step": 49
    },
    {
      "epoch": 0.05136106831022085,
      "grad_norm": 2.5246975421905518,
      "learning_rate": 6.164383561643836e-07,
      "loss": 2.5916,
      "step": 50
    },
    {
      "epoch": 0.05136106831022085,
      "eval_loss": 2.6723692417144775,
      "eval_runtime": 46.7206,
      "eval_samples_per_second": 21.404,
      "eval_steps_per_second": 0.899,
      "step": 50
    },
    {
      "epoch": 0.05238828967642527,
      "grad_norm": 2.390143632888794,
      "learning_rate": 6.335616438356165e-07,
      "loss": 2.5773,
      "step": 51
    },
    {
      "epoch": 0.05341551104262968,
      "grad_norm": 2.0792484283447266,
      "learning_rate": 6.506849315068493e-07,
      "loss": 2.6249,
      "step": 52
    },
    {
      "epoch": 0.054442732408834106,
      "grad_norm": 1.9995019435882568,
      "learning_rate": 6.678082191780823e-07,
      "loss": 2.5425,
      "step": 53
    },
    {
      "epoch": 0.05546995377503852,
      "grad_norm": 1.985452651977539,
      "learning_rate": 6.849315068493151e-07,
      "loss": 2.5986,
      "step": 54
    },
    {
      "epoch": 0.05649717514124294,
      "grad_norm": 2.10184907913208,
      "learning_rate": 7.020547945205481e-07,
      "loss": 2.5713,
      "step": 55
    },
    {
      "epoch": 0.05752439650744735,
      "grad_norm": 2.0898985862731934,
      "learning_rate": 7.191780821917808e-07,
      "loss": 2.535,
      "step": 56
    },
    {
      "epoch": 0.05855161787365177,
      "grad_norm": 2.0239579677581787,
      "learning_rate": 7.363013698630137e-07,
      "loss": 2.5667,
      "step": 57
    },
    {
      "epoch": 0.05957883923985619,
      "grad_norm": 2.207080125808716,
      "learning_rate": 7.534246575342466e-07,
      "loss": 2.4999,
      "step": 58
    },
    {
      "epoch": 0.06060606060606061,
      "grad_norm": 1.8541918992996216,
      "learning_rate": 7.705479452054795e-07,
      "loss": 2.5582,
      "step": 59
    },
    {
      "epoch": 0.061633281972265024,
      "grad_norm": 1.8530476093292236,
      "learning_rate": 7.876712328767124e-07,
      "loss": 2.5535,
      "step": 60
    },
    {
      "epoch": 0.061633281972265024,
      "eval_loss": 2.607963800430298,
      "eval_runtime": 45.9754,
      "eval_samples_per_second": 21.751,
      "eval_steps_per_second": 0.914,
      "step": 60
    },
    {
      "epoch": 0.06266050333846944,
      "grad_norm": 1.7978090047836304,
      "learning_rate": 8.047945205479453e-07,
      "loss": 2.53,
      "step": 61
    },
    {
      "epoch": 0.06368772470467386,
      "grad_norm": 1.8173762559890747,
      "learning_rate": 8.219178082191781e-07,
      "loss": 2.5144,
      "step": 62
    },
    {
      "epoch": 0.06471494607087827,
      "grad_norm": 2.0847485065460205,
      "learning_rate": 8.39041095890411e-07,
      "loss": 2.5525,
      "step": 63
    },
    {
      "epoch": 0.0657421674370827,
      "grad_norm": 1.8691741228103638,
      "learning_rate": 8.561643835616439e-07,
      "loss": 2.5451,
      "step": 64
    },
    {
      "epoch": 0.0667693888032871,
      "grad_norm": 2.0242726802825928,
      "learning_rate": 8.732876712328768e-07,
      "loss": 2.4145,
      "step": 65
    },
    {
      "epoch": 0.06779661016949153,
      "grad_norm": 1.8268908262252808,
      "learning_rate": 8.904109589041097e-07,
      "loss": 2.5024,
      "step": 66
    },
    {
      "epoch": 0.06882383153569595,
      "grad_norm": 1.8703057765960693,
      "learning_rate": 9.075342465753426e-07,
      "loss": 2.461,
      "step": 67
    },
    {
      "epoch": 0.06985105290190036,
      "grad_norm": 1.8747144937515259,
      "learning_rate": 9.246575342465753e-07,
      "loss": 2.5122,
      "step": 68
    },
    {
      "epoch": 0.07087827426810478,
      "grad_norm": 1.744924783706665,
      "learning_rate": 9.417808219178083e-07,
      "loss": 2.4673,
      "step": 69
    },
    {
      "epoch": 0.07190549563430919,
      "grad_norm": 1.7492460012435913,
      "learning_rate": 9.589041095890411e-07,
      "loss": 2.5598,
      "step": 70
    },
    {
      "epoch": 0.07190549563430919,
      "eval_loss": 2.547417640686035,
      "eval_runtime": 49.7488,
      "eval_samples_per_second": 20.101,
      "eval_steps_per_second": 0.844,
      "step": 70
    },
    {
      "epoch": 0.07293271700051361,
      "grad_norm": 1.9218230247497559,
      "learning_rate": 9.76027397260274e-07,
      "loss": 2.4562,
      "step": 71
    },
    {
      "epoch": 0.07395993836671803,
      "grad_norm": 2.0127363204956055,
      "learning_rate": 9.931506849315068e-07,
      "loss": 2.4141,
      "step": 72
    },
    {
      "epoch": 0.07498715973292244,
      "grad_norm": 1.8700369596481323,
      "learning_rate": 1.0102739726027399e-06,
      "loss": 2.4696,
      "step": 73
    },
    {
      "epoch": 0.07601438109912687,
      "grad_norm": 2.1706016063690186,
      "learning_rate": 1.0273972602739727e-06,
      "loss": 2.4404,
      "step": 74
    },
    {
      "epoch": 0.07704160246533127,
      "grad_norm": 2.470412254333496,
      "learning_rate": 1.0445205479452056e-06,
      "loss": 2.4409,
      "step": 75
    },
    {
      "epoch": 0.0780688238315357,
      "grad_norm": 1.906569242477417,
      "learning_rate": 1.0616438356164384e-06,
      "loss": 2.4313,
      "step": 76
    },
    {
      "epoch": 0.07909604519774012,
      "grad_norm": 2.122936487197876,
      "learning_rate": 1.0787671232876712e-06,
      "loss": 2.4369,
      "step": 77
    },
    {
      "epoch": 0.08012326656394453,
      "grad_norm": 2.079127311706543,
      "learning_rate": 1.095890410958904e-06,
      "loss": 2.449,
      "step": 78
    },
    {
      "epoch": 0.08115048793014895,
      "grad_norm": 1.9173146486282349,
      "learning_rate": 1.1130136986301371e-06,
      "loss": 2.3782,
      "step": 79
    },
    {
      "epoch": 0.08217770929635336,
      "grad_norm": 1.9856903553009033,
      "learning_rate": 1.13013698630137e-06,
      "loss": 2.4215,
      "step": 80
    },
    {
      "epoch": 0.08217770929635336,
      "eval_loss": 2.479673385620117,
      "eval_runtime": 46.2211,
      "eval_samples_per_second": 21.635,
      "eval_steps_per_second": 0.909,
      "step": 80
    },
    {
      "epoch": 0.08320493066255778,
      "grad_norm": 2.3764865398406982,
      "learning_rate": 1.1472602739726028e-06,
      "loss": 2.3744,
      "step": 81
    },
    {
      "epoch": 0.08423215202876219,
      "grad_norm": 1.980668544769287,
      "learning_rate": 1.1643835616438357e-06,
      "loss": 2.3856,
      "step": 82
    },
    {
      "epoch": 0.08525937339496661,
      "grad_norm": 2.2589704990386963,
      "learning_rate": 1.1815068493150685e-06,
      "loss": 2.3853,
      "step": 83
    },
    {
      "epoch": 0.08628659476117104,
      "grad_norm": 2.171780824661255,
      "learning_rate": 1.1986301369863014e-06,
      "loss": 2.3603,
      "step": 84
    },
    {
      "epoch": 0.08731381612737545,
      "grad_norm": 2.25736403465271,
      "learning_rate": 1.2157534246575344e-06,
      "loss": 2.3189,
      "step": 85
    },
    {
      "epoch": 0.08834103749357987,
      "grad_norm": 2.3643763065338135,
      "learning_rate": 1.2328767123287673e-06,
      "loss": 2.3795,
      "step": 86
    },
    {
      "epoch": 0.08936825885978428,
      "grad_norm": 2.384516954421997,
      "learning_rate": 1.25e-06,
      "loss": 2.3257,
      "step": 87
    },
    {
      "epoch": 0.0903954802259887,
      "grad_norm": 2.5072567462921143,
      "learning_rate": 1.267123287671233e-06,
      "loss": 2.3125,
      "step": 88
    },
    {
      "epoch": 0.09142270159219312,
      "grad_norm": 2.2726807594299316,
      "learning_rate": 1.284246575342466e-06,
      "loss": 2.3297,
      "step": 89
    },
    {
      "epoch": 0.09244992295839753,
      "grad_norm": 2.4817421436309814,
      "learning_rate": 1.3013698630136986e-06,
      "loss": 2.2878,
      "step": 90
    },
    {
      "epoch": 0.09244992295839753,
      "eval_loss": 2.4065146446228027,
      "eval_runtime": 47.9372,
      "eval_samples_per_second": 20.861,
      "eval_steps_per_second": 0.876,
      "step": 90
    },
    {
      "epoch": 0.09347714432460195,
      "grad_norm": 2.2219603061676025,
      "learning_rate": 1.3184931506849317e-06,
      "loss": 2.2978,
      "step": 91
    },
    {
      "epoch": 0.09450436569080636,
      "grad_norm": 2.8790745735168457,
      "learning_rate": 1.3356164383561645e-06,
      "loss": 2.313,
      "step": 92
    },
    {
      "epoch": 0.09553158705701079,
      "grad_norm": 2.589200496673584,
      "learning_rate": 1.3527397260273976e-06,
      "loss": 2.2937,
      "step": 93
    },
    {
      "epoch": 0.09655880842321521,
      "grad_norm": 2.760016679763794,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 2.277,
      "step": 94
    },
    {
      "epoch": 0.09758602978941962,
      "grad_norm": 2.2641494274139404,
      "learning_rate": 1.386986301369863e-06,
      "loss": 2.3336,
      "step": 95
    },
    {
      "epoch": 0.09861325115562404,
      "grad_norm": 2.2786386013031006,
      "learning_rate": 1.4041095890410961e-06,
      "loss": 2.233,
      "step": 96
    },
    {
      "epoch": 0.09964047252182845,
      "grad_norm": 2.5590052604675293,
      "learning_rate": 1.421232876712329e-06,
      "loss": 2.2851,
      "step": 97
    },
    {
      "epoch": 0.10066769388803287,
      "grad_norm": 2.5609471797943115,
      "learning_rate": 1.4383561643835616e-06,
      "loss": 2.1848,
      "step": 98
    },
    {
      "epoch": 0.1016949152542373,
      "grad_norm": 2.442521810531616,
      "learning_rate": 1.4554794520547946e-06,
      "loss": 2.2073,
      "step": 99
    },
    {
      "epoch": 0.1027221366204417,
      "grad_norm": 2.3735859394073486,
      "learning_rate": 1.4726027397260275e-06,
      "loss": 2.2353,
      "step": 100
    },
    {
      "epoch": 0.1027221366204417,
      "eval_loss": 2.3457911014556885,
      "eval_runtime": 51.6469,
      "eval_samples_per_second": 19.362,
      "eval_steps_per_second": 0.813,
      "step": 100
    },
    {
      "epoch": 0.10374935798664613,
      "grad_norm": 2.175354242324829,
      "learning_rate": 1.4897260273972605e-06,
      "loss": 2.2259,
      "step": 101
    },
    {
      "epoch": 0.10477657935285054,
      "grad_norm": 2.217677116394043,
      "learning_rate": 1.5068493150684932e-06,
      "loss": 2.2082,
      "step": 102
    },
    {
      "epoch": 0.10580380071905496,
      "grad_norm": 2.4618067741394043,
      "learning_rate": 1.5239726027397262e-06,
      "loss": 2.1649,
      "step": 103
    },
    {
      "epoch": 0.10683102208525937,
      "grad_norm": 2.232060432434082,
      "learning_rate": 1.541095890410959e-06,
      "loss": 2.1962,
      "step": 104
    },
    {
      "epoch": 0.10785824345146379,
      "grad_norm": 2.9140872955322266,
      "learning_rate": 1.5582191780821921e-06,
      "loss": 2.2446,
      "step": 105
    },
    {
      "epoch": 0.10888546481766821,
      "grad_norm": 2.232158660888672,
      "learning_rate": 1.5753424657534248e-06,
      "loss": 2.2179,
      "step": 106
    },
    {
      "epoch": 0.10991268618387262,
      "grad_norm": 2.804938793182373,
      "learning_rate": 1.5924657534246576e-06,
      "loss": 2.2449,
      "step": 107
    },
    {
      "epoch": 0.11093990755007704,
      "grad_norm": 2.3690710067749023,
      "learning_rate": 1.6095890410958907e-06,
      "loss": 2.225,
      "step": 108
    },
    {
      "epoch": 0.11196712891628145,
      "grad_norm": 2.7717416286468506,
      "learning_rate": 1.6267123287671235e-06,
      "loss": 2.0922,
      "step": 109
    },
    {
      "epoch": 0.11299435028248588,
      "grad_norm": 3.053602933883667,
      "learning_rate": 1.6438356164383561e-06,
      "loss": 2.1702,
      "step": 110
    },
    {
      "epoch": 0.11299435028248588,
      "eval_loss": 2.298161029815674,
      "eval_runtime": 44.0776,
      "eval_samples_per_second": 22.687,
      "eval_steps_per_second": 0.953,
      "step": 110
    },
    {
      "epoch": 0.1140215716486903,
      "grad_norm": 2.8673532009124756,
      "learning_rate": 1.6609589041095892e-06,
      "loss": 2.0938,
      "step": 111
    },
    {
      "epoch": 0.1150487930148947,
      "grad_norm": 2.57185959815979,
      "learning_rate": 1.678082191780822e-06,
      "loss": 2.1892,
      "step": 112
    },
    {
      "epoch": 0.11607601438109913,
      "grad_norm": 2.5278828144073486,
      "learning_rate": 1.695205479452055e-06,
      "loss": 2.1251,
      "step": 113
    },
    {
      "epoch": 0.11710323574730354,
      "grad_norm": 2.8581748008728027,
      "learning_rate": 1.7123287671232877e-06,
      "loss": 2.1699,
      "step": 114
    },
    {
      "epoch": 0.11813045711350796,
      "grad_norm": 2.924959897994995,
      "learning_rate": 1.7294520547945206e-06,
      "loss": 2.1005,
      "step": 115
    },
    {
      "epoch": 0.11915767847971238,
      "grad_norm": 2.477034091949463,
      "learning_rate": 1.7465753424657536e-06,
      "loss": 2.2016,
      "step": 116
    },
    {
      "epoch": 0.12018489984591679,
      "grad_norm": 2.8719305992126465,
      "learning_rate": 1.7636986301369865e-06,
      "loss": 2.1588,
      "step": 117
    },
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 2.4403953552246094,
      "learning_rate": 1.7808219178082193e-06,
      "loss": 2.1445,
      "step": 118
    },
    {
      "epoch": 0.12223934257832562,
      "grad_norm": 2.7994632720947266,
      "learning_rate": 1.7979452054794521e-06,
      "loss": 2.1369,
      "step": 119
    },
    {
      "epoch": 0.12326656394453005,
      "grad_norm": 2.163196563720703,
      "learning_rate": 1.8150684931506852e-06,
      "loss": 2.2195,
      "step": 120
    },
    {
      "epoch": 0.12326656394453005,
      "eval_loss": 2.2648775577545166,
      "eval_runtime": 49.47,
      "eval_samples_per_second": 20.214,
      "eval_steps_per_second": 0.849,
      "step": 120
    },
    {
      "epoch": 0.12429378531073447,
      "grad_norm": 2.6743483543395996,
      "learning_rate": 1.832191780821918e-06,
      "loss": 2.1521,
      "step": 121
    },
    {
      "epoch": 0.12532100667693888,
      "grad_norm": 2.54795503616333,
      "learning_rate": 1.8493150684931507e-06,
      "loss": 2.1158,
      "step": 122
    },
    {
      "epoch": 0.1263482280431433,
      "grad_norm": 2.9093267917633057,
      "learning_rate": 1.8664383561643837e-06,
      "loss": 2.0707,
      "step": 123
    },
    {
      "epoch": 0.12737544940934772,
      "grad_norm": 2.416940212249756,
      "learning_rate": 1.8835616438356166e-06,
      "loss": 2.0901,
      "step": 124
    },
    {
      "epoch": 0.12840267077555212,
      "grad_norm": 2.8333663940429688,
      "learning_rate": 1.9006849315068496e-06,
      "loss": 2.0545,
      "step": 125
    },
    {
      "epoch": 0.12942989214175654,
      "grad_norm": 2.483051061630249,
      "learning_rate": 1.9178082191780823e-06,
      "loss": 2.1274,
      "step": 126
    },
    {
      "epoch": 0.13045711350796096,
      "grad_norm": 2.3486859798431396,
      "learning_rate": 1.9349315068493153e-06,
      "loss": 2.0478,
      "step": 127
    },
    {
      "epoch": 0.1314843348741654,
      "grad_norm": 3.1512887477874756,
      "learning_rate": 1.952054794520548e-06,
      "loss": 2.1041,
      "step": 128
    },
    {
      "epoch": 0.1325115562403698,
      "grad_norm": 2.902873992919922,
      "learning_rate": 1.969178082191781e-06,
      "loss": 2.1332,
      "step": 129
    },
    {
      "epoch": 0.1335387776065742,
      "grad_norm": 2.779132127761841,
      "learning_rate": 1.9863013698630136e-06,
      "loss": 2.1297,
      "step": 130
    },
    {
      "epoch": 0.1335387776065742,
      "eval_loss": 2.2391879558563232,
      "eval_runtime": 55.3018,
      "eval_samples_per_second": 18.083,
      "eval_steps_per_second": 0.759,
      "step": 130
    },
    {
      "epoch": 0.13456599897277863,
      "grad_norm": 2.703019618988037,
      "learning_rate": 2.0034246575342467e-06,
      "loss": 2.0506,
      "step": 131
    },
    {
      "epoch": 0.13559322033898305,
      "grad_norm": 3.0187575817108154,
      "learning_rate": 2.0205479452054797e-06,
      "loss": 2.1291,
      "step": 132
    },
    {
      "epoch": 0.13662044170518747,
      "grad_norm": 2.7629971504211426,
      "learning_rate": 2.037671232876713e-06,
      "loss": 2.1003,
      "step": 133
    },
    {
      "epoch": 0.1376476630713919,
      "grad_norm": 3.175776720046997,
      "learning_rate": 2.0547945205479454e-06,
      "loss": 2.0429,
      "step": 134
    },
    {
      "epoch": 0.1386748844375963,
      "grad_norm": 3.0342514514923096,
      "learning_rate": 2.071917808219178e-06,
      "loss": 2.0858,
      "step": 135
    },
    {
      "epoch": 0.1397021058038007,
      "grad_norm": 3.0861928462982178,
      "learning_rate": 2.089041095890411e-06,
      "loss": 2.0964,
      "step": 136
    },
    {
      "epoch": 0.14072932717000514,
      "grad_norm": 3.151945114135742,
      "learning_rate": 2.106164383561644e-06,
      "loss": 2.0194,
      "step": 137
    },
    {
      "epoch": 0.14175654853620956,
      "grad_norm": 2.9069676399230957,
      "learning_rate": 2.123287671232877e-06,
      "loss": 2.1131,
      "step": 138
    },
    {
      "epoch": 0.14278376990241398,
      "grad_norm": 3.2285282611846924,
      "learning_rate": 2.14041095890411e-06,
      "loss": 2.0643,
      "step": 139
    },
    {
      "epoch": 0.14381099126861838,
      "grad_norm": 2.500481367111206,
      "learning_rate": 2.1575342465753425e-06,
      "loss": 2.0625,
      "step": 140
    },
    {
      "epoch": 0.14381099126861838,
      "eval_loss": 2.211498975753784,
      "eval_runtime": 55.1046,
      "eval_samples_per_second": 18.147,
      "eval_steps_per_second": 0.762,
      "step": 140
    },
    {
      "epoch": 0.1448382126348228,
      "grad_norm": 4.004592418670654,
      "learning_rate": 2.1746575342465755e-06,
      "loss": 2.0644,
      "step": 141
    },
    {
      "epoch": 0.14586543400102722,
      "grad_norm": 2.9258391857147217,
      "learning_rate": 2.191780821917808e-06,
      "loss": 2.0174,
      "step": 142
    },
    {
      "epoch": 0.14689265536723164,
      "grad_norm": 3.0073509216308594,
      "learning_rate": 2.2089041095890412e-06,
      "loss": 2.1417,
      "step": 143
    },
    {
      "epoch": 0.14791987673343607,
      "grad_norm": 3.040708065032959,
      "learning_rate": 2.2260273972602743e-06,
      "loss": 2.0357,
      "step": 144
    },
    {
      "epoch": 0.14894709809964046,
      "grad_norm": 2.9062869548797607,
      "learning_rate": 2.243150684931507e-06,
      "loss": 2.0122,
      "step": 145
    },
    {
      "epoch": 0.14997431946584489,
      "grad_norm": 3.28969669342041,
      "learning_rate": 2.26027397260274e-06,
      "loss": 2.0255,
      "step": 146
    },
    {
      "epoch": 0.1510015408320493,
      "grad_norm": 3.608752727508545,
      "learning_rate": 2.2773972602739726e-06,
      "loss": 2.0184,
      "step": 147
    },
    {
      "epoch": 0.15202876219825373,
      "grad_norm": 2.900341033935547,
      "learning_rate": 2.2945205479452057e-06,
      "loss": 2.0532,
      "step": 148
    },
    {
      "epoch": 0.15305598356445815,
      "grad_norm": 2.881843328475952,
      "learning_rate": 2.3116438356164387e-06,
      "loss": 2.1336,
      "step": 149
    },
    {
      "epoch": 0.15408320493066255,
      "grad_norm": 3.4043259620666504,
      "learning_rate": 2.3287671232876713e-06,
      "loss": 2.0536,
      "step": 150
    },
    {
      "epoch": 0.15408320493066255,
      "eval_loss": 2.192880153656006,
      "eval_runtime": 53.7838,
      "eval_samples_per_second": 18.593,
      "eval_steps_per_second": 0.781,
      "step": 150
    },
    {
      "epoch": 0.15511042629686697,
      "grad_norm": 2.693107843399048,
      "learning_rate": 2.3458904109589044e-06,
      "loss": 1.9503,
      "step": 151
    },
    {
      "epoch": 0.1561376476630714,
      "grad_norm": 2.885260581970215,
      "learning_rate": 2.363013698630137e-06,
      "loss": 2.0081,
      "step": 152
    },
    {
      "epoch": 0.15716486902927582,
      "grad_norm": 3.4459924697875977,
      "learning_rate": 2.38013698630137e-06,
      "loss": 2.0649,
      "step": 153
    },
    {
      "epoch": 0.15819209039548024,
      "grad_norm": 2.8447093963623047,
      "learning_rate": 2.3972602739726027e-06,
      "loss": 2.0476,
      "step": 154
    },
    {
      "epoch": 0.15921931176168463,
      "grad_norm": 2.755176305770874,
      "learning_rate": 2.4143835616438358e-06,
      "loss": 2.0006,
      "step": 155
    },
    {
      "epoch": 0.16024653312788906,
      "grad_norm": 2.8677561283111572,
      "learning_rate": 2.431506849315069e-06,
      "loss": 1.9663,
      "step": 156
    },
    {
      "epoch": 0.16127375449409348,
      "grad_norm": 2.783351182937622,
      "learning_rate": 2.4486301369863015e-06,
      "loss": 2.0301,
      "step": 157
    },
    {
      "epoch": 0.1623009758602979,
      "grad_norm": 2.7545998096466064,
      "learning_rate": 2.4657534246575345e-06,
      "loss": 2.0438,
      "step": 158
    },
    {
      "epoch": 0.1633281972265023,
      "grad_norm": 2.8140594959259033,
      "learning_rate": 2.482876712328767e-06,
      "loss": 2.0055,
      "step": 159
    },
    {
      "epoch": 0.16435541859270672,
      "grad_norm": 2.6072335243225098,
      "learning_rate": 2.5e-06,
      "loss": 2.0262,
      "step": 160
    },
    {
      "epoch": 0.16435541859270672,
      "eval_loss": 2.179842710494995,
      "eval_runtime": 52.6159,
      "eval_samples_per_second": 19.006,
      "eval_steps_per_second": 0.798,
      "step": 160
    },
    {
      "epoch": 0.16538263995891114,
      "grad_norm": 2.830212354660034,
      "learning_rate": 2.5171232876712333e-06,
      "loss": 2.0154,
      "step": 161
    },
    {
      "epoch": 0.16640986132511557,
      "grad_norm": 2.677140712738037,
      "learning_rate": 2.534246575342466e-06,
      "loss": 1.9933,
      "step": 162
    },
    {
      "epoch": 0.16743708269132,
      "grad_norm": 2.837759017944336,
      "learning_rate": 2.551369863013699e-06,
      "loss": 1.9602,
      "step": 163
    },
    {
      "epoch": 0.16846430405752438,
      "grad_norm": 3.9556589126586914,
      "learning_rate": 2.568493150684932e-06,
      "loss": 2.0121,
      "step": 164
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 2.8715925216674805,
      "learning_rate": 2.585616438356164e-06,
      "loss": 2.0219,
      "step": 165
    },
    {
      "epoch": 0.17051874678993323,
      "grad_norm": 3.5073623657226562,
      "learning_rate": 2.6027397260273973e-06,
      "loss": 2.0202,
      "step": 166
    },
    {
      "epoch": 0.17154596815613765,
      "grad_norm": 2.751962900161743,
      "learning_rate": 2.6198630136986303e-06,
      "loss": 1.9528,
      "step": 167
    },
    {
      "epoch": 0.17257318952234207,
      "grad_norm": 3.7753474712371826,
      "learning_rate": 2.6369863013698634e-06,
      "loss": 2.0208,
      "step": 168
    },
    {
      "epoch": 0.17360041088854647,
      "grad_norm": 2.642289876937866,
      "learning_rate": 2.654109589041096e-06,
      "loss": 2.0386,
      "step": 169
    },
    {
      "epoch": 0.1746276322547509,
      "grad_norm": 2.82173490524292,
      "learning_rate": 2.671232876712329e-06,
      "loss": 2.0355,
      "step": 170
    },
    {
      "epoch": 0.1746276322547509,
      "eval_loss": 2.1699249744415283,
      "eval_runtime": 50.0341,
      "eval_samples_per_second": 19.986,
      "eval_steps_per_second": 0.839,
      "step": 170
    },
    {
      "epoch": 0.17565485362095531,
      "grad_norm": 3.342280149459839,
      "learning_rate": 2.688356164383562e-06,
      "loss": 1.9672,
      "step": 171
    },
    {
      "epoch": 0.17668207498715974,
      "grad_norm": 2.6548173427581787,
      "learning_rate": 2.705479452054795e-06,
      "loss": 1.9669,
      "step": 172
    },
    {
      "epoch": 0.17770929635336416,
      "grad_norm": 2.6548173427581787,
      "learning_rate": 2.705479452054795e-06,
      "loss": 1.961,
      "step": 173
    },
    {
      "epoch": 0.17873651771956856,
      "grad_norm": 3.4306116104125977,
      "learning_rate": 2.7226027397260274e-06,
      "loss": 1.9173,
      "step": 174
    },
    {
      "epoch": 0.17976373908577298,
      "grad_norm": 2.7991764545440674,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 2.0339,
      "step": 175
    },
    {
      "epoch": 0.1807909604519774,
      "grad_norm": 3.6040217876434326,
      "learning_rate": 2.7568493150684935e-06,
      "loss": 2.0218,
      "step": 176
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 3.0253515243530273,
      "learning_rate": 2.773972602739726e-06,
      "loss": 1.9245,
      "step": 177
    },
    {
      "epoch": 0.18284540318438625,
      "grad_norm": 3.405141830444336,
      "learning_rate": 2.791095890410959e-06,
      "loss": 2.0136,
      "step": 178
    },
    {
      "epoch": 0.18387262455059064,
      "grad_norm": 2.7518203258514404,
      "learning_rate": 2.8082191780821922e-06,
      "loss": 1.8925,
      "step": 179
    },
    {
      "epoch": 0.18489984591679506,
      "grad_norm": 3.1903367042541504,
      "learning_rate": 2.8253424657534253e-06,
      "loss": 2.0523,
      "step": 180
    },
    {
      "epoch": 0.18489984591679506,
      "eval_loss": 2.158029794692993,
      "eval_runtime": 48.6162,
      "eval_samples_per_second": 20.569,
      "eval_steps_per_second": 0.864,
      "step": 180
    },
    {
      "epoch": 0.1859270672829995,
      "grad_norm": 3.0331413745880127,
      "learning_rate": 2.842465753424658e-06,
      "loss": 2.0167,
      "step": 181
    },
    {
      "epoch": 0.1869542886492039,
      "grad_norm": 2.919433116912842,
      "learning_rate": 2.8595890410958905e-06,
      "loss": 2.0197,
      "step": 182
    },
    {
      "epoch": 0.18798151001540833,
      "grad_norm": 2.9013049602508545,
      "learning_rate": 2.876712328767123e-06,
      "loss": 2.0301,
      "step": 183
    },
    {
      "epoch": 0.18900873138161273,
      "grad_norm": 2.738800525665283,
      "learning_rate": 2.8938356164383562e-06,
      "loss": 1.894,
      "step": 184
    },
    {
      "epoch": 0.19003595274781715,
      "grad_norm": 2.670806407928467,
      "learning_rate": 2.9109589041095893e-06,
      "loss": 1.9127,
      "step": 185
    },
    {
      "epoch": 0.19106317411402157,
      "grad_norm": 3.0435004234313965,
      "learning_rate": 2.9280821917808223e-06,
      "loss": 1.8996,
      "step": 186
    },
    {
      "epoch": 0.192090395480226,
      "grad_norm": 3.2139949798583984,
      "learning_rate": 2.945205479452055e-06,
      "loss": 1.9255,
      "step": 187
    },
    {
      "epoch": 0.19311761684643042,
      "grad_norm": 3.252965211868286,
      "learning_rate": 2.962328767123288e-06,
      "loss": 1.9809,
      "step": 188
    },
    {
      "epoch": 0.1941448382126348,
      "grad_norm": 2.9325191974639893,
      "learning_rate": 2.979452054794521e-06,
      "loss": 1.9691,
      "step": 189
    },
    {
      "epoch": 0.19517205957883924,
      "grad_norm": 3.5879862308502197,
      "learning_rate": 2.9965753424657533e-06,
      "loss": 1.9205,
      "step": 190
    },
    {
      "epoch": 0.19517205957883924,
      "eval_loss": 2.13791561126709,
      "eval_runtime": 44.8545,
      "eval_samples_per_second": 22.294,
      "eval_steps_per_second": 0.936,
      "step": 190
    },
    {
      "epoch": 0.19619928094504366,
      "grad_norm": 3.0025253295898438,
      "learning_rate": 3.0136986301369864e-06,
      "loss": 1.9717,
      "step": 191
    },
    {
      "epoch": 0.19722650231124808,
      "grad_norm": 3.1785972118377686,
      "learning_rate": 3.0308219178082194e-06,
      "loss": 1.9535,
      "step": 192
    },
    {
      "epoch": 0.1982537236774525,
      "grad_norm": 3.121065378189087,
      "learning_rate": 3.0479452054794525e-06,
      "loss": 1.9026,
      "step": 193
    },
    {
      "epoch": 0.1992809450436569,
      "grad_norm": 2.838779926300049,
      "learning_rate": 3.065068493150685e-06,
      "loss": 1.941,
      "step": 194
    },
    {
      "epoch": 0.20030816640986132,
      "grad_norm": 2.8916969299316406,
      "learning_rate": 3.082191780821918e-06,
      "loss": 1.8983,
      "step": 195
    },
    {
      "epoch": 0.20133538777606574,
      "grad_norm": 2.9195168018341064,
      "learning_rate": 3.099315068493151e-06,
      "loss": 1.918,
      "step": 196
    },
    {
      "epoch": 0.20236260914227017,
      "grad_norm": 3.235989809036255,
      "learning_rate": 3.1164383561643843e-06,
      "loss": 1.9111,
      "step": 197
    },
    {
      "epoch": 0.2033898305084746,
      "grad_norm": 3.0316455364227295,
      "learning_rate": 3.1335616438356165e-06,
      "loss": 1.8839,
      "step": 198
    },
    {
      "epoch": 0.20441705187467898,
      "grad_norm": 2.963610887527466,
      "learning_rate": 3.1506849315068495e-06,
      "loss": 2.0303,
      "step": 199
    },
    {
      "epoch": 0.2054442732408834,
      "grad_norm": 2.944054365158081,
      "learning_rate": 3.167808219178082e-06,
      "loss": 1.8822,
      "step": 200
    },
    {
      "epoch": 0.2054442732408834,
      "eval_loss": 2.1433050632476807,
      "eval_runtime": 45.8205,
      "eval_samples_per_second": 21.824,
      "eval_steps_per_second": 0.917,
      "step": 200
    },
    {
      "epoch": 0.20647149460708783,
      "grad_norm": 2.8648695945739746,
      "learning_rate": 3.184931506849315e-06,
      "loss": 1.8841,
      "step": 201
    },
    {
      "epoch": 0.20749871597329225,
      "grad_norm": 3.2838797569274902,
      "learning_rate": 3.2020547945205483e-06,
      "loss": 1.8623,
      "step": 202
    },
    {
      "epoch": 0.20852593733949668,
      "grad_norm": 2.7965173721313477,
      "learning_rate": 3.2191780821917813e-06,
      "loss": 1.8521,
      "step": 203
    },
    {
      "epoch": 0.20955315870570107,
      "grad_norm": 3.3449349403381348,
      "learning_rate": 3.236301369863014e-06,
      "loss": 1.84,
      "step": 204
    },
    {
      "epoch": 0.2105803800719055,
      "grad_norm": 3.4888951778411865,
      "learning_rate": 3.253424657534247e-06,
      "loss": 1.824,
      "step": 205
    },
    {
      "epoch": 0.21160760143810992,
      "grad_norm": 2.991349220275879,
      "learning_rate": 3.2705479452054796e-06,
      "loss": 1.9101,
      "step": 206
    },
    {
      "epoch": 0.21263482280431434,
      "grad_norm": 3.313509702682495,
      "learning_rate": 3.2876712328767123e-06,
      "loss": 1.8912,
      "step": 207
    },
    {
      "epoch": 0.21366204417051873,
      "grad_norm": 2.8260746002197266,
      "learning_rate": 3.3047945205479453e-06,
      "loss": 1.8555,
      "step": 208
    },
    {
      "epoch": 0.21468926553672316,
      "grad_norm": 2.7131688594818115,
      "learning_rate": 3.3219178082191784e-06,
      "loss": 1.8693,
      "step": 209
    },
    {
      "epoch": 0.21571648690292758,
      "grad_norm": 2.5836338996887207,
      "learning_rate": 3.3390410958904114e-06,
      "loss": 1.8965,
      "step": 210
    },
    {
      "epoch": 0.21571648690292758,
      "eval_loss": 2.1232542991638184,
      "eval_runtime": 42.3352,
      "eval_samples_per_second": 23.621,
      "eval_steps_per_second": 0.992,
      "step": 210
    },
    {
      "epoch": 0.216743708269132,
      "grad_norm": 2.8341355323791504,
      "learning_rate": 3.356164383561644e-06,
      "loss": 1.9131,
      "step": 211
    },
    {
      "epoch": 0.21777092963533642,
      "grad_norm": 2.490853786468506,
      "learning_rate": 3.373287671232877e-06,
      "loss": 1.9626,
      "step": 212
    },
    {
      "epoch": 0.21879815100154082,
      "grad_norm": 2.597956895828247,
      "learning_rate": 3.39041095890411e-06,
      "loss": 1.9203,
      "step": 213
    },
    {
      "epoch": 0.21982537236774524,
      "grad_norm": 3.1104283332824707,
      "learning_rate": 3.4075342465753424e-06,
      "loss": 1.9442,
      "step": 214
    },
    {
      "epoch": 0.22085259373394966,
      "grad_norm": 2.5698931217193604,
      "learning_rate": 3.4246575342465754e-06,
      "loss": 1.8385,
      "step": 215
    },
    {
      "epoch": 0.2218798151001541,
      "grad_norm": 3.1637446880340576,
      "learning_rate": 3.4417808219178085e-06,
      "loss": 1.9005,
      "step": 216
    },
    {
      "epoch": 0.2229070364663585,
      "grad_norm": 2.764070749282837,
      "learning_rate": 3.458904109589041e-06,
      "loss": 1.8886,
      "step": 217
    },
    {
      "epoch": 0.2239342578325629,
      "grad_norm": 3.303781032562256,
      "learning_rate": 3.476027397260274e-06,
      "loss": 1.9085,
      "step": 218
    },
    {
      "epoch": 0.22496147919876733,
      "grad_norm": 2.9543044567108154,
      "learning_rate": 3.4931506849315072e-06,
      "loss": 1.9014,
      "step": 219
    },
    {
      "epoch": 0.22598870056497175,
      "grad_norm": 2.7277028560638428,
      "learning_rate": 3.5102739726027403e-06,
      "loss": 1.9017,
      "step": 220
    },
    {
      "epoch": 0.22598870056497175,
      "eval_loss": 2.1024081707000732,
      "eval_runtime": 45.4966,
      "eval_samples_per_second": 21.98,
      "eval_steps_per_second": 0.923,
      "step": 220
    },
    {
      "epoch": 0.22701592193117617,
      "grad_norm": 2.864140033721924,
      "learning_rate": 3.527397260273973e-06,
      "loss": 1.8935,
      "step": 221
    },
    {
      "epoch": 0.2280431432973806,
      "grad_norm": 3.4133172035217285,
      "learning_rate": 3.5445205479452056e-06,
      "loss": 1.874,
      "step": 222
    },
    {
      "epoch": 0.229070364663585,
      "grad_norm": 3.33343243598938,
      "learning_rate": 3.5616438356164386e-06,
      "loss": 1.8789,
      "step": 223
    },
    {
      "epoch": 0.2300975860297894,
      "grad_norm": 3.5947320461273193,
      "learning_rate": 3.5787671232876712e-06,
      "loss": 1.9529,
      "step": 224
    },
    {
      "epoch": 0.23112480739599384,
      "grad_norm": 3.2194948196411133,
      "learning_rate": 3.5958904109589043e-06,
      "loss": 1.9205,
      "step": 225
    },
    {
      "epoch": 0.23215202876219826,
      "grad_norm": 3.3137149810791016,
      "learning_rate": 3.6130136986301373e-06,
      "loss": 1.8584,
      "step": 226
    },
    {
      "epoch": 0.23317925012840268,
      "grad_norm": 3.3358561992645264,
      "learning_rate": 3.6301369863013704e-06,
      "loss": 1.8817,
      "step": 227
    },
    {
      "epoch": 0.23420647149460708,
      "grad_norm": 3.8612711429595947,
      "learning_rate": 3.647260273972603e-06,
      "loss": 1.9189,
      "step": 228
    },
    {
      "epoch": 0.2352336928608115,
      "grad_norm": 3.0039336681365967,
      "learning_rate": 3.664383561643836e-06,
      "loss": 1.92,
      "step": 229
    },
    {
      "epoch": 0.23626091422701592,
      "grad_norm": 3.8983118534088135,
      "learning_rate": 3.6815068493150687e-06,
      "loss": 1.8512,
      "step": 230
    },
    {
      "epoch": 0.23626091422701592,
      "eval_loss": 2.1237711906433105,
      "eval_runtime": 45.6126,
      "eval_samples_per_second": 21.924,
      "eval_steps_per_second": 0.921,
      "step": 230
    },
    {
      "epoch": 0.23728813559322035,
      "grad_norm": 3.061300039291382,
      "learning_rate": 3.6986301369863014e-06,
      "loss": 1.8772,
      "step": 231
    },
    {
      "epoch": 0.23831535695942477,
      "grad_norm": 3.3102450370788574,
      "learning_rate": 3.7157534246575344e-06,
      "loss": 1.8395,
      "step": 232
    },
    {
      "epoch": 0.23934257832562916,
      "grad_norm": 2.9634461402893066,
      "learning_rate": 3.7328767123287675e-06,
      "loss": 1.9447,
      "step": 233
    },
    {
      "epoch": 0.24036979969183359,
      "grad_norm": 3.0937695503234863,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 1.8628,
      "step": 234
    },
    {
      "epoch": 0.241397021058038,
      "grad_norm": 2.8129842281341553,
      "learning_rate": 3.767123287671233e-06,
      "loss": 1.8891,
      "step": 235
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 2.9616780281066895,
      "learning_rate": 3.784246575342466e-06,
      "loss": 1.9112,
      "step": 236
    },
    {
      "epoch": 0.24345146379044685,
      "grad_norm": 2.837721109390259,
      "learning_rate": 3.8013698630136993e-06,
      "loss": 1.8514,
      "step": 237
    },
    {
      "epoch": 0.24447868515665125,
      "grad_norm": 3.137784004211426,
      "learning_rate": 3.818493150684932e-06,
      "loss": 1.8764,
      "step": 238
    },
    {
      "epoch": 0.24550590652285567,
      "grad_norm": 2.9692037105560303,
      "learning_rate": 3.8356164383561645e-06,
      "loss": 1.8201,
      "step": 239
    },
    {
      "epoch": 0.2465331278890601,
      "grad_norm": 2.6718621253967285,
      "learning_rate": 3.852739726027397e-06,
      "loss": 1.9309,
      "step": 240
    },
    {
      "epoch": 0.2465331278890601,
      "eval_loss": 2.101331949234009,
      "eval_runtime": 45.6346,
      "eval_samples_per_second": 21.913,
      "eval_steps_per_second": 0.92,
      "step": 240
    },
    {
      "epoch": 0.24756034925526452,
      "grad_norm": 3.411193609237671,
      "learning_rate": 3.869863013698631e-06,
      "loss": 1.9171,
      "step": 241
    },
    {
      "epoch": 0.24858757062146894,
      "grad_norm": 2.8848962783813477,
      "learning_rate": 3.886986301369863e-06,
      "loss": 1.8587,
      "step": 242
    },
    {
      "epoch": 0.24961479198767333,
      "grad_norm": 2.756298065185547,
      "learning_rate": 3.904109589041096e-06,
      "loss": 1.8614,
      "step": 243
    },
    {
      "epoch": 0.25064201335387776,
      "grad_norm": 2.8048956394195557,
      "learning_rate": 3.921232876712329e-06,
      "loss": 1.8225,
      "step": 244
    },
    {
      "epoch": 0.2516692347200822,
      "grad_norm": 2.9872403144836426,
      "learning_rate": 3.938356164383562e-06,
      "loss": 1.8367,
      "step": 245
    },
    {
      "epoch": 0.2526964560862866,
      "grad_norm": 2.710156202316284,
      "learning_rate": 3.9554794520547955e-06,
      "loss": 1.809,
      "step": 246
    },
    {
      "epoch": 0.253723677452491,
      "grad_norm": 3.2177000045776367,
      "learning_rate": 3.972602739726027e-06,
      "loss": 1.8883,
      "step": 247
    },
    {
      "epoch": 0.25475089881869545,
      "grad_norm": 3.160964250564575,
      "learning_rate": 3.989726027397261e-06,
      "loss": 1.8267,
      "step": 248
    },
    {
      "epoch": 0.25577812018489987,
      "grad_norm": 2.853498935699463,
      "learning_rate": 4.006849315068493e-06,
      "loss": 1.8852,
      "step": 249
    },
    {
      "epoch": 0.25680534155110424,
      "grad_norm": 2.571742296218872,
      "learning_rate": 4.023972602739726e-06,
      "loss": 1.8833,
      "step": 250
    },
    {
      "epoch": 0.25680534155110424,
      "eval_loss": 2.0988597869873047,
      "eval_runtime": 47.7045,
      "eval_samples_per_second": 20.962,
      "eval_steps_per_second": 0.88,
      "step": 250
    },
    {
      "epoch": 0.25783256291730866,
      "grad_norm": 2.776742458343506,
      "learning_rate": 4.0410958904109595e-06,
      "loss": 1.8662,
      "step": 251
    },
    {
      "epoch": 0.2588597842835131,
      "grad_norm": 2.8038573265075684,
      "learning_rate": 4.058219178082192e-06,
      "loss": 1.8528,
      "step": 252
    },
    {
      "epoch": 0.2598870056497175,
      "grad_norm": 2.557623863220215,
      "learning_rate": 4.075342465753426e-06,
      "loss": 1.8765,
      "step": 253
    },
    {
      "epoch": 0.26091422701592193,
      "grad_norm": 3.0477073192596436,
      "learning_rate": 4.092465753424658e-06,
      "loss": 1.8434,
      "step": 254
    },
    {
      "epoch": 0.26194144838212635,
      "grad_norm": 2.6221649646759033,
      "learning_rate": 4.109589041095891e-06,
      "loss": 1.8348,
      "step": 255
    },
    {
      "epoch": 0.2629686697483308,
      "grad_norm": 3.0849292278289795,
      "learning_rate": 4.1267123287671235e-06,
      "loss": 1.9057,
      "step": 256
    },
    {
      "epoch": 0.2639958911145352,
      "grad_norm": 2.795184373855591,
      "learning_rate": 4.143835616438356e-06,
      "loss": 1.8528,
      "step": 257
    },
    {
      "epoch": 0.2650231124807396,
      "grad_norm": 2.887193441390991,
      "learning_rate": 4.16095890410959e-06,
      "loss": 1.9176,
      "step": 258
    },
    {
      "epoch": 0.26605033384694404,
      "grad_norm": 2.776801347732544,
      "learning_rate": 4.178082191780822e-06,
      "loss": 1.8911,
      "step": 259
    },
    {
      "epoch": 0.2670775552131484,
      "grad_norm": 2.980426788330078,
      "learning_rate": 4.195205479452055e-06,
      "loss": 1.8288,
      "step": 260
    },
    {
      "epoch": 0.2670775552131484,
      "eval_loss": 2.0636370182037354,
      "eval_runtime": 49.3081,
      "eval_samples_per_second": 20.281,
      "eval_steps_per_second": 0.852,
      "step": 260
    },
    {
      "epoch": 0.26810477657935283,
      "grad_norm": 2.7921340465545654,
      "learning_rate": 4.212328767123288e-06,
      "loss": 1.8405,
      "step": 261
    },
    {
      "epoch": 0.26913199794555726,
      "grad_norm": 2.67714262008667,
      "learning_rate": 4.229452054794521e-06,
      "loss": 1.8342,
      "step": 262
    },
    {
      "epoch": 0.2701592193117617,
      "grad_norm": 2.819868326187134,
      "learning_rate": 4.246575342465754e-06,
      "loss": 1.8466,
      "step": 263
    },
    {
      "epoch": 0.2711864406779661,
      "grad_norm": 2.611250638961792,
      "learning_rate": 4.263698630136986e-06,
      "loss": 1.8314,
      "step": 264
    },
    {
      "epoch": 0.2722136620441705,
      "grad_norm": 3.0406076908111572,
      "learning_rate": 4.28082191780822e-06,
      "loss": 1.7924,
      "step": 265
    },
    {
      "epoch": 0.27324088341037495,
      "grad_norm": 3.212790012359619,
      "learning_rate": 4.297945205479452e-06,
      "loss": 1.9479,
      "step": 266
    },
    {
      "epoch": 0.27426810477657937,
      "grad_norm": 2.4767045974731445,
      "learning_rate": 4.315068493150685e-06,
      "loss": 1.835,
      "step": 267
    },
    {
      "epoch": 0.2752953261427838,
      "grad_norm": 3.191835641860962,
      "learning_rate": 4.3321917808219185e-06,
      "loss": 1.8261,
      "step": 268
    },
    {
      "epoch": 0.2763225475089882,
      "grad_norm": 2.9970273971557617,
      "learning_rate": 4.349315068493151e-06,
      "loss": 1.8427,
      "step": 269
    },
    {
      "epoch": 0.2773497688751926,
      "grad_norm": 2.888843297958374,
      "learning_rate": 4.3664383561643846e-06,
      "loss": 1.819,
      "step": 270
    },
    {
      "epoch": 0.2773497688751926,
      "eval_loss": 2.054966926574707,
      "eval_runtime": 45.9212,
      "eval_samples_per_second": 21.776,
      "eval_steps_per_second": 0.915,
      "step": 270
    },
    {
      "epoch": 0.278376990241397,
      "grad_norm": 2.7771849632263184,
      "learning_rate": 4.383561643835616e-06,
      "loss": 1.8178,
      "step": 271
    },
    {
      "epoch": 0.2794042116076014,
      "grad_norm": 2.9749014377593994,
      "learning_rate": 4.40068493150685e-06,
      "loss": 1.8982,
      "step": 272
    },
    {
      "epoch": 0.28043143297380585,
      "grad_norm": 3.076021432876587,
      "learning_rate": 4.4178082191780825e-06,
      "loss": 1.8575,
      "step": 273
    },
    {
      "epoch": 0.2814586543400103,
      "grad_norm": 3.2113568782806396,
      "learning_rate": 4.434931506849315e-06,
      "loss": 1.8241,
      "step": 274
    },
    {
      "epoch": 0.2824858757062147,
      "grad_norm": 3.322063684463501,
      "learning_rate": 4.4520547945205486e-06,
      "loss": 1.7922,
      "step": 275
    },
    {
      "epoch": 0.2835130970724191,
      "grad_norm": 3.2778918743133545,
      "learning_rate": 4.469178082191781e-06,
      "loss": 1.8412,
      "step": 276
    },
    {
      "epoch": 0.28454031843862354,
      "grad_norm": 3.1802618503570557,
      "learning_rate": 4.486301369863014e-06,
      "loss": 1.8283,
      "step": 277
    },
    {
      "epoch": 0.28556753980482796,
      "grad_norm": 2.9214258193969727,
      "learning_rate": 4.503424657534247e-06,
      "loss": 1.8817,
      "step": 278
    },
    {
      "epoch": 0.28659476117103233,
      "grad_norm": 2.8809854984283447,
      "learning_rate": 4.52054794520548e-06,
      "loss": 1.7662,
      "step": 279
    },
    {
      "epoch": 0.28762198253723675,
      "grad_norm": 2.925537586212158,
      "learning_rate": 4.537671232876713e-06,
      "loss": 1.8147,
      "step": 280
    },
    {
      "epoch": 0.28762198253723675,
      "eval_loss": 2.054513454437256,
      "eval_runtime": 51.5036,
      "eval_samples_per_second": 19.416,
      "eval_steps_per_second": 0.815,
      "step": 280
    },
    {
      "epoch": 0.2886492039034412,
      "grad_norm": 2.9421887397766113,
      "learning_rate": 4.554794520547945e-06,
      "loss": 1.8581,
      "step": 281
    },
    {
      "epoch": 0.2896764252696456,
      "grad_norm": 2.9733567237854004,
      "learning_rate": 4.571917808219179e-06,
      "loss": 1.8703,
      "step": 282
    },
    {
      "epoch": 0.29070364663585,
      "grad_norm": 2.5352156162261963,
      "learning_rate": 4.589041095890411e-06,
      "loss": 1.8505,
      "step": 283
    },
    {
      "epoch": 0.29173086800205444,
      "grad_norm": 2.773847818374634,
      "learning_rate": 4.606164383561644e-06,
      "loss": 1.8114,
      "step": 284
    },
    {
      "epoch": 0.29275808936825887,
      "grad_norm": 2.709632635116577,
      "learning_rate": 4.6232876712328774e-06,
      "loss": 1.815,
      "step": 285
    },
    {
      "epoch": 0.2937853107344633,
      "grad_norm": 3.1234357357025146,
      "learning_rate": 4.64041095890411e-06,
      "loss": 1.8391,
      "step": 286
    },
    {
      "epoch": 0.2948125321006677,
      "grad_norm": 2.9376144409179688,
      "learning_rate": 4.657534246575343e-06,
      "loss": 1.7998,
      "step": 287
    },
    {
      "epoch": 0.29583975346687214,
      "grad_norm": 3.0812032222747803,
      "learning_rate": 4.674657534246575e-06,
      "loss": 1.8462,
      "step": 288
    },
    {
      "epoch": 0.2968669748330765,
      "grad_norm": 2.922722578048706,
      "learning_rate": 4.691780821917809e-06,
      "loss": 1.8263,
      "step": 289
    },
    {
      "epoch": 0.2978941961992809,
      "grad_norm": 2.9484992027282715,
      "learning_rate": 4.7089041095890414e-06,
      "loss": 1.8754,
      "step": 290
    },
    {
      "epoch": 0.2978941961992809,
      "eval_loss": 2.032902240753174,
      "eval_runtime": 47.8711,
      "eval_samples_per_second": 20.889,
      "eval_steps_per_second": 0.877,
      "step": 290
    },
    {
      "epoch": 0.29892141756548535,
      "grad_norm": 3.0068976879119873,
      "learning_rate": 4.726027397260274e-06,
      "loss": 1.8256,
      "step": 291
    },
    {
      "epoch": 0.29994863893168977,
      "grad_norm": 2.6469006538391113,
      "learning_rate": 4.7431506849315075e-06,
      "loss": 1.8354,
      "step": 292
    },
    {
      "epoch": 0.3009758602978942,
      "grad_norm": 3.06986927986145,
      "learning_rate": 4.76027397260274e-06,
      "loss": 1.8402,
      "step": 293
    },
    {
      "epoch": 0.3020030816640986,
      "grad_norm": 2.8409981727600098,
      "learning_rate": 4.777397260273973e-06,
      "loss": 1.7915,
      "step": 294
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 3.1027376651763916,
      "learning_rate": 4.7945205479452054e-06,
      "loss": 1.8153,
      "step": 295
    },
    {
      "epoch": 0.30405752439650746,
      "grad_norm": 2.7221434116363525,
      "learning_rate": 4.811643835616439e-06,
      "loss": 1.8098,
      "step": 296
    },
    {
      "epoch": 0.3050847457627119,
      "grad_norm": 3.0559370517730713,
      "learning_rate": 4.8287671232876716e-06,
      "loss": 1.7504,
      "step": 297
    },
    {
      "epoch": 0.3061119671289163,
      "grad_norm": 2.9628236293792725,
      "learning_rate": 4.845890410958904e-06,
      "loss": 1.8019,
      "step": 298
    },
    {
      "epoch": 0.3071391884951207,
      "grad_norm": 2.6248221397399902,
      "learning_rate": 4.863013698630138e-06,
      "loss": 1.8001,
      "step": 299
    },
    {
      "epoch": 0.3081664098613251,
      "grad_norm": 3.0477702617645264,
      "learning_rate": 4.88013698630137e-06,
      "loss": 1.7915,
      "step": 300
    },
    {
      "epoch": 0.3081664098613251,
      "eval_loss": 2.0522429943084717,
      "eval_runtime": 46.796,
      "eval_samples_per_second": 21.369,
      "eval_steps_per_second": 0.898,
      "step": 300
    },
    {
      "epoch": 0.3091936312275295,
      "grad_norm": 2.784707546234131,
      "learning_rate": 4.897260273972603e-06,
      "loss": 1.7757,
      "step": 301
    },
    {
      "epoch": 0.31022085259373394,
      "grad_norm": 2.8283841609954834,
      "learning_rate": 4.914383561643836e-06,
      "loss": 1.8975,
      "step": 302
    },
    {
      "epoch": 0.31124807395993837,
      "grad_norm": 2.879342555999756,
      "learning_rate": 4.931506849315069e-06,
      "loss": 1.8208,
      "step": 303
    },
    {
      "epoch": 0.3122752953261428,
      "grad_norm": 3.2111175060272217,
      "learning_rate": 4.948630136986302e-06,
      "loss": 1.8631,
      "step": 304
    },
    {
      "epoch": 0.3133025166923472,
      "grad_norm": 2.876168966293335,
      "learning_rate": 4.965753424657534e-06,
      "loss": 1.8863,
      "step": 305
    },
    {
      "epoch": 0.31432973805855163,
      "grad_norm": 2.8958466053009033,
      "learning_rate": 4.982876712328768e-06,
      "loss": 1.7565,
      "step": 306
    },
    {
      "epoch": 0.31535695942475606,
      "grad_norm": 2.6702723503112793,
      "learning_rate": 5e-06,
      "loss": 1.7844,
      "step": 307
    },
    {
      "epoch": 0.3163841807909605,
      "grad_norm": 2.557966470718384,
      "learning_rate": 4.998096688237534e-06,
      "loss": 1.7472,
      "step": 308
    },
    {
      "epoch": 0.31741140215716485,
      "grad_norm": 2.9257633686065674,
      "learning_rate": 4.996193376475067e-06,
      "loss": 1.853,
      "step": 309
    },
    {
      "epoch": 0.31843862352336927,
      "grad_norm": 2.497093439102173,
      "learning_rate": 4.9942900647126005e-06,
      "loss": 1.7313,
      "step": 310
    },
    {
      "epoch": 0.31843862352336927,
      "eval_loss": 2.0336520671844482,
      "eval_runtime": 43.3773,
      "eval_samples_per_second": 23.054,
      "eval_steps_per_second": 0.968,
      "step": 310
    },
    {
      "epoch": 0.3194658448895737,
      "grad_norm": 3.0313165187835693,
      "learning_rate": 4.992386752950134e-06,
      "loss": 1.7397,
      "step": 311
    },
    {
      "epoch": 0.3204930662557781,
      "grad_norm": 2.629685163497925,
      "learning_rate": 4.990483441187667e-06,
      "loss": 1.7839,
      "step": 312
    },
    {
      "epoch": 0.32152028762198254,
      "grad_norm": 2.6772103309631348,
      "learning_rate": 4.9885801294252e-06,
      "loss": 1.8038,
      "step": 313
    },
    {
      "epoch": 0.32254750898818696,
      "grad_norm": 2.893958568572998,
      "learning_rate": 4.986676817662733e-06,
      "loss": 1.8072,
      "step": 314
    },
    {
      "epoch": 0.3235747303543914,
      "grad_norm": 2.5317318439483643,
      "learning_rate": 4.9847735059002665e-06,
      "loss": 1.7632,
      "step": 315
    },
    {
      "epoch": 0.3246019517205958,
      "grad_norm": 2.836867332458496,
      "learning_rate": 4.9828701941378e-06,
      "loss": 1.8499,
      "step": 316
    },
    {
      "epoch": 0.3256291730868002,
      "grad_norm": 2.5564308166503906,
      "learning_rate": 4.980966882375333e-06,
      "loss": 1.7849,
      "step": 317
    },
    {
      "epoch": 0.3266563944530046,
      "grad_norm": 2.4335806369781494,
      "learning_rate": 4.979063570612867e-06,
      "loss": 1.7425,
      "step": 318
    },
    {
      "epoch": 0.327683615819209,
      "grad_norm": 2.7140371799468994,
      "learning_rate": 4.9771602588504e-06,
      "loss": 1.7173,
      "step": 319
    },
    {
      "epoch": 0.32871083718541344,
      "grad_norm": 2.574345111846924,
      "learning_rate": 4.975256947087933e-06,
      "loss": 1.8307,
      "step": 320
    },
    {
      "epoch": 0.32871083718541344,
      "eval_loss": 2.026960611343384,
      "eval_runtime": 53.8982,
      "eval_samples_per_second": 18.553,
      "eval_steps_per_second": 0.779,
      "step": 320
    },
    {
      "epoch": 0.32973805855161786,
      "grad_norm": 2.6877281665802,
      "learning_rate": 4.973353635325467e-06,
      "loss": 1.7216,
      "step": 321
    },
    {
      "epoch": 0.3307652799178223,
      "grad_norm": 2.8152849674224854,
      "learning_rate": 4.971450323563e-06,
      "loss": 1.7831,
      "step": 322
    },
    {
      "epoch": 0.3317925012840267,
      "grad_norm": 2.4208791255950928,
      "learning_rate": 4.969547011800533e-06,
      "loss": 1.7266,
      "step": 323
    },
    {
      "epoch": 0.33281972265023113,
      "grad_norm": 2.765162229537964,
      "learning_rate": 4.967643700038067e-06,
      "loss": 1.7971,
      "step": 324
    },
    {
      "epoch": 0.33384694401643555,
      "grad_norm": 2.5018413066864014,
      "learning_rate": 4.9657403882756e-06,
      "loss": 1.7842,
      "step": 325
    },
    {
      "epoch": 0.33487416538264,
      "grad_norm": 2.901268482208252,
      "learning_rate": 4.9638370765131335e-06,
      "loss": 1.7522,
      "step": 326
    },
    {
      "epoch": 0.3359013867488444,
      "grad_norm": 2.40067195892334,
      "learning_rate": 4.961933764750667e-06,
      "loss": 1.7371,
      "step": 327
    },
    {
      "epoch": 0.33692860811504877,
      "grad_norm": 2.6963725090026855,
      "learning_rate": 4.9600304529882e-06,
      "loss": 1.8036,
      "step": 328
    },
    {
      "epoch": 0.3379558294812532,
      "grad_norm": 2.483790397644043,
      "learning_rate": 4.9581271412257335e-06,
      "loss": 1.815,
      "step": 329
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 2.7459559440612793,
      "learning_rate": 4.956223829463266e-06,
      "loss": 1.777,
      "step": 330
    },
    {
      "epoch": 0.3389830508474576,
      "eval_loss": 2.046339988708496,
      "eval_runtime": 53.909,
      "eval_samples_per_second": 18.55,
      "eval_steps_per_second": 0.779,
      "step": 330
    },
    {
      "epoch": 0.34001027221366203,
      "grad_norm": 2.5724353790283203,
      "learning_rate": 4.954320517700799e-06,
      "loss": 1.7022,
      "step": 331
    },
    {
      "epoch": 0.34103749357986646,
      "grad_norm": 2.510002613067627,
      "learning_rate": 4.952417205938333e-06,
      "loss": 1.802,
      "step": 332
    },
    {
      "epoch": 0.3420647149460709,
      "grad_norm": 2.8722083568573,
      "learning_rate": 4.950513894175866e-06,
      "loss": 1.7772,
      "step": 333
    },
    {
      "epoch": 0.3430919363122753,
      "grad_norm": 2.586747407913208,
      "learning_rate": 4.9486105824133995e-06,
      "loss": 1.7664,
      "step": 334
    },
    {
      "epoch": 0.3441191576784797,
      "grad_norm": 2.8008205890655518,
      "learning_rate": 4.946707270650933e-06,
      "loss": 1.7892,
      "step": 335
    },
    {
      "epoch": 0.34514637904468415,
      "grad_norm": 2.632664203643799,
      "learning_rate": 4.944803958888466e-06,
      "loss": 1.7967,
      "step": 336
    },
    {
      "epoch": 0.34617360041088857,
      "grad_norm": 2.8898301124572754,
      "learning_rate": 4.942900647126e-06,
      "loss": 1.8126,
      "step": 337
    },
    {
      "epoch": 0.34720082177709294,
      "grad_norm": 2.719019651412964,
      "learning_rate": 4.940997335363533e-06,
      "loss": 1.7544,
      "step": 338
    },
    {
      "epoch": 0.34822804314329736,
      "grad_norm": 2.9995970726013184,
      "learning_rate": 4.939094023601066e-06,
      "loss": 1.7142,
      "step": 339
    },
    {
      "epoch": 0.3492552645095018,
      "grad_norm": 2.9737157821655273,
      "learning_rate": 4.9371907118386e-06,
      "loss": 1.7659,
      "step": 340
    },
    {
      "epoch": 0.3492552645095018,
      "eval_loss": 2.024946928024292,
      "eval_runtime": 52.8069,
      "eval_samples_per_second": 18.937,
      "eval_steps_per_second": 0.795,
      "step": 340
    },
    {
      "epoch": 0.3502824858757062,
      "grad_norm": 2.8651719093322754,
      "learning_rate": 4.935287400076133e-06,
      "loss": 1.7767,
      "step": 341
    },
    {
      "epoch": 0.35130970724191063,
      "grad_norm": 2.6293981075286865,
      "learning_rate": 4.933384088313666e-06,
      "loss": 1.7815,
      "step": 342
    },
    {
      "epoch": 0.35233692860811505,
      "grad_norm": 3.230978012084961,
      "learning_rate": 4.9314807765512e-06,
      "loss": 1.7423,
      "step": 343
    },
    {
      "epoch": 0.3533641499743195,
      "grad_norm": 3.228300094604492,
      "learning_rate": 4.929577464788733e-06,
      "loss": 1.6845,
      "step": 344
    },
    {
      "epoch": 0.3543913713405239,
      "grad_norm": 3.036733388900757,
      "learning_rate": 4.9276741530262665e-06,
      "loss": 1.7524,
      "step": 345
    },
    {
      "epoch": 0.3554185927067283,
      "grad_norm": 2.778538227081299,
      "learning_rate": 4.9257708412638e-06,
      "loss": 1.7433,
      "step": 346
    },
    {
      "epoch": 0.35644581407293274,
      "grad_norm": 3.0958566665649414,
      "learning_rate": 4.923867529501332e-06,
      "loss": 1.76,
      "step": 347
    },
    {
      "epoch": 0.3574730354391371,
      "grad_norm": 3.1120026111602783,
      "learning_rate": 4.921964217738866e-06,
      "loss": 1.7399,
      "step": 348
    },
    {
      "epoch": 0.35850025680534153,
      "grad_norm": 2.70768141746521,
      "learning_rate": 4.920060905976399e-06,
      "loss": 1.8169,
      "step": 349
    },
    {
      "epoch": 0.35952747817154596,
      "grad_norm": 2.917478561401367,
      "learning_rate": 4.9181575942139325e-06,
      "loss": 1.7948,
      "step": 350
    },
    {
      "epoch": 0.35952747817154596,
      "eval_loss": 2.0208380222320557,
      "eval_runtime": 47.4243,
      "eval_samples_per_second": 21.086,
      "eval_steps_per_second": 0.886,
      "step": 350
    },
    {
      "epoch": 0.3605546995377504,
      "grad_norm": 3.0564756393432617,
      "learning_rate": 4.916254282451466e-06,
      "loss": 1.7861,
      "step": 351
    },
    {
      "epoch": 0.3615819209039548,
      "grad_norm": 2.709195137023926,
      "learning_rate": 4.914350970688999e-06,
      "loss": 1.8323,
      "step": 352
    },
    {
      "epoch": 0.3626091422701592,
      "grad_norm": 2.8797669410705566,
      "learning_rate": 4.9124476589265325e-06,
      "loss": 1.722,
      "step": 353
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.560702085494995,
      "learning_rate": 4.910544347164066e-06,
      "loss": 1.7289,
      "step": 354
    },
    {
      "epoch": 0.36466358500256807,
      "grad_norm": 3.3781847953796387,
      "learning_rate": 4.908641035401599e-06,
      "loss": 1.8096,
      "step": 355
    },
    {
      "epoch": 0.3656908063687725,
      "grad_norm": 2.582310676574707,
      "learning_rate": 4.906737723639133e-06,
      "loss": 1.8752,
      "step": 356
    },
    {
      "epoch": 0.3667180277349769,
      "grad_norm": 3.506385564804077,
      "learning_rate": 4.904834411876665e-06,
      "loss": 1.7404,
      "step": 357
    },
    {
      "epoch": 0.3677452491011813,
      "grad_norm": 2.4360411167144775,
      "learning_rate": 4.902931100114199e-06,
      "loss": 1.7812,
      "step": 358
    },
    {
      "epoch": 0.3687724704673857,
      "grad_norm": 3.77639102935791,
      "learning_rate": 4.901027788351733e-06,
      "loss": 1.7103,
      "step": 359
    },
    {
      "epoch": 0.3697996918335901,
      "grad_norm": 2.887723922729492,
      "learning_rate": 4.899124476589266e-06,
      "loss": 1.7485,
      "step": 360
    },
    {
      "epoch": 0.3697996918335901,
      "eval_loss": 2.0298168659210205,
      "eval_runtime": 46.0993,
      "eval_samples_per_second": 21.692,
      "eval_steps_per_second": 0.911,
      "step": 360
    },
    {
      "epoch": 0.37082691319979455,
      "grad_norm": 3.5861997604370117,
      "learning_rate": 4.8972211648267994e-06,
      "loss": 1.7189,
      "step": 361
    },
    {
      "epoch": 0.371854134565999,
      "grad_norm": 2.590808868408203,
      "learning_rate": 4.895317853064333e-06,
      "loss": 1.7683,
      "step": 362
    },
    {
      "epoch": 0.3728813559322034,
      "grad_norm": 3.4451487064361572,
      "learning_rate": 4.893414541301866e-06,
      "loss": 1.7436,
      "step": 363
    },
    {
      "epoch": 0.3739085772984078,
      "grad_norm": 2.406073808670044,
      "learning_rate": 4.891511229539399e-06,
      "loss": 1.7378,
      "step": 364
    },
    {
      "epoch": 0.37493579866461224,
      "grad_norm": 2.7162997722625732,
      "learning_rate": 4.889607917776932e-06,
      "loss": 1.7558,
      "step": 365
    },
    {
      "epoch": 0.37596302003081666,
      "grad_norm": 2.564753770828247,
      "learning_rate": 4.887704606014465e-06,
      "loss": 1.7559,
      "step": 366
    },
    {
      "epoch": 0.37699024139702103,
      "grad_norm": 2.7196900844573975,
      "learning_rate": 4.885801294251999e-06,
      "loss": 1.7683,
      "step": 367
    },
    {
      "epoch": 0.37801746276322545,
      "grad_norm": 2.9334428310394287,
      "learning_rate": 4.883897982489532e-06,
      "loss": 1.6932,
      "step": 368
    },
    {
      "epoch": 0.3790446841294299,
      "grad_norm": 2.4899954795837402,
      "learning_rate": 4.8819946707270655e-06,
      "loss": 1.7943,
      "step": 369
    },
    {
      "epoch": 0.3800719054956343,
      "grad_norm": 2.820594310760498,
      "learning_rate": 4.880091358964599e-06,
      "loss": 1.7729,
      "step": 370
    },
    {
      "epoch": 0.3800719054956343,
      "eval_loss": 2.0204291343688965,
      "eval_runtime": 44.9993,
      "eval_samples_per_second": 22.223,
      "eval_steps_per_second": 0.933,
      "step": 370
    },
    {
      "epoch": 0.3810991268618387,
      "grad_norm": 2.4717929363250732,
      "learning_rate": 4.878188047202132e-06,
      "loss": 1.6813,
      "step": 371
    },
    {
      "epoch": 0.38212634822804314,
      "grad_norm": 3.1815855503082275,
      "learning_rate": 4.876284735439666e-06,
      "loss": 1.8588,
      "step": 372
    },
    {
      "epoch": 0.38315356959424757,
      "grad_norm": 2.7330422401428223,
      "learning_rate": 4.874381423677199e-06,
      "loss": 1.8245,
      "step": 373
    },
    {
      "epoch": 0.384180790960452,
      "grad_norm": 3.129444122314453,
      "learning_rate": 4.8724781119147315e-06,
      "loss": 1.7494,
      "step": 374
    },
    {
      "epoch": 0.3852080123266564,
      "grad_norm": 2.75551700592041,
      "learning_rate": 4.870574800152265e-06,
      "loss": 1.7761,
      "step": 375
    },
    {
      "epoch": 0.38623523369286084,
      "grad_norm": 2.9313466548919678,
      "learning_rate": 4.868671488389798e-06,
      "loss": 1.7076,
      "step": 376
    },
    {
      "epoch": 0.3872624550590652,
      "grad_norm": 2.7590556144714355,
      "learning_rate": 4.8667681766273316e-06,
      "loss": 1.7105,
      "step": 377
    },
    {
      "epoch": 0.3882896764252696,
      "grad_norm": 2.8224680423736572,
      "learning_rate": 4.864864864864866e-06,
      "loss": 1.6853,
      "step": 378
    },
    {
      "epoch": 0.38931689779147405,
      "grad_norm": 2.5148062705993652,
      "learning_rate": 4.862961553102399e-06,
      "loss": 1.7477,
      "step": 379
    },
    {
      "epoch": 0.39034411915767847,
      "grad_norm": 3.144247055053711,
      "learning_rate": 4.8610582413399325e-06,
      "loss": 1.7726,
      "step": 380
    },
    {
      "epoch": 0.39034411915767847,
      "eval_loss": 2.006993532180786,
      "eval_runtime": 50.8909,
      "eval_samples_per_second": 19.65,
      "eval_steps_per_second": 0.825,
      "step": 380
    },
    {
      "epoch": 0.3913713405238829,
      "grad_norm": 2.6818125247955322,
      "learning_rate": 4.859154929577465e-06,
      "loss": 1.7351,
      "step": 381
    },
    {
      "epoch": 0.3923985618900873,
      "grad_norm": 2.8543903827667236,
      "learning_rate": 4.857251617814998e-06,
      "loss": 1.7499,
      "step": 382
    },
    {
      "epoch": 0.39342578325629174,
      "grad_norm": 2.5483179092407227,
      "learning_rate": 4.855348306052532e-06,
      "loss": 1.7849,
      "step": 383
    },
    {
      "epoch": 0.39445300462249616,
      "grad_norm": 2.7169013023376465,
      "learning_rate": 4.853444994290065e-06,
      "loss": 1.7664,
      "step": 384
    },
    {
      "epoch": 0.3954802259887006,
      "grad_norm": 2.8304696083068848,
      "learning_rate": 4.8515416825275985e-06,
      "loss": 1.73,
      "step": 385
    },
    {
      "epoch": 0.396507447354905,
      "grad_norm": 2.4217448234558105,
      "learning_rate": 4.849638370765132e-06,
      "loss": 1.7915,
      "step": 386
    },
    {
      "epoch": 0.3975346687211094,
      "grad_norm": 2.523256301879883,
      "learning_rate": 4.847735059002665e-06,
      "loss": 1.7195,
      "step": 387
    },
    {
      "epoch": 0.3985618900873138,
      "grad_norm": 2.813638925552368,
      "learning_rate": 4.8458317472401985e-06,
      "loss": 1.6766,
      "step": 388
    },
    {
      "epoch": 0.3995891114535182,
      "grad_norm": 2.6701548099517822,
      "learning_rate": 4.843928435477732e-06,
      "loss": 1.7666,
      "step": 389
    },
    {
      "epoch": 0.40061633281972264,
      "grad_norm": 2.5568342208862305,
      "learning_rate": 4.842025123715265e-06,
      "loss": 1.723,
      "step": 390
    },
    {
      "epoch": 0.40061633281972264,
      "eval_loss": 2.0098180770874023,
      "eval_runtime": 44.4503,
      "eval_samples_per_second": 22.497,
      "eval_steps_per_second": 0.945,
      "step": 390
    },
    {
      "epoch": 0.40164355418592707,
      "grad_norm": 2.6287341117858887,
      "learning_rate": 4.840121811952798e-06,
      "loss": 1.759,
      "step": 391
    },
    {
      "epoch": 0.4026707755521315,
      "grad_norm": 2.934105157852173,
      "learning_rate": 4.838218500190331e-06,
      "loss": 1.7501,
      "step": 392
    },
    {
      "epoch": 0.4036979969183359,
      "grad_norm": 2.6563878059387207,
      "learning_rate": 4.8363151884278645e-06,
      "loss": 1.7624,
      "step": 393
    },
    {
      "epoch": 0.40472521828454033,
      "grad_norm": 2.538278102874756,
      "learning_rate": 4.834411876665398e-06,
      "loss": 1.7457,
      "step": 394
    },
    {
      "epoch": 0.40575243965074476,
      "grad_norm": 2.801630735397339,
      "learning_rate": 4.832508564902931e-06,
      "loss": 1.7549,
      "step": 395
    },
    {
      "epoch": 0.4067796610169492,
      "grad_norm": 2.571937322616577,
      "learning_rate": 4.830605253140465e-06,
      "loss": 1.6954,
      "step": 396
    },
    {
      "epoch": 0.40780688238315355,
      "grad_norm": 2.5652592182159424,
      "learning_rate": 4.828701941377998e-06,
      "loss": 1.7328,
      "step": 397
    },
    {
      "epoch": 0.40883410374935797,
      "grad_norm": 2.6840121746063232,
      "learning_rate": 4.826798629615531e-06,
      "loss": 1.6934,
      "step": 398
    },
    {
      "epoch": 0.4098613251155624,
      "grad_norm": 3.0103495121002197,
      "learning_rate": 4.824895317853065e-06,
      "loss": 1.7086,
      "step": 399
    },
    {
      "epoch": 0.4108885464817668,
      "grad_norm": 2.7361223697662354,
      "learning_rate": 4.822992006090598e-06,
      "loss": 1.7301,
      "step": 400
    },
    {
      "epoch": 0.4108885464817668,
      "eval_loss": 1.9975080490112305,
      "eval_runtime": 44.5413,
      "eval_samples_per_second": 22.451,
      "eval_steps_per_second": 0.943,
      "step": 400
    },
    {
      "epoch": 0.41191576784797124,
      "grad_norm": 2.6805901527404785,
      "learning_rate": 4.821088694328131e-06,
      "loss": 1.6896,
      "step": 401
    },
    {
      "epoch": 0.41294298921417566,
      "grad_norm": 3.231706142425537,
      "learning_rate": 4.819185382565665e-06,
      "loss": 1.7134,
      "step": 402
    },
    {
      "epoch": 0.4139702105803801,
      "grad_norm": 2.5561299324035645,
      "learning_rate": 4.817282070803198e-06,
      "loss": 1.764,
      "step": 403
    },
    {
      "epoch": 0.4149974319465845,
      "grad_norm": 3.3125340938568115,
      "learning_rate": 4.8153787590407315e-06,
      "loss": 1.76,
      "step": 404
    },
    {
      "epoch": 0.41602465331278893,
      "grad_norm": 2.750051498413086,
      "learning_rate": 4.813475447278265e-06,
      "loss": 1.6882,
      "step": 405
    },
    {
      "epoch": 0.41705187467899335,
      "grad_norm": 4.006515979766846,
      "learning_rate": 4.811572135515798e-06,
      "loss": 1.7892,
      "step": 406
    },
    {
      "epoch": 0.4180790960451977,
      "grad_norm": 2.5276904106140137,
      "learning_rate": 4.809668823753332e-06,
      "loss": 1.6995,
      "step": 407
    },
    {
      "epoch": 0.41910631741140214,
      "grad_norm": 2.959444046020508,
      "learning_rate": 4.807765511990864e-06,
      "loss": 1.7058,
      "step": 408
    },
    {
      "epoch": 0.42013353877760656,
      "grad_norm": 2.8472464084625244,
      "learning_rate": 4.8058622002283975e-06,
      "loss": 1.7258,
      "step": 409
    },
    {
      "epoch": 0.421160760143811,
      "grad_norm": 3.0240273475646973,
      "learning_rate": 4.803958888465931e-06,
      "loss": 1.7789,
      "step": 410
    },
    {
      "epoch": 0.421160760143811,
      "eval_loss": 1.9919871091842651,
      "eval_runtime": 41.727,
      "eval_samples_per_second": 23.965,
      "eval_steps_per_second": 1.007,
      "step": 410
    },
    {
      "epoch": 0.4221879815100154,
      "grad_norm": 2.6211893558502197,
      "learning_rate": 4.802055576703464e-06,
      "loss": 1.7315,
      "step": 411
    },
    {
      "epoch": 0.42321520287621983,
      "grad_norm": 2.6072895526885986,
      "learning_rate": 4.8001522649409976e-06,
      "loss": 1.8176,
      "step": 412
    },
    {
      "epoch": 0.42424242424242425,
      "grad_norm": 2.9197702407836914,
      "learning_rate": 4.798248953178531e-06,
      "loss": 1.6947,
      "step": 413
    },
    {
      "epoch": 0.4252696456086287,
      "grad_norm": 2.4383556842803955,
      "learning_rate": 4.796345641416064e-06,
      "loss": 1.6728,
      "step": 414
    },
    {
      "epoch": 0.4262968669748331,
      "grad_norm": 2.9951915740966797,
      "learning_rate": 4.794442329653598e-06,
      "loss": 1.7249,
      "step": 415
    },
    {
      "epoch": 0.42732408834103747,
      "grad_norm": 2.6095519065856934,
      "learning_rate": 4.792539017891131e-06,
      "loss": 1.7735,
      "step": 416
    },
    {
      "epoch": 0.4283513097072419,
      "grad_norm": 2.6429390907287598,
      "learning_rate": 4.790635706128664e-06,
      "loss": 1.704,
      "step": 417
    },
    {
      "epoch": 0.4293785310734463,
      "grad_norm": 2.921812057495117,
      "learning_rate": 4.788732394366197e-06,
      "loss": 1.7355,
      "step": 418
    },
    {
      "epoch": 0.43040575243965074,
      "grad_norm": 2.6050965785980225,
      "learning_rate": 4.786829082603731e-06,
      "loss": 1.6636,
      "step": 419
    },
    {
      "epoch": 0.43143297380585516,
      "grad_norm": 2.672070264816284,
      "learning_rate": 4.7849257708412645e-06,
      "loss": 1.6627,
      "step": 420
    },
    {
      "epoch": 0.43143297380585516,
      "eval_loss": 1.9990200996398926,
      "eval_runtime": 45.0672,
      "eval_samples_per_second": 22.189,
      "eval_steps_per_second": 0.932,
      "step": 420
    },
    {
      "epoch": 0.4324601951720596,
      "grad_norm": 2.6813488006591797,
      "learning_rate": 4.783022459078798e-06,
      "loss": 1.7453,
      "step": 421
    },
    {
      "epoch": 0.433487416538264,
      "grad_norm": 2.6587321758270264,
      "learning_rate": 4.781119147316331e-06,
      "loss": 1.6795,
      "step": 422
    },
    {
      "epoch": 0.4345146379044684,
      "grad_norm": 2.827758312225342,
      "learning_rate": 4.7792158355538645e-06,
      "loss": 1.7098,
      "step": 423
    },
    {
      "epoch": 0.43554185927067285,
      "grad_norm": 2.8615901470184326,
      "learning_rate": 4.777312523791398e-06,
      "loss": 1.7171,
      "step": 424
    },
    {
      "epoch": 0.43656908063687727,
      "grad_norm": 2.9509294033050537,
      "learning_rate": 4.77540921202893e-06,
      "loss": 1.7018,
      "step": 425
    },
    {
      "epoch": 0.43759630200308164,
      "grad_norm": 2.7994418144226074,
      "learning_rate": 4.773505900266464e-06,
      "loss": 1.7328,
      "step": 426
    },
    {
      "epoch": 0.43862352336928606,
      "grad_norm": 2.7476799488067627,
      "learning_rate": 4.771602588503997e-06,
      "loss": 1.7116,
      "step": 427
    },
    {
      "epoch": 0.4396507447354905,
      "grad_norm": 2.7955048084259033,
      "learning_rate": 4.7696992767415305e-06,
      "loss": 1.7187,
      "step": 428
    },
    {
      "epoch": 0.4406779661016949,
      "grad_norm": 2.4326071739196777,
      "learning_rate": 4.767795964979064e-06,
      "loss": 1.75,
      "step": 429
    },
    {
      "epoch": 0.44170518746789933,
      "grad_norm": 2.7708163261413574,
      "learning_rate": 4.765892653216597e-06,
      "loss": 1.7562,
      "step": 430
    },
    {
      "epoch": 0.44170518746789933,
      "eval_loss": 2.0262997150421143,
      "eval_runtime": 46.1649,
      "eval_samples_per_second": 21.661,
      "eval_steps_per_second": 0.91,
      "step": 430
    },
    {
      "epoch": 0.44273240883410375,
      "grad_norm": 2.872743606567383,
      "learning_rate": 4.763989341454131e-06,
      "loss": 1.7527,
      "step": 431
    },
    {
      "epoch": 0.4437596302003082,
      "grad_norm": 2.6387035846710205,
      "learning_rate": 4.762086029691664e-06,
      "loss": 1.6742,
      "step": 432
    },
    {
      "epoch": 0.4447868515665126,
      "grad_norm": 2.6522979736328125,
      "learning_rate": 4.760182717929197e-06,
      "loss": 1.6811,
      "step": 433
    },
    {
      "epoch": 0.445814072932717,
      "grad_norm": 2.644155263900757,
      "learning_rate": 4.758279406166731e-06,
      "loss": 1.7083,
      "step": 434
    },
    {
      "epoch": 0.44684129429892144,
      "grad_norm": 2.530224561691284,
      "learning_rate": 4.756376094404263e-06,
      "loss": 1.6318,
      "step": 435
    },
    {
      "epoch": 0.4478685156651258,
      "grad_norm": 2.431049108505249,
      "learning_rate": 4.7544727826417966e-06,
      "loss": 1.7162,
      "step": 436
    },
    {
      "epoch": 0.44889573703133023,
      "grad_norm": 2.8091282844543457,
      "learning_rate": 4.75256947087933e-06,
      "loss": 1.7717,
      "step": 437
    },
    {
      "epoch": 0.44992295839753466,
      "grad_norm": 2.5520501136779785,
      "learning_rate": 4.750666159116863e-06,
      "loss": 1.7201,
      "step": 438
    },
    {
      "epoch": 0.4509501797637391,
      "grad_norm": 2.596370220184326,
      "learning_rate": 4.7487628473543975e-06,
      "loss": 1.6882,
      "step": 439
    },
    {
      "epoch": 0.4519774011299435,
      "grad_norm": 2.667034387588501,
      "learning_rate": 4.746859535591931e-06,
      "loss": 1.6499,
      "step": 440
    },
    {
      "epoch": 0.4519774011299435,
      "eval_loss": 2.008521556854248,
      "eval_runtime": 49.3626,
      "eval_samples_per_second": 20.258,
      "eval_steps_per_second": 0.851,
      "step": 440
    },
    {
      "epoch": 0.4530046224961479,
      "grad_norm": 2.636491537094116,
      "learning_rate": 4.744956223829464e-06,
      "loss": 1.6987,
      "step": 441
    },
    {
      "epoch": 0.45403184386235235,
      "grad_norm": 2.4653050899505615,
      "learning_rate": 4.743052912066997e-06,
      "loss": 1.6775,
      "step": 442
    },
    {
      "epoch": 0.45505906522855677,
      "grad_norm": 2.3866467475891113,
      "learning_rate": 4.74114960030453e-06,
      "loss": 1.7234,
      "step": 443
    },
    {
      "epoch": 0.4560862865947612,
      "grad_norm": 2.675262928009033,
      "learning_rate": 4.7392462885420635e-06,
      "loss": 1.7363,
      "step": 444
    },
    {
      "epoch": 0.4571135079609656,
      "grad_norm": 2.6233296394348145,
      "learning_rate": 4.737342976779597e-06,
      "loss": 1.6236,
      "step": 445
    },
    {
      "epoch": 0.45814072932717,
      "grad_norm": 2.5638058185577393,
      "learning_rate": 4.73543966501713e-06,
      "loss": 1.7097,
      "step": 446
    },
    {
      "epoch": 0.4591679506933744,
      "grad_norm": 2.7206926345825195,
      "learning_rate": 4.7335363532546636e-06,
      "loss": 1.6909,
      "step": 447
    },
    {
      "epoch": 0.4601951720595788,
      "grad_norm": 2.546241283416748,
      "learning_rate": 4.731633041492197e-06,
      "loss": 1.6951,
      "step": 448
    },
    {
      "epoch": 0.46122239342578325,
      "grad_norm": 2.6549105644226074,
      "learning_rate": 4.72972972972973e-06,
      "loss": 1.7052,
      "step": 449
    },
    {
      "epoch": 0.4622496147919877,
      "grad_norm": 2.6649065017700195,
      "learning_rate": 4.727826417967264e-06,
      "loss": 1.7406,
      "step": 450
    },
    {
      "epoch": 0.4622496147919877,
      "eval_loss": 1.9886436462402344,
      "eval_runtime": 47.48,
      "eval_samples_per_second": 21.061,
      "eval_steps_per_second": 0.885,
      "step": 450
    },
    {
      "epoch": 0.4632768361581921,
      "grad_norm": 2.7138187885284424,
      "learning_rate": 4.725923106204797e-06,
      "loss": 1.7154,
      "step": 451
    },
    {
      "epoch": 0.4643040575243965,
      "grad_norm": 2.6095476150512695,
      "learning_rate": 4.72401979444233e-06,
      "loss": 1.686,
      "step": 452
    },
    {
      "epoch": 0.46533127889060094,
      "grad_norm": 2.866363048553467,
      "learning_rate": 4.722116482679863e-06,
      "loss": 1.721,
      "step": 453
    },
    {
      "epoch": 0.46635850025680536,
      "grad_norm": 2.4947352409362793,
      "learning_rate": 4.720213170917396e-06,
      "loss": 1.708,
      "step": 454
    },
    {
      "epoch": 0.46738572162300973,
      "grad_norm": 2.3888256549835205,
      "learning_rate": 4.71830985915493e-06,
      "loss": 1.7356,
      "step": 455
    },
    {
      "epoch": 0.46841294298921415,
      "grad_norm": 2.7065343856811523,
      "learning_rate": 4.716406547392463e-06,
      "loss": 1.6943,
      "step": 456
    },
    {
      "epoch": 0.4694401643554186,
      "grad_norm": 2.424285888671875,
      "learning_rate": 4.714503235629996e-06,
      "loss": 1.7385,
      "step": 457
    },
    {
      "epoch": 0.470467385721623,
      "grad_norm": 2.4175925254821777,
      "learning_rate": 4.71259992386753e-06,
      "loss": 1.6654,
      "step": 458
    },
    {
      "epoch": 0.4714946070878274,
      "grad_norm": 2.3203656673431396,
      "learning_rate": 4.710696612105063e-06,
      "loss": 1.6977,
      "step": 459
    },
    {
      "epoch": 0.47252182845403184,
      "grad_norm": 2.6660172939300537,
      "learning_rate": 4.708793300342596e-06,
      "loss": 1.7055,
      "step": 460
    },
    {
      "epoch": 0.47252182845403184,
      "eval_loss": 1.9864659309387207,
      "eval_runtime": 47.6232,
      "eval_samples_per_second": 20.998,
      "eval_steps_per_second": 0.882,
      "step": 460
    },
    {
      "epoch": 0.47354904982023627,
      "grad_norm": 2.515615701675415,
      "learning_rate": 4.70688998858013e-06,
      "loss": 1.689,
      "step": 461
    },
    {
      "epoch": 0.4745762711864407,
      "grad_norm": 2.429264783859253,
      "learning_rate": 4.704986676817663e-06,
      "loss": 1.6796,
      "step": 462
    },
    {
      "epoch": 0.4756034925526451,
      "grad_norm": 2.469543933868408,
      "learning_rate": 4.7030833650551965e-06,
      "loss": 1.7164,
      "step": 463
    },
    {
      "epoch": 0.47663071391884954,
      "grad_norm": 2.5921683311462402,
      "learning_rate": 4.70118005329273e-06,
      "loss": 1.6746,
      "step": 464
    },
    {
      "epoch": 0.4776579352850539,
      "grad_norm": 2.6612918376922607,
      "learning_rate": 4.699276741530263e-06,
      "loss": 1.7598,
      "step": 465
    },
    {
      "epoch": 0.4786851566512583,
      "grad_norm": 2.4633305072784424,
      "learning_rate": 4.697373429767797e-06,
      "loss": 1.7241,
      "step": 466
    },
    {
      "epoch": 0.47971237801746275,
      "grad_norm": 2.613023281097412,
      "learning_rate": 4.69547011800533e-06,
      "loss": 1.7088,
      "step": 467
    },
    {
      "epoch": 0.48073959938366717,
      "grad_norm": 2.5329010486602783,
      "learning_rate": 4.693566806242863e-06,
      "loss": 1.7163,
      "step": 468
    },
    {
      "epoch": 0.4817668207498716,
      "grad_norm": 2.7887041568756104,
      "learning_rate": 4.691663494480397e-06,
      "loss": 1.6969,
      "step": 469
    },
    {
      "epoch": 0.482794042116076,
      "grad_norm": 2.72991943359375,
      "learning_rate": 4.689760182717929e-06,
      "loss": 1.7332,
      "step": 470
    },
    {
      "epoch": 0.482794042116076,
      "eval_loss": 1.965472936630249,
      "eval_runtime": 46.1059,
      "eval_samples_per_second": 21.689,
      "eval_steps_per_second": 0.911,
      "step": 470
    },
    {
      "epoch": 0.48382126348228044,
      "grad_norm": 2.613785982131958,
      "learning_rate": 4.6878568709554626e-06,
      "loss": 1.6272,
      "step": 471
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 2.605299472808838,
      "learning_rate": 4.685953559192996e-06,
      "loss": 1.6966,
      "step": 472
    },
    {
      "epoch": 0.4858757062146893,
      "grad_norm": 2.672013998031616,
      "learning_rate": 4.684050247430529e-06,
      "loss": 1.7132,
      "step": 473
    },
    {
      "epoch": 0.4869029275808937,
      "grad_norm": 2.576486825942993,
      "learning_rate": 4.682146935668063e-06,
      "loss": 1.6449,
      "step": 474
    },
    {
      "epoch": 0.4879301489470981,
      "grad_norm": 2.626840114593506,
      "learning_rate": 4.680243623905596e-06,
      "loss": 1.6993,
      "step": 475
    },
    {
      "epoch": 0.4889573703133025,
      "grad_norm": 2.974418878555298,
      "learning_rate": 4.678340312143129e-06,
      "loss": 1.7116,
      "step": 476
    },
    {
      "epoch": 0.4899845916795069,
      "grad_norm": 2.541242837905884,
      "learning_rate": 4.676437000380663e-06,
      "loss": 1.6693,
      "step": 477
    },
    {
      "epoch": 0.49101181304571134,
      "grad_norm": 2.5494985580444336,
      "learning_rate": 4.674533688618196e-06,
      "loss": 1.7819,
      "step": 478
    },
    {
      "epoch": 0.49203903441191577,
      "grad_norm": 2.606137990951538,
      "learning_rate": 4.6726303768557295e-06,
      "loss": 1.7278,
      "step": 479
    },
    {
      "epoch": 0.4930662557781202,
      "grad_norm": 2.5032176971435547,
      "learning_rate": 4.670727065093263e-06,
      "loss": 1.6318,
      "step": 480
    },
    {
      "epoch": 0.4930662557781202,
      "eval_loss": 1.9851018190383911,
      "eval_runtime": 46.0386,
      "eval_samples_per_second": 21.721,
      "eval_steps_per_second": 0.912,
      "step": 480
    },
    {
      "epoch": 0.4940934771443246,
      "grad_norm": 2.703259229660034,
      "learning_rate": 4.668823753330796e-06,
      "loss": 1.6646,
      "step": 481
    },
    {
      "epoch": 0.49512069851052903,
      "grad_norm": 2.681807041168213,
      "learning_rate": 4.6669204415683295e-06,
      "loss": 1.6759,
      "step": 482
    },
    {
      "epoch": 0.49614791987673346,
      "grad_norm": 2.7280473709106445,
      "learning_rate": 4.665017129805863e-06,
      "loss": 1.7388,
      "step": 483
    },
    {
      "epoch": 0.4971751412429379,
      "grad_norm": 2.420909881591797,
      "learning_rate": 4.663113818043396e-06,
      "loss": 1.729,
      "step": 484
    },
    {
      "epoch": 0.49820236260914225,
      "grad_norm": 2.5764737129211426,
      "learning_rate": 4.66121050628093e-06,
      "loss": 1.6903,
      "step": 485
    },
    {
      "epoch": 0.49922958397534667,
      "grad_norm": 2.389531135559082,
      "learning_rate": 4.659307194518463e-06,
      "loss": 1.7191,
      "step": 486
    },
    {
      "epoch": 0.5002568053415511,
      "grad_norm": 2.5207972526550293,
      "learning_rate": 4.6574038827559955e-06,
      "loss": 1.6892,
      "step": 487
    },
    {
      "epoch": 0.5012840267077555,
      "grad_norm": 2.5662360191345215,
      "learning_rate": 4.655500570993529e-06,
      "loss": 1.6978,
      "step": 488
    },
    {
      "epoch": 0.50231124807396,
      "grad_norm": 2.50262713432312,
      "learning_rate": 4.653597259231062e-06,
      "loss": 1.7379,
      "step": 489
    },
    {
      "epoch": 0.5033384694401644,
      "grad_norm": 2.56643009185791,
      "learning_rate": 4.651693947468596e-06,
      "loss": 1.6233,
      "step": 490
    },
    {
      "epoch": 0.5033384694401644,
      "eval_loss": 1.9815036058425903,
      "eval_runtime": 48.0257,
      "eval_samples_per_second": 20.822,
      "eval_steps_per_second": 0.875,
      "step": 490
    },
    {
      "epoch": 0.5043656908063687,
      "grad_norm": 2.504117965698242,
      "learning_rate": 4.649790635706129e-06,
      "loss": 1.7664,
      "step": 491
    },
    {
      "epoch": 0.5053929121725732,
      "grad_norm": 2.906843423843384,
      "learning_rate": 4.647887323943662e-06,
      "loss": 1.6731,
      "step": 492
    },
    {
      "epoch": 0.5064201335387776,
      "grad_norm": 2.586747169494629,
      "learning_rate": 4.645984012181196e-06,
      "loss": 1.68,
      "step": 493
    },
    {
      "epoch": 0.507447354904982,
      "grad_norm": 2.3300511837005615,
      "learning_rate": 4.644080700418729e-06,
      "loss": 1.5998,
      "step": 494
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 2.8711049556732178,
      "learning_rate": 4.642177388656262e-06,
      "loss": 1.7198,
      "step": 495
    },
    {
      "epoch": 0.5095017976373909,
      "grad_norm": 2.6463162899017334,
      "learning_rate": 4.640274076893796e-06,
      "loss": 1.6814,
      "step": 496
    },
    {
      "epoch": 0.5105290190035953,
      "grad_norm": 2.650481700897217,
      "learning_rate": 4.638370765131328e-06,
      "loss": 1.7353,
      "step": 497
    },
    {
      "epoch": 0.5115562403697997,
      "grad_norm": 2.4464986324310303,
      "learning_rate": 4.636467453368862e-06,
      "loss": 1.6579,
      "step": 498
    },
    {
      "epoch": 0.5125834617360041,
      "grad_norm": 2.438021659851074,
      "learning_rate": 4.634564141606396e-06,
      "loss": 1.6485,
      "step": 499
    },
    {
      "epoch": 0.5136106831022085,
      "grad_norm": 2.5354347229003906,
      "learning_rate": 4.632660829843929e-06,
      "loss": 1.7123,
      "step": 500
    },
    {
      "epoch": 0.5136106831022085,
      "eval_loss": 1.9738385677337646,
      "eval_runtime": 52.5306,
      "eval_samples_per_second": 19.037,
      "eval_steps_per_second": 0.8,
      "step": 500
    },
    {
      "epoch": 0.514637904468413,
      "grad_norm": 2.3702070713043213,
      "learning_rate": 4.630757518081463e-06,
      "loss": 1.6423,
      "step": 501
    },
    {
      "epoch": 0.5156651258346173,
      "grad_norm": 2.5336198806762695,
      "learning_rate": 4.628854206318996e-06,
      "loss": 1.771,
      "step": 502
    },
    {
      "epoch": 0.5166923472008218,
      "grad_norm": 2.572751998901367,
      "learning_rate": 4.626950894556529e-06,
      "loss": 1.677,
      "step": 503
    },
    {
      "epoch": 0.5177195685670262,
      "grad_norm": 2.3865134716033936,
      "learning_rate": 4.625047582794062e-06,
      "loss": 1.7111,
      "step": 504
    },
    {
      "epoch": 0.5187467899332306,
      "grad_norm": 2.5152785778045654,
      "learning_rate": 4.623144271031595e-06,
      "loss": 1.6343,
      "step": 505
    },
    {
      "epoch": 0.519774011299435,
      "grad_norm": 2.5121307373046875,
      "learning_rate": 4.6212409592691286e-06,
      "loss": 1.6357,
      "step": 506
    },
    {
      "epoch": 0.5208012326656395,
      "grad_norm": 2.506908655166626,
      "learning_rate": 4.619337647506662e-06,
      "loss": 1.7295,
      "step": 507
    },
    {
      "epoch": 0.5218284540318439,
      "grad_norm": 2.636082172393799,
      "learning_rate": 4.617434335744195e-06,
      "loss": 1.6916,
      "step": 508
    },
    {
      "epoch": 0.5228556753980482,
      "grad_norm": 2.5987415313720703,
      "learning_rate": 4.615531023981729e-06,
      "loss": 1.659,
      "step": 509
    },
    {
      "epoch": 0.5238828967642527,
      "grad_norm": 2.642770290374756,
      "learning_rate": 4.613627712219262e-06,
      "loss": 1.6789,
      "step": 510
    },
    {
      "epoch": 0.5238828967642527,
      "eval_loss": 1.9659523963928223,
      "eval_runtime": 42.9453,
      "eval_samples_per_second": 23.285,
      "eval_steps_per_second": 0.978,
      "step": 510
    },
    {
      "epoch": 0.5249101181304571,
      "grad_norm": 2.438389301300049,
      "learning_rate": 4.611724400456795e-06,
      "loss": 1.68,
      "step": 511
    },
    {
      "epoch": 0.5259373394966615,
      "grad_norm": 2.536546468734741,
      "learning_rate": 4.609821088694329e-06,
      "loss": 1.6847,
      "step": 512
    },
    {
      "epoch": 0.5269645608628659,
      "grad_norm": 2.4056830406188965,
      "learning_rate": 4.607917776931862e-06,
      "loss": 1.6577,
      "step": 513
    },
    {
      "epoch": 0.5279917822290704,
      "grad_norm": 2.81195330619812,
      "learning_rate": 4.606014465169395e-06,
      "loss": 1.7149,
      "step": 514
    },
    {
      "epoch": 0.5290190035952748,
      "grad_norm": 2.684948205947876,
      "learning_rate": 4.604111153406928e-06,
      "loss": 1.717,
      "step": 515
    },
    {
      "epoch": 0.5300462249614792,
      "grad_norm": 2.583092212677002,
      "learning_rate": 4.602207841644461e-06,
      "loss": 1.6238,
      "step": 516
    },
    {
      "epoch": 0.5310734463276836,
      "grad_norm": 3.022528886795044,
      "learning_rate": 4.600304529881995e-06,
      "loss": 1.7262,
      "step": 517
    },
    {
      "epoch": 0.5321006676938881,
      "grad_norm": 2.6941866874694824,
      "learning_rate": 4.598401218119528e-06,
      "loss": 1.7055,
      "step": 518
    },
    {
      "epoch": 0.5331278890600925,
      "grad_norm": 2.5207839012145996,
      "learning_rate": 4.5964979063570614e-06,
      "loss": 1.7117,
      "step": 519
    },
    {
      "epoch": 0.5341551104262968,
      "grad_norm": 2.8056678771972656,
      "learning_rate": 4.594594594594596e-06,
      "loss": 1.657,
      "step": 520
    },
    {
      "epoch": 0.5341551104262968,
      "eval_loss": 1.9712116718292236,
      "eval_runtime": 49.57,
      "eval_samples_per_second": 20.173,
      "eval_steps_per_second": 0.847,
      "step": 520
    },
    {
      "epoch": 0.5351823317925013,
      "grad_norm": 2.6119930744171143,
      "learning_rate": 4.592691282832128e-06,
      "loss": 1.6897,
      "step": 521
    },
    {
      "epoch": 0.5362095531587057,
      "grad_norm": 2.7165610790252686,
      "learning_rate": 4.5907879710696615e-06,
      "loss": 1.7038,
      "step": 522
    },
    {
      "epoch": 0.5372367745249101,
      "grad_norm": 2.3613736629486084,
      "learning_rate": 4.588884659307195e-06,
      "loss": 1.6222,
      "step": 523
    },
    {
      "epoch": 0.5382639958911145,
      "grad_norm": 2.647672176361084,
      "learning_rate": 4.586981347544728e-06,
      "loss": 1.702,
      "step": 524
    },
    {
      "epoch": 0.539291217257319,
      "grad_norm": 2.704634428024292,
      "learning_rate": 4.585078035782262e-06,
      "loss": 1.6566,
      "step": 525
    },
    {
      "epoch": 0.5403184386235234,
      "grad_norm": 2.6634597778320312,
      "learning_rate": 4.583174724019795e-06,
      "loss": 1.6903,
      "step": 526
    },
    {
      "epoch": 0.5413456599897278,
      "grad_norm": 2.5745162963867188,
      "learning_rate": 4.581271412257328e-06,
      "loss": 1.6375,
      "step": 527
    },
    {
      "epoch": 0.5423728813559322,
      "grad_norm": 2.5577778816223145,
      "learning_rate": 4.579368100494862e-06,
      "loss": 1.614,
      "step": 528
    },
    {
      "epoch": 0.5434001027221366,
      "grad_norm": 2.7684454917907715,
      "learning_rate": 4.577464788732395e-06,
      "loss": 1.7168,
      "step": 529
    },
    {
      "epoch": 0.544427324088341,
      "grad_norm": 2.582108736038208,
      "learning_rate": 4.575561476969928e-06,
      "loss": 1.6897,
      "step": 530
    },
    {
      "epoch": 0.544427324088341,
      "eval_loss": 1.9821091890335083,
      "eval_runtime": 50.9409,
      "eval_samples_per_second": 19.631,
      "eval_steps_per_second": 0.824,
      "step": 530
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 2.516486406326294,
      "learning_rate": 4.573658165207461e-06,
      "loss": 1.7273,
      "step": 531
    },
    {
      "epoch": 0.5464817668207499,
      "grad_norm": 2.786806344985962,
      "learning_rate": 4.571754853444994e-06,
      "loss": 1.6898,
      "step": 532
    },
    {
      "epoch": 0.5475089881869543,
      "grad_norm": 2.3137319087982178,
      "learning_rate": 4.569851541682528e-06,
      "loss": 1.676,
      "step": 533
    },
    {
      "epoch": 0.5485362095531587,
      "grad_norm": 2.7268903255462646,
      "learning_rate": 4.567948229920061e-06,
      "loss": 1.6483,
      "step": 534
    },
    {
      "epoch": 0.5495634309193631,
      "grad_norm": 2.4385902881622314,
      "learning_rate": 4.566044918157594e-06,
      "loss": 1.6172,
      "step": 535
    },
    {
      "epoch": 0.5505906522855676,
      "grad_norm": 2.420576333999634,
      "learning_rate": 4.564141606395128e-06,
      "loss": 1.6569,
      "step": 536
    },
    {
      "epoch": 0.551617873651772,
      "grad_norm": 2.592290163040161,
      "learning_rate": 4.562238294632661e-06,
      "loss": 1.7085,
      "step": 537
    },
    {
      "epoch": 0.5526450950179764,
      "grad_norm": 2.485285520553589,
      "learning_rate": 4.5603349828701945e-06,
      "loss": 1.6699,
      "step": 538
    },
    {
      "epoch": 0.5536723163841808,
      "grad_norm": 2.844710111618042,
      "learning_rate": 4.558431671107728e-06,
      "loss": 1.6873,
      "step": 539
    },
    {
      "epoch": 0.5546995377503852,
      "grad_norm": 2.5279181003570557,
      "learning_rate": 4.556528359345261e-06,
      "loss": 1.7285,
      "step": 540
    },
    {
      "epoch": 0.5546995377503852,
      "eval_loss": 1.9825981855392456,
      "eval_runtime": 52.461,
      "eval_samples_per_second": 19.062,
      "eval_steps_per_second": 0.801,
      "step": 540
    },
    {
      "epoch": 0.5557267591165896,
      "grad_norm": 2.673997163772583,
      "learning_rate": 4.5546250475827946e-06,
      "loss": 1.6658,
      "step": 541
    },
    {
      "epoch": 0.556753980482794,
      "grad_norm": 2.6023194789886475,
      "learning_rate": 4.552721735820328e-06,
      "loss": 1.6262,
      "step": 542
    },
    {
      "epoch": 0.5577812018489985,
      "grad_norm": 2.6327662467956543,
      "learning_rate": 4.550818424057861e-06,
      "loss": 1.6574,
      "step": 543
    },
    {
      "epoch": 0.5588084232152029,
      "grad_norm": 2.5532140731811523,
      "learning_rate": 4.548915112295395e-06,
      "loss": 1.6607,
      "step": 544
    },
    {
      "epoch": 0.5598356445814073,
      "grad_norm": 2.523966073989868,
      "learning_rate": 4.547011800532928e-06,
      "loss": 1.6752,
      "step": 545
    },
    {
      "epoch": 0.5608628659476117,
      "grad_norm": 2.6338107585906982,
      "learning_rate": 4.545108488770461e-06,
      "loss": 1.6258,
      "step": 546
    },
    {
      "epoch": 0.5618900873138162,
      "grad_norm": 2.500096559524536,
      "learning_rate": 4.543205177007995e-06,
      "loss": 1.7189,
      "step": 547
    },
    {
      "epoch": 0.5629173086800205,
      "grad_norm": 2.644413471221924,
      "learning_rate": 4.541301865245527e-06,
      "loss": 1.6594,
      "step": 548
    },
    {
      "epoch": 0.5639445300462249,
      "grad_norm": 2.451965093612671,
      "learning_rate": 4.539398553483061e-06,
      "loss": 1.6984,
      "step": 549
    },
    {
      "epoch": 0.5649717514124294,
      "grad_norm": 2.7554423809051514,
      "learning_rate": 4.537495241720594e-06,
      "loss": 1.6626,
      "step": 550
    },
    {
      "epoch": 0.5649717514124294,
      "eval_loss": 1.9897041320800781,
      "eval_runtime": 45.9349,
      "eval_samples_per_second": 21.77,
      "eval_steps_per_second": 0.914,
      "step": 550
    },
    {
      "epoch": 0.5659989727786338,
      "grad_norm": 2.5962674617767334,
      "learning_rate": 4.535591929958127e-06,
      "loss": 1.7015,
      "step": 551
    },
    {
      "epoch": 0.5670261941448382,
      "grad_norm": 2.7917397022247314,
      "learning_rate": 4.533688618195661e-06,
      "loss": 1.6811,
      "step": 552
    },
    {
      "epoch": 0.5680534155110426,
      "grad_norm": 2.4205498695373535,
      "learning_rate": 4.531785306433194e-06,
      "loss": 1.6659,
      "step": 553
    },
    {
      "epoch": 0.5690806368772471,
      "grad_norm": 2.74898624420166,
      "learning_rate": 4.529881994670727e-06,
      "loss": 1.6521,
      "step": 554
    },
    {
      "epoch": 0.5701078582434514,
      "grad_norm": 2.859898328781128,
      "learning_rate": 4.527978682908261e-06,
      "loss": 1.6445,
      "step": 555
    },
    {
      "epoch": 0.5711350796096559,
      "grad_norm": 2.5526154041290283,
      "learning_rate": 4.526075371145794e-06,
      "loss": 1.6679,
      "step": 556
    },
    {
      "epoch": 0.5721623009758603,
      "grad_norm": 2.96175217628479,
      "learning_rate": 4.5241720593833275e-06,
      "loss": 1.6784,
      "step": 557
    },
    {
      "epoch": 0.5731895223420647,
      "grad_norm": 2.5037007331848145,
      "learning_rate": 4.52226874762086e-06,
      "loss": 1.6689,
      "step": 558
    },
    {
      "epoch": 0.5742167437082691,
      "grad_norm": 2.6513259410858154,
      "learning_rate": 4.520365435858393e-06,
      "loss": 1.7038,
      "step": 559
    },
    {
      "epoch": 0.5752439650744735,
      "grad_norm": 2.762770175933838,
      "learning_rate": 4.518462124095928e-06,
      "loss": 1.728,
      "step": 560
    },
    {
      "epoch": 0.5752439650744735,
      "eval_loss": 1.972934603691101,
      "eval_runtime": 44.2572,
      "eval_samples_per_second": 22.595,
      "eval_steps_per_second": 0.949,
      "step": 560
    },
    {
      "epoch": 0.576271186440678,
      "grad_norm": 2.637406349182129,
      "learning_rate": 4.516558812333461e-06,
      "loss": 1.682,
      "step": 561
    },
    {
      "epoch": 0.5772984078068824,
      "grad_norm": 2.545583724975586,
      "learning_rate": 4.514655500570994e-06,
      "loss": 1.6669,
      "step": 562
    },
    {
      "epoch": 0.5783256291730868,
      "grad_norm": 2.5655033588409424,
      "learning_rate": 4.512752188808528e-06,
      "loss": 1.7279,
      "step": 563
    },
    {
      "epoch": 0.5793528505392912,
      "grad_norm": 2.6856470108032227,
      "learning_rate": 4.510848877046061e-06,
      "loss": 1.7331,
      "step": 564
    },
    {
      "epoch": 0.5803800719054957,
      "grad_norm": 2.605252981185913,
      "learning_rate": 4.5089455652835936e-06,
      "loss": 1.6904,
      "step": 565
    },
    {
      "epoch": 0.5814072932717,
      "grad_norm": 2.610267400741577,
      "learning_rate": 4.507042253521127e-06,
      "loss": 1.6607,
      "step": 566
    },
    {
      "epoch": 0.5824345146379045,
      "grad_norm": 2.6955432891845703,
      "learning_rate": 4.50513894175866e-06,
      "loss": 1.6166,
      "step": 567
    },
    {
      "epoch": 0.5834617360041089,
      "grad_norm": 2.5661842823028564,
      "learning_rate": 4.503235629996194e-06,
      "loss": 1.6955,
      "step": 568
    },
    {
      "epoch": 0.5844889573703133,
      "grad_norm": 2.6393115520477295,
      "learning_rate": 4.501332318233727e-06,
      "loss": 1.6933,
      "step": 569
    },
    {
      "epoch": 0.5855161787365177,
      "grad_norm": 2.612645387649536,
      "learning_rate": 4.49942900647126e-06,
      "loss": 1.6585,
      "step": 570
    },
    {
      "epoch": 0.5855161787365177,
      "eval_loss": 1.9599064588546753,
      "eval_runtime": 46.0676,
      "eval_samples_per_second": 21.707,
      "eval_steps_per_second": 0.912,
      "step": 570
    },
    {
      "epoch": 0.5865434001027221,
      "grad_norm": 2.4182891845703125,
      "learning_rate": 4.497525694708794e-06,
      "loss": 1.707,
      "step": 571
    },
    {
      "epoch": 0.5875706214689266,
      "grad_norm": 2.7214412689208984,
      "learning_rate": 4.495622382946327e-06,
      "loss": 1.6966,
      "step": 572
    },
    {
      "epoch": 0.588597842835131,
      "grad_norm": 2.5138018131256104,
      "learning_rate": 4.4937190711838605e-06,
      "loss": 1.6373,
      "step": 573
    },
    {
      "epoch": 0.5896250642013354,
      "grad_norm": 2.5814292430877686,
      "learning_rate": 4.491815759421394e-06,
      "loss": 1.7313,
      "step": 574
    },
    {
      "epoch": 0.5906522855675398,
      "grad_norm": 2.6236367225646973,
      "learning_rate": 4.489912447658926e-06,
      "loss": 1.6647,
      "step": 575
    },
    {
      "epoch": 0.5916795069337443,
      "grad_norm": 2.468557357788086,
      "learning_rate": 4.48800913589646e-06,
      "loss": 1.6412,
      "step": 576
    },
    {
      "epoch": 0.5927067282999486,
      "grad_norm": 2.454141855239868,
      "learning_rate": 4.486105824133993e-06,
      "loss": 1.6762,
      "step": 577
    },
    {
      "epoch": 0.593733949666153,
      "grad_norm": 3.0009915828704834,
      "learning_rate": 4.4842025123715264e-06,
      "loss": 1.6847,
      "step": 578
    },
    {
      "epoch": 0.5947611710323575,
      "grad_norm": 2.574394702911377,
      "learning_rate": 4.48229920060906e-06,
      "loss": 1.6421,
      "step": 579
    },
    {
      "epoch": 0.5957883923985619,
      "grad_norm": 2.6978182792663574,
      "learning_rate": 4.480395888846594e-06,
      "loss": 1.6198,
      "step": 580
    },
    {
      "epoch": 0.5957883923985619,
      "eval_loss": 1.9570144414901733,
      "eval_runtime": 44.7661,
      "eval_samples_per_second": 22.338,
      "eval_steps_per_second": 0.938,
      "step": 580
    },
    {
      "epoch": 0.5968156137647663,
      "grad_norm": 2.652461051940918,
      "learning_rate": 4.478492577084127e-06,
      "loss": 1.7334,
      "step": 581
    },
    {
      "epoch": 0.5978428351309707,
      "grad_norm": 2.6292548179626465,
      "learning_rate": 4.47658926532166e-06,
      "loss": 1.6242,
      "step": 582
    },
    {
      "epoch": 0.5988700564971752,
      "grad_norm": 2.474919080734253,
      "learning_rate": 4.474685953559193e-06,
      "loss": 1.6732,
      "step": 583
    },
    {
      "epoch": 0.5998972778633795,
      "grad_norm": 2.7282516956329346,
      "learning_rate": 4.472782641796727e-06,
      "loss": 1.802,
      "step": 584
    },
    {
      "epoch": 0.600924499229584,
      "grad_norm": 2.770307779312134,
      "learning_rate": 4.47087933003426e-06,
      "loss": 1.6156,
      "step": 585
    },
    {
      "epoch": 0.6019517205957884,
      "grad_norm": 2.570681095123291,
      "learning_rate": 4.468976018271793e-06,
      "loss": 1.6309,
      "step": 586
    },
    {
      "epoch": 0.6029789419619929,
      "grad_norm": 2.4075732231140137,
      "learning_rate": 4.467072706509327e-06,
      "loss": 1.6436,
      "step": 587
    },
    {
      "epoch": 0.6040061633281972,
      "grad_norm": 2.4570937156677246,
      "learning_rate": 4.46516939474686e-06,
      "loss": 1.6737,
      "step": 588
    },
    {
      "epoch": 0.6050333846944016,
      "grad_norm": 2.407274007797241,
      "learning_rate": 4.463266082984393e-06,
      "loss": 1.6179,
      "step": 589
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.745486259460449,
      "learning_rate": 4.461362771221927e-06,
      "loss": 1.7458,
      "step": 590
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 1.9700438976287842,
      "eval_runtime": 45.3018,
      "eval_samples_per_second": 22.074,
      "eval_steps_per_second": 0.927,
      "step": 590
    },
    {
      "epoch": 0.6070878274268104,
      "grad_norm": 2.4258809089660645,
      "learning_rate": 4.45945945945946e-06,
      "loss": 1.6372,
      "step": 591
    },
    {
      "epoch": 0.6081150487930149,
      "grad_norm": 2.683662176132202,
      "learning_rate": 4.457556147696993e-06,
      "loss": 1.6649,
      "step": 592
    },
    {
      "epoch": 0.6091422701592193,
      "grad_norm": 2.879727602005005,
      "learning_rate": 4.455652835934526e-06,
      "loss": 1.7376,
      "step": 593
    },
    {
      "epoch": 0.6101694915254238,
      "grad_norm": 2.5620689392089844,
      "learning_rate": 4.453749524172059e-06,
      "loss": 1.7203,
      "step": 594
    },
    {
      "epoch": 0.6111967128916281,
      "grad_norm": 2.662031888961792,
      "learning_rate": 4.451846212409593e-06,
      "loss": 1.6835,
      "step": 595
    },
    {
      "epoch": 0.6122239342578326,
      "grad_norm": 2.6269922256469727,
      "learning_rate": 4.449942900647126e-06,
      "loss": 1.6726,
      "step": 596
    },
    {
      "epoch": 0.613251155624037,
      "grad_norm": 2.515852212905884,
      "learning_rate": 4.4480395888846595e-06,
      "loss": 1.6045,
      "step": 597
    },
    {
      "epoch": 0.6142783769902413,
      "grad_norm": 3.006873607635498,
      "learning_rate": 4.446136277122193e-06,
      "loss": 1.6827,
      "step": 598
    },
    {
      "epoch": 0.6153055983564458,
      "grad_norm": 2.2904858589172363,
      "learning_rate": 4.444232965359726e-06,
      "loss": 1.6123,
      "step": 599
    },
    {
      "epoch": 0.6163328197226502,
      "grad_norm": 2.7345471382141113,
      "learning_rate": 4.4423296535972596e-06,
      "loss": 1.6697,
      "step": 600
    },
    {
      "epoch": 0.6163328197226502,
      "eval_loss": 1.9713997840881348,
      "eval_runtime": 45.6715,
      "eval_samples_per_second": 21.895,
      "eval_steps_per_second": 0.92,
      "step": 600
    },
    {
      "epoch": 0.6173600410888547,
      "grad_norm": 2.955853223800659,
      "learning_rate": 4.440426341834793e-06,
      "loss": 1.6195,
      "step": 601
    },
    {
      "epoch": 0.618387262455059,
      "grad_norm": 2.4079766273498535,
      "learning_rate": 4.438523030072326e-06,
      "loss": 1.6589,
      "step": 602
    },
    {
      "epoch": 0.6194144838212635,
      "grad_norm": 2.8158490657806396,
      "learning_rate": 4.43661971830986e-06,
      "loss": 1.6895,
      "step": 603
    },
    {
      "epoch": 0.6204417051874679,
      "grad_norm": 2.491471529006958,
      "learning_rate": 4.434716406547393e-06,
      "loss": 1.6351,
      "step": 604
    },
    {
      "epoch": 0.6214689265536724,
      "grad_norm": 2.4093260765075684,
      "learning_rate": 4.432813094784926e-06,
      "loss": 1.6402,
      "step": 605
    },
    {
      "epoch": 0.6224961479198767,
      "grad_norm": 2.751018762588501,
      "learning_rate": 4.43090978302246e-06,
      "loss": 1.6962,
      "step": 606
    },
    {
      "epoch": 0.6235233692860811,
      "grad_norm": 2.4410409927368164,
      "learning_rate": 4.429006471259993e-06,
      "loss": 1.6549,
      "step": 607
    },
    {
      "epoch": 0.6245505906522856,
      "grad_norm": 2.360455274581909,
      "learning_rate": 4.4271031594975265e-06,
      "loss": 1.6449,
      "step": 608
    },
    {
      "epoch": 0.6255778120184899,
      "grad_norm": 2.5058023929595947,
      "learning_rate": 4.425199847735059e-06,
      "loss": 1.617,
      "step": 609
    },
    {
      "epoch": 0.6266050333846944,
      "grad_norm": 2.3429927825927734,
      "learning_rate": 4.423296535972592e-06,
      "loss": 1.6262,
      "step": 610
    },
    {
      "epoch": 0.6266050333846944,
      "eval_loss": 1.9597365856170654,
      "eval_runtime": 42.2034,
      "eval_samples_per_second": 23.695,
      "eval_steps_per_second": 0.995,
      "step": 610
    },
    {
      "epoch": 0.6276322547508988,
      "grad_norm": 2.401761054992676,
      "learning_rate": 4.421393224210126e-06,
      "loss": 1.6497,
      "step": 611
    },
    {
      "epoch": 0.6286594761171033,
      "grad_norm": 2.433439254760742,
      "learning_rate": 4.419489912447659e-06,
      "loss": 1.6117,
      "step": 612
    },
    {
      "epoch": 0.6296866974833076,
      "grad_norm": 2.5147452354431152,
      "learning_rate": 4.4175866006851924e-06,
      "loss": 1.7041,
      "step": 613
    },
    {
      "epoch": 0.6307139188495121,
      "grad_norm": 2.4174447059631348,
      "learning_rate": 4.415683288922726e-06,
      "loss": 1.6652,
      "step": 614
    },
    {
      "epoch": 0.6317411402157165,
      "grad_norm": 2.424800157546997,
      "learning_rate": 4.413779977160259e-06,
      "loss": 1.6407,
      "step": 615
    },
    {
      "epoch": 0.632768361581921,
      "grad_norm": 2.6760315895080566,
      "learning_rate": 4.4118766653977925e-06,
      "loss": 1.6566,
      "step": 616
    },
    {
      "epoch": 0.6337955829481253,
      "grad_norm": 2.694810628890991,
      "learning_rate": 4.409973353635326e-06,
      "loss": 1.7125,
      "step": 617
    },
    {
      "epoch": 0.6348228043143297,
      "grad_norm": 2.475126266479492,
      "learning_rate": 4.408070041872859e-06,
      "loss": 1.6663,
      "step": 618
    },
    {
      "epoch": 0.6358500256805342,
      "grad_norm": 2.734738349914551,
      "learning_rate": 4.406166730110393e-06,
      "loss": 1.6115,
      "step": 619
    },
    {
      "epoch": 0.6368772470467385,
      "grad_norm": 2.4914753437042236,
      "learning_rate": 4.404263418347925e-06,
      "loss": 1.6096,
      "step": 620
    },
    {
      "epoch": 0.6368772470467385,
      "eval_loss": 1.952081322669983,
      "eval_runtime": 45.4451,
      "eval_samples_per_second": 22.005,
      "eval_steps_per_second": 0.924,
      "step": 620
    },
    {
      "epoch": 0.637904468412943,
      "grad_norm": 2.3727800846099854,
      "learning_rate": 4.402360106585459e-06,
      "loss": 1.6275,
      "step": 621
    },
    {
      "epoch": 0.6389316897791474,
      "grad_norm": 2.576402187347412,
      "learning_rate": 4.400456794822993e-06,
      "loss": 1.6252,
      "step": 622
    },
    {
      "epoch": 0.6399589111453519,
      "grad_norm": 2.560185432434082,
      "learning_rate": 4.398553483060526e-06,
      "loss": 1.6522,
      "step": 623
    },
    {
      "epoch": 0.6409861325115562,
      "grad_norm": 2.462785005569458,
      "learning_rate": 4.396650171298059e-06,
      "loss": 1.6124,
      "step": 624
    },
    {
      "epoch": 0.6420133538777607,
      "grad_norm": 2.3942930698394775,
      "learning_rate": 4.394746859535593e-06,
      "loss": 1.6231,
      "step": 625
    },
    {
      "epoch": 0.6430405752439651,
      "grad_norm": 2.4870522022247314,
      "learning_rate": 4.392843547773125e-06,
      "loss": 1.6086,
      "step": 626
    },
    {
      "epoch": 0.6440677966101694,
      "grad_norm": 2.5702786445617676,
      "learning_rate": 4.390940236010659e-06,
      "loss": 1.6228,
      "step": 627
    },
    {
      "epoch": 0.6450950179763739,
      "grad_norm": 2.743983268737793,
      "learning_rate": 4.389036924248192e-06,
      "loss": 1.6394,
      "step": 628
    },
    {
      "epoch": 0.6461222393425783,
      "grad_norm": 2.59942364692688,
      "learning_rate": 4.387133612485725e-06,
      "loss": 1.6721,
      "step": 629
    },
    {
      "epoch": 0.6471494607087828,
      "grad_norm": 2.4473204612731934,
      "learning_rate": 4.385230300723259e-06,
      "loss": 1.6407,
      "step": 630
    },
    {
      "epoch": 0.6471494607087828,
      "eval_loss": 1.9426974058151245,
      "eval_runtime": 45.2318,
      "eval_samples_per_second": 22.108,
      "eval_steps_per_second": 0.929,
      "step": 630
    },
    {
      "epoch": 0.6481766820749871,
      "grad_norm": 2.6036875247955322,
      "learning_rate": 4.383326988960792e-06,
      "loss": 1.6307,
      "step": 631
    },
    {
      "epoch": 0.6492039034411916,
      "grad_norm": 2.691011428833008,
      "learning_rate": 4.3814236771983255e-06,
      "loss": 1.6059,
      "step": 632
    },
    {
      "epoch": 0.650231124807396,
      "grad_norm": 2.585758924484253,
      "learning_rate": 4.379520365435859e-06,
      "loss": 1.6781,
      "step": 633
    },
    {
      "epoch": 0.6512583461736005,
      "grad_norm": 2.9522950649261475,
      "learning_rate": 4.377617053673392e-06,
      "loss": 1.6559,
      "step": 634
    },
    {
      "epoch": 0.6522855675398048,
      "grad_norm": 2.5368709564208984,
      "learning_rate": 4.3757137419109256e-06,
      "loss": 1.6445,
      "step": 635
    },
    {
      "epoch": 0.6533127889060092,
      "grad_norm": 2.5002150535583496,
      "learning_rate": 4.373810430148459e-06,
      "loss": 1.7184,
      "step": 636
    },
    {
      "epoch": 0.6543400102722137,
      "grad_norm": 2.7623229026794434,
      "learning_rate": 4.3719071183859914e-06,
      "loss": 1.6195,
      "step": 637
    },
    {
      "epoch": 0.655367231638418,
      "grad_norm": 2.6188151836395264,
      "learning_rate": 4.370003806623525e-06,
      "loss": 1.6686,
      "step": 638
    },
    {
      "epoch": 0.6563944530046225,
      "grad_norm": 2.685776948928833,
      "learning_rate": 4.368100494861058e-06,
      "loss": 1.6134,
      "step": 639
    },
    {
      "epoch": 0.6574216743708269,
      "grad_norm": 2.5703907012939453,
      "learning_rate": 4.3661971830985915e-06,
      "loss": 1.665,
      "step": 640
    },
    {
      "epoch": 0.6574216743708269,
      "eval_loss": 1.9504806995391846,
      "eval_runtime": 47.1947,
      "eval_samples_per_second": 21.189,
      "eval_steps_per_second": 0.89,
      "step": 640
    },
    {
      "epoch": 0.6584488957370314,
      "grad_norm": 2.511617660522461,
      "learning_rate": 4.364293871336126e-06,
      "loss": 1.6201,
      "step": 641
    },
    {
      "epoch": 0.6594761171032357,
      "grad_norm": 2.549302339553833,
      "learning_rate": 4.362390559573659e-06,
      "loss": 1.647,
      "step": 642
    },
    {
      "epoch": 0.6605033384694402,
      "grad_norm": 2.348747730255127,
      "learning_rate": 4.3604872478111925e-06,
      "loss": 1.5993,
      "step": 643
    },
    {
      "epoch": 0.6615305598356446,
      "grad_norm": 2.7393345832824707,
      "learning_rate": 4.358583936048725e-06,
      "loss": 1.543,
      "step": 644
    },
    {
      "epoch": 0.662557781201849,
      "grad_norm": 2.4242193698883057,
      "learning_rate": 4.356680624286258e-06,
      "loss": 1.6726,
      "step": 645
    },
    {
      "epoch": 0.6635850025680534,
      "grad_norm": 2.6446714401245117,
      "learning_rate": 4.354777312523792e-06,
      "loss": 1.6312,
      "step": 646
    },
    {
      "epoch": 0.6646122239342578,
      "grad_norm": 2.7407124042510986,
      "learning_rate": 4.352874000761325e-06,
      "loss": 1.6868,
      "step": 647
    },
    {
      "epoch": 0.6656394453004623,
      "grad_norm": 2.3706371784210205,
      "learning_rate": 4.3509706889988584e-06,
      "loss": 1.6301,
      "step": 648
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.5197198390960693,
      "learning_rate": 4.349067377236392e-06,
      "loss": 1.702,
      "step": 649
    },
    {
      "epoch": 0.6676938880328711,
      "grad_norm": 2.4792377948760986,
      "learning_rate": 4.347164065473925e-06,
      "loss": 1.6891,
      "step": 650
    },
    {
      "epoch": 0.6676938880328711,
      "eval_loss": 1.9589039087295532,
      "eval_runtime": 47.1512,
      "eval_samples_per_second": 21.208,
      "eval_steps_per_second": 0.891,
      "step": 650
    },
    {
      "epoch": 0.6687211093990755,
      "grad_norm": 2.4003701210021973,
      "learning_rate": 4.3452607537114585e-06,
      "loss": 1.7015,
      "step": 651
    },
    {
      "epoch": 0.66974833076528,
      "grad_norm": 2.5363786220550537,
      "learning_rate": 4.343357441948992e-06,
      "loss": 1.6133,
      "step": 652
    },
    {
      "epoch": 0.6707755521314843,
      "grad_norm": 2.5183844566345215,
      "learning_rate": 4.341454130186525e-06,
      "loss": 1.6574,
      "step": 653
    },
    {
      "epoch": 0.6718027734976888,
      "grad_norm": 2.4668304920196533,
      "learning_rate": 4.339550818424058e-06,
      "loss": 1.6253,
      "step": 654
    },
    {
      "epoch": 0.6728299948638932,
      "grad_norm": 2.4829814434051514,
      "learning_rate": 4.337647506661591e-06,
      "loss": 1.6099,
      "step": 655
    },
    {
      "epoch": 0.6738572162300975,
      "grad_norm": 2.3683176040649414,
      "learning_rate": 4.3357441948991245e-06,
      "loss": 1.6385,
      "step": 656
    },
    {
      "epoch": 0.674884437596302,
      "grad_norm": 2.4906842708587646,
      "learning_rate": 4.333840883136658e-06,
      "loss": 1.6755,
      "step": 657
    },
    {
      "epoch": 0.6759116589625064,
      "grad_norm": 2.558776617050171,
      "learning_rate": 4.331937571374191e-06,
      "loss": 1.6152,
      "step": 658
    },
    {
      "epoch": 0.6769388803287109,
      "grad_norm": 2.4081523418426514,
      "learning_rate": 4.3300342596117246e-06,
      "loss": 1.6224,
      "step": 659
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 2.488370180130005,
      "learning_rate": 4.328130947849258e-06,
      "loss": 1.6298,
      "step": 660
    },
    {
      "epoch": 0.6779661016949152,
      "eval_loss": 1.9484621286392212,
      "eval_runtime": 52.549,
      "eval_samples_per_second": 19.03,
      "eval_steps_per_second": 0.799,
      "step": 660
    },
    {
      "epoch": 0.6789933230611197,
      "grad_norm": 2.495748281478882,
      "learning_rate": 4.326227636086791e-06,
      "loss": 1.5852,
      "step": 661
    },
    {
      "epoch": 0.6800205444273241,
      "grad_norm": 2.525031566619873,
      "learning_rate": 4.324324324324325e-06,
      "loss": 1.6348,
      "step": 662
    },
    {
      "epoch": 0.6810477657935285,
      "grad_norm": 2.427615165710449,
      "learning_rate": 4.322421012561858e-06,
      "loss": 1.6352,
      "step": 663
    },
    {
      "epoch": 0.6820749871597329,
      "grad_norm": 2.5084927082061768,
      "learning_rate": 4.320517700799391e-06,
      "loss": 1.6334,
      "step": 664
    },
    {
      "epoch": 0.6831022085259374,
      "grad_norm": 2.6018898487091064,
      "learning_rate": 4.318614389036925e-06,
      "loss": 1.7212,
      "step": 665
    },
    {
      "epoch": 0.6841294298921418,
      "grad_norm": 2.44184947013855,
      "learning_rate": 4.316711077274458e-06,
      "loss": 1.6802,
      "step": 666
    },
    {
      "epoch": 0.6851566512583461,
      "grad_norm": 2.542123317718506,
      "learning_rate": 4.3148077655119915e-06,
      "loss": 1.5912,
      "step": 667
    },
    {
      "epoch": 0.6861838726245506,
      "grad_norm": 2.5208137035369873,
      "learning_rate": 4.312904453749525e-06,
      "loss": 1.6965,
      "step": 668
    },
    {
      "epoch": 0.687211093990755,
      "grad_norm": 2.391904592514038,
      "learning_rate": 4.311001141987058e-06,
      "loss": 1.6388,
      "step": 669
    },
    {
      "epoch": 0.6882383153569595,
      "grad_norm": 2.629565477371216,
      "learning_rate": 4.3090978302245916e-06,
      "loss": 1.6475,
      "step": 670
    },
    {
      "epoch": 0.6882383153569595,
      "eval_loss": 1.9572818279266357,
      "eval_runtime": 48.5354,
      "eval_samples_per_second": 20.604,
      "eval_steps_per_second": 0.865,
      "step": 670
    },
    {
      "epoch": 0.6892655367231638,
      "grad_norm": 2.3132376670837402,
      "learning_rate": 4.307194518462124e-06,
      "loss": 1.6399,
      "step": 671
    },
    {
      "epoch": 0.6902927580893683,
      "grad_norm": 2.5003411769866943,
      "learning_rate": 4.3052912066996574e-06,
      "loss": 1.7136,
      "step": 672
    },
    {
      "epoch": 0.6913199794555727,
      "grad_norm": 2.545541286468506,
      "learning_rate": 4.303387894937191e-06,
      "loss": 1.6825,
      "step": 673
    },
    {
      "epoch": 0.6923472008217771,
      "grad_norm": 2.342423439025879,
      "learning_rate": 4.301484583174724e-06,
      "loss": 1.6088,
      "step": 674
    },
    {
      "epoch": 0.6933744221879815,
      "grad_norm": 2.5516135692596436,
      "learning_rate": 4.2995812714122575e-06,
      "loss": 1.5934,
      "step": 675
    },
    {
      "epoch": 0.6944016435541859,
      "grad_norm": 2.438675880432129,
      "learning_rate": 4.297677959649791e-06,
      "loss": 1.6236,
      "step": 676
    },
    {
      "epoch": 0.6954288649203904,
      "grad_norm": 2.767305612564087,
      "learning_rate": 4.295774647887324e-06,
      "loss": 1.6514,
      "step": 677
    },
    {
      "epoch": 0.6964560862865947,
      "grad_norm": 2.591994047164917,
      "learning_rate": 4.293871336124858e-06,
      "loss": 1.6643,
      "step": 678
    },
    {
      "epoch": 0.6974833076527992,
      "grad_norm": 2.4734244346618652,
      "learning_rate": 4.291968024362391e-06,
      "loss": 1.6362,
      "step": 679
    },
    {
      "epoch": 0.6985105290190036,
      "grad_norm": 2.6181890964508057,
      "learning_rate": 4.290064712599924e-06,
      "loss": 1.6666,
      "step": 680
    },
    {
      "epoch": 0.6985105290190036,
      "eval_loss": 1.9585388898849487,
      "eval_runtime": 46.8094,
      "eval_samples_per_second": 21.363,
      "eval_steps_per_second": 0.897,
      "step": 680
    },
    {
      "epoch": 0.699537750385208,
      "grad_norm": 2.465571403503418,
      "learning_rate": 4.288161400837458e-06,
      "loss": 1.6588,
      "step": 681
    },
    {
      "epoch": 0.7005649717514124,
      "grad_norm": 2.664262294769287,
      "learning_rate": 4.286258089074991e-06,
      "loss": 1.5862,
      "step": 682
    },
    {
      "epoch": 0.7015921931176169,
      "grad_norm": 2.677140474319458,
      "learning_rate": 4.284354777312524e-06,
      "loss": 1.6549,
      "step": 683
    },
    {
      "epoch": 0.7026194144838213,
      "grad_norm": 2.559051275253296,
      "learning_rate": 4.282451465550058e-06,
      "loss": 1.609,
      "step": 684
    },
    {
      "epoch": 0.7036466358500256,
      "grad_norm": 2.428424119949341,
      "learning_rate": 4.280548153787591e-06,
      "loss": 1.6243,
      "step": 685
    },
    {
      "epoch": 0.7046738572162301,
      "grad_norm": 2.54136061668396,
      "learning_rate": 4.2786448420251245e-06,
      "loss": 1.6301,
      "step": 686
    },
    {
      "epoch": 0.7057010785824345,
      "grad_norm": 2.377699136734009,
      "learning_rate": 4.276741530262658e-06,
      "loss": 1.5903,
      "step": 687
    },
    {
      "epoch": 0.706728299948639,
      "grad_norm": 2.5548322200775146,
      "learning_rate": 4.27483821850019e-06,
      "loss": 1.6161,
      "step": 688
    },
    {
      "epoch": 0.7077555213148433,
      "grad_norm": 2.5993409156799316,
      "learning_rate": 4.272934906737724e-06,
      "loss": 1.6882,
      "step": 689
    },
    {
      "epoch": 0.7087827426810478,
      "grad_norm": 2.4249117374420166,
      "learning_rate": 4.271031594975257e-06,
      "loss": 1.6723,
      "step": 690
    },
    {
      "epoch": 0.7087827426810478,
      "eval_loss": 1.9526821374893188,
      "eval_runtime": 48.2168,
      "eval_samples_per_second": 20.74,
      "eval_steps_per_second": 0.871,
      "step": 690
    },
    {
      "epoch": 0.7098099640472522,
      "grad_norm": 2.6965081691741943,
      "learning_rate": 4.2691282832127905e-06,
      "loss": 1.6636,
      "step": 691
    },
    {
      "epoch": 0.7108371854134566,
      "grad_norm": 2.6205356121063232,
      "learning_rate": 4.267224971450324e-06,
      "loss": 1.6231,
      "step": 692
    },
    {
      "epoch": 0.711864406779661,
      "grad_norm": 2.5924129486083984,
      "learning_rate": 4.265321659687857e-06,
      "loss": 1.5634,
      "step": 693
    },
    {
      "epoch": 0.7128916281458655,
      "grad_norm": 2.9710943698883057,
      "learning_rate": 4.2634183479253906e-06,
      "loss": 1.6134,
      "step": 694
    },
    {
      "epoch": 0.7139188495120699,
      "grad_norm": 2.6252782344818115,
      "learning_rate": 4.261515036162924e-06,
      "loss": 1.5997,
      "step": 695
    },
    {
      "epoch": 0.7149460708782742,
      "grad_norm": 2.6178832054138184,
      "learning_rate": 4.259611724400457e-06,
      "loss": 1.6255,
      "step": 696
    },
    {
      "epoch": 0.7159732922444787,
      "grad_norm": 2.82548189163208,
      "learning_rate": 4.257708412637991e-06,
      "loss": 1.6167,
      "step": 697
    },
    {
      "epoch": 0.7170005136106831,
      "grad_norm": 2.6430351734161377,
      "learning_rate": 4.255805100875523e-06,
      "loss": 1.5585,
      "step": 698
    },
    {
      "epoch": 0.7180277349768875,
      "grad_norm": 2.85046124458313,
      "learning_rate": 4.2539017891130565e-06,
      "loss": 1.5873,
      "step": 699
    },
    {
      "epoch": 0.7190549563430919,
      "grad_norm": 2.5643832683563232,
      "learning_rate": 4.25199847735059e-06,
      "loss": 1.6705,
      "step": 700
    },
    {
      "epoch": 0.7190549563430919,
      "eval_loss": 1.945617437362671,
      "eval_runtime": 54.2778,
      "eval_samples_per_second": 18.424,
      "eval_steps_per_second": 0.774,
      "step": 700
    },
    {
      "epoch": 0.7200821777092964,
      "grad_norm": 2.7856431007385254,
      "learning_rate": 4.250095165588123e-06,
      "loss": 1.5833,
      "step": 701
    },
    {
      "epoch": 0.7211093990755008,
      "grad_norm": 2.9881107807159424,
      "learning_rate": 4.2481918538256575e-06,
      "loss": 1.6438,
      "step": 702
    },
    {
      "epoch": 0.7221366204417052,
      "grad_norm": 2.602938652038574,
      "learning_rate": 4.246288542063191e-06,
      "loss": 1.6135,
      "step": 703
    },
    {
      "epoch": 0.7231638418079096,
      "grad_norm": 2.997450590133667,
      "learning_rate": 4.244385230300724e-06,
      "loss": 1.6419,
      "step": 704
    },
    {
      "epoch": 0.724191063174114,
      "grad_norm": 2.5592496395111084,
      "learning_rate": 4.242481918538257e-06,
      "loss": 1.5886,
      "step": 705
    },
    {
      "epoch": 0.7252182845403184,
      "grad_norm": 2.4073948860168457,
      "learning_rate": 4.24057860677579e-06,
      "loss": 1.6219,
      "step": 706
    },
    {
      "epoch": 0.7262455059065228,
      "grad_norm": 2.545997381210327,
      "learning_rate": 4.2386752950133234e-06,
      "loss": 1.6684,
      "step": 707
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.468264579772949,
      "learning_rate": 4.236771983250857e-06,
      "loss": 1.6155,
      "step": 708
    },
    {
      "epoch": 0.7282999486389317,
      "grad_norm": 2.6635043621063232,
      "learning_rate": 4.23486867148839e-06,
      "loss": 1.5956,
      "step": 709
    },
    {
      "epoch": 0.7293271700051361,
      "grad_norm": 2.8751156330108643,
      "learning_rate": 4.2329653597259235e-06,
      "loss": 1.6211,
      "step": 710
    },
    {
      "epoch": 0.7293271700051361,
      "eval_loss": 1.9319628477096558,
      "eval_runtime": 43.9701,
      "eval_samples_per_second": 22.743,
      "eval_steps_per_second": 0.955,
      "step": 710
    },
    {
      "epoch": 0.7303543913713405,
      "grad_norm": 2.5999035835266113,
      "learning_rate": 4.231062047963457e-06,
      "loss": 1.6442,
      "step": 711
    },
    {
      "epoch": 0.731381612737545,
      "grad_norm": 2.611095666885376,
      "learning_rate": 4.22915873620099e-06,
      "loss": 1.6745,
      "step": 712
    },
    {
      "epoch": 0.7324088341037494,
      "grad_norm": 2.7235333919525146,
      "learning_rate": 4.227255424438524e-06,
      "loss": 1.6353,
      "step": 713
    },
    {
      "epoch": 0.7334360554699538,
      "grad_norm": 2.453639507293701,
      "learning_rate": 4.225352112676057e-06,
      "loss": 1.6657,
      "step": 714
    },
    {
      "epoch": 0.7344632768361582,
      "grad_norm": 2.4031429290771484,
      "learning_rate": 4.2234488009135895e-06,
      "loss": 1.603,
      "step": 715
    },
    {
      "epoch": 0.7354904982023626,
      "grad_norm": 2.631052017211914,
      "learning_rate": 4.221545489151123e-06,
      "loss": 1.5681,
      "step": 716
    },
    {
      "epoch": 0.736517719568567,
      "grad_norm": 2.518758773803711,
      "learning_rate": 4.219642177388656e-06,
      "loss": 1.5941,
      "step": 717
    },
    {
      "epoch": 0.7375449409347714,
      "grad_norm": 2.699491500854492,
      "learning_rate": 4.2177388656261896e-06,
      "loss": 1.6327,
      "step": 718
    },
    {
      "epoch": 0.7385721623009759,
      "grad_norm": 2.749356985092163,
      "learning_rate": 4.215835553863723e-06,
      "loss": 1.5864,
      "step": 719
    },
    {
      "epoch": 0.7395993836671803,
      "grad_norm": 2.5699031352996826,
      "learning_rate": 4.213932242101256e-06,
      "loss": 1.5444,
      "step": 720
    },
    {
      "epoch": 0.7395993836671803,
      "eval_loss": 1.9521623849868774,
      "eval_runtime": 46.5749,
      "eval_samples_per_second": 21.471,
      "eval_steps_per_second": 0.902,
      "step": 720
    },
    {
      "epoch": 0.7406266050333847,
      "grad_norm": 3.197425127029419,
      "learning_rate": 4.21202893033879e-06,
      "loss": 1.6158,
      "step": 721
    },
    {
      "epoch": 0.7416538263995891,
      "grad_norm": 2.619459390640259,
      "learning_rate": 4.210125618576323e-06,
      "loss": 1.6328,
      "step": 722
    },
    {
      "epoch": 0.7426810477657936,
      "grad_norm": 2.792086601257324,
      "learning_rate": 4.208222306813856e-06,
      "loss": 1.6333,
      "step": 723
    },
    {
      "epoch": 0.743708269131998,
      "grad_norm": 2.7023141384124756,
      "learning_rate": 4.20631899505139e-06,
      "loss": 1.6787,
      "step": 724
    },
    {
      "epoch": 0.7447354904982023,
      "grad_norm": 2.6319735050201416,
      "learning_rate": 4.204415683288923e-06,
      "loss": 1.54,
      "step": 725
    },
    {
      "epoch": 0.7457627118644068,
      "grad_norm": 2.7032439708709717,
      "learning_rate": 4.2025123715264565e-06,
      "loss": 1.641,
      "step": 726
    },
    {
      "epoch": 0.7467899332306112,
      "grad_norm": 2.9133267402648926,
      "learning_rate": 4.20060905976399e-06,
      "loss": 1.5901,
      "step": 727
    },
    {
      "epoch": 0.7478171545968156,
      "grad_norm": 2.492403745651245,
      "learning_rate": 4.198705748001523e-06,
      "loss": 1.5516,
      "step": 728
    },
    {
      "epoch": 0.74884437596302,
      "grad_norm": 2.604520082473755,
      "learning_rate": 4.1968024362390566e-06,
      "loss": 1.6574,
      "step": 729
    },
    {
      "epoch": 0.7498715973292245,
      "grad_norm": 2.521969795227051,
      "learning_rate": 4.19489912447659e-06,
      "loss": 1.6099,
      "step": 730
    },
    {
      "epoch": 0.7498715973292245,
      "eval_loss": 1.9449840784072876,
      "eval_runtime": 44.9334,
      "eval_samples_per_second": 22.255,
      "eval_steps_per_second": 0.935,
      "step": 730
    },
    {
      "epoch": 0.7508988186954288,
      "grad_norm": 2.4645657539367676,
      "learning_rate": 4.192995812714123e-06,
      "loss": 1.6752,
      "step": 731
    },
    {
      "epoch": 0.7519260400616333,
      "grad_norm": 2.7852981090545654,
      "learning_rate": 4.191092500951656e-06,
      "loss": 1.6176,
      "step": 732
    },
    {
      "epoch": 0.7529532614278377,
      "grad_norm": 2.2817575931549072,
      "learning_rate": 4.189189189189189e-06,
      "loss": 1.5934,
      "step": 733
    },
    {
      "epoch": 0.7539804827940421,
      "grad_norm": 2.6218173503875732,
      "learning_rate": 4.1872858774267225e-06,
      "loss": 1.5902,
      "step": 734
    },
    {
      "epoch": 0.7550077041602465,
      "grad_norm": 2.6671013832092285,
      "learning_rate": 4.185382565664256e-06,
      "loss": 1.657,
      "step": 735
    },
    {
      "epoch": 0.7560349255264509,
      "grad_norm": 2.669757843017578,
      "learning_rate": 4.183479253901789e-06,
      "loss": 1.5884,
      "step": 736
    },
    {
      "epoch": 0.7570621468926554,
      "grad_norm": 2.7210471630096436,
      "learning_rate": 4.181575942139323e-06,
      "loss": 1.5558,
      "step": 737
    },
    {
      "epoch": 0.7580893682588598,
      "grad_norm": 2.414276123046875,
      "learning_rate": 4.179672630376856e-06,
      "loss": 1.6124,
      "step": 738
    },
    {
      "epoch": 0.7591165896250642,
      "grad_norm": 2.4643094539642334,
      "learning_rate": 4.177769318614389e-06,
      "loss": 1.5904,
      "step": 739
    },
    {
      "epoch": 0.7601438109912686,
      "grad_norm": 2.5447018146514893,
      "learning_rate": 4.175866006851923e-06,
      "loss": 1.6705,
      "step": 740
    },
    {
      "epoch": 0.7601438109912686,
      "eval_loss": 1.9437954425811768,
      "eval_runtime": 49.2702,
      "eval_samples_per_second": 20.296,
      "eval_steps_per_second": 0.852,
      "step": 740
    },
    {
      "epoch": 0.7611710323574731,
      "grad_norm": 2.41048002243042,
      "learning_rate": 4.173962695089456e-06,
      "loss": 1.6398,
      "step": 741
    },
    {
      "epoch": 0.7621982537236774,
      "grad_norm": 2.3949906826019287,
      "learning_rate": 4.1720593833269894e-06,
      "loss": 1.6198,
      "step": 742
    },
    {
      "epoch": 0.7632254750898819,
      "grad_norm": 2.36966609954834,
      "learning_rate": 4.170156071564523e-06,
      "loss": 1.6275,
      "step": 743
    },
    {
      "epoch": 0.7642526964560863,
      "grad_norm": 2.623347043991089,
      "learning_rate": 4.168252759802056e-06,
      "loss": 1.7059,
      "step": 744
    },
    {
      "epoch": 0.7652799178222907,
      "grad_norm": 2.384788751602173,
      "learning_rate": 4.1663494480395895e-06,
      "loss": 1.6015,
      "step": 745
    },
    {
      "epoch": 0.7663071391884951,
      "grad_norm": 2.586009979248047,
      "learning_rate": 4.164446136277123e-06,
      "loss": 1.6445,
      "step": 746
    },
    {
      "epoch": 0.7673343605546995,
      "grad_norm": 2.396684169769287,
      "learning_rate": 4.162542824514656e-06,
      "loss": 1.6539,
      "step": 747
    },
    {
      "epoch": 0.768361581920904,
      "grad_norm": 2.416243553161621,
      "learning_rate": 4.16063951275219e-06,
      "loss": 1.6142,
      "step": 748
    },
    {
      "epoch": 0.7693888032871083,
      "grad_norm": 2.480912208557129,
      "learning_rate": 4.158736200989722e-06,
      "loss": 1.5768,
      "step": 749
    },
    {
      "epoch": 0.7704160246533128,
      "grad_norm": 2.5516929626464844,
      "learning_rate": 4.1568328892272555e-06,
      "loss": 1.6074,
      "step": 750
    },
    {
      "epoch": 0.7704160246533128,
      "eval_loss": 1.9409116506576538,
      "eval_runtime": 45.3465,
      "eval_samples_per_second": 22.052,
      "eval_steps_per_second": 0.926,
      "step": 750
    },
    {
      "epoch": 0.7714432460195172,
      "grad_norm": 2.4248714447021484,
      "learning_rate": 4.154929577464789e-06,
      "loss": 1.6308,
      "step": 751
    },
    {
      "epoch": 0.7724704673857217,
      "grad_norm": 2.520810604095459,
      "learning_rate": 4.153026265702322e-06,
      "loss": 1.6227,
      "step": 752
    },
    {
      "epoch": 0.773497688751926,
      "grad_norm": 2.5938422679901123,
      "learning_rate": 4.1511229539398556e-06,
      "loss": 1.5601,
      "step": 753
    },
    {
      "epoch": 0.7745249101181304,
      "grad_norm": 2.4538512229919434,
      "learning_rate": 4.149219642177389e-06,
      "loss": 1.6006,
      "step": 754
    },
    {
      "epoch": 0.7755521314843349,
      "grad_norm": 2.4936258792877197,
      "learning_rate": 4.147316330414922e-06,
      "loss": 1.6281,
      "step": 755
    },
    {
      "epoch": 0.7765793528505393,
      "grad_norm": 2.5774738788604736,
      "learning_rate": 4.145413018652456e-06,
      "loss": 1.5328,
      "step": 756
    },
    {
      "epoch": 0.7776065742167437,
      "grad_norm": 2.3616905212402344,
      "learning_rate": 4.143509706889989e-06,
      "loss": 1.629,
      "step": 757
    },
    {
      "epoch": 0.7786337955829481,
      "grad_norm": 2.4983086585998535,
      "learning_rate": 4.141606395127522e-06,
      "loss": 1.6298,
      "step": 758
    },
    {
      "epoch": 0.7796610169491526,
      "grad_norm": 2.4313876628875732,
      "learning_rate": 4.139703083365056e-06,
      "loss": 1.6113,
      "step": 759
    },
    {
      "epoch": 0.7806882383153569,
      "grad_norm": 2.6290934085845947,
      "learning_rate": 4.137799771602588e-06,
      "loss": 1.6632,
      "step": 760
    },
    {
      "epoch": 0.7806882383153569,
      "eval_loss": 1.950036644935608,
      "eval_runtime": 47.4523,
      "eval_samples_per_second": 21.074,
      "eval_steps_per_second": 0.885,
      "step": 760
    },
    {
      "epoch": 0.7817154596815614,
      "grad_norm": 2.3637750148773193,
      "learning_rate": 4.135896459840122e-06,
      "loss": 1.6268,
      "step": 761
    },
    {
      "epoch": 0.7827426810477658,
      "grad_norm": 2.5537843704223633,
      "learning_rate": 4.133993148077656e-06,
      "loss": 1.5736,
      "step": 762
    },
    {
      "epoch": 0.7837699024139703,
      "grad_norm": 2.4945273399353027,
      "learning_rate": 4.132089836315189e-06,
      "loss": 1.612,
      "step": 763
    },
    {
      "epoch": 0.7847971237801746,
      "grad_norm": 2.506983757019043,
      "learning_rate": 4.1301865245527226e-06,
      "loss": 1.6112,
      "step": 764
    },
    {
      "epoch": 0.785824345146379,
      "grad_norm": 2.599721908569336,
      "learning_rate": 4.128283212790256e-06,
      "loss": 1.5698,
      "step": 765
    },
    {
      "epoch": 0.7868515665125835,
      "grad_norm": 2.4561266899108887,
      "learning_rate": 4.1263799010277884e-06,
      "loss": 1.5354,
      "step": 766
    },
    {
      "epoch": 0.7878787878787878,
      "grad_norm": 2.445420026779175,
      "learning_rate": 4.124476589265322e-06,
      "loss": 1.6555,
      "step": 767
    },
    {
      "epoch": 0.7889060092449923,
      "grad_norm": 2.576533079147339,
      "learning_rate": 4.122573277502855e-06,
      "loss": 1.5915,
      "step": 768
    },
    {
      "epoch": 0.7899332306111967,
      "grad_norm": 2.503164291381836,
      "learning_rate": 4.1206699657403885e-06,
      "loss": 1.5893,
      "step": 769
    },
    {
      "epoch": 0.7909604519774012,
      "grad_norm": 2.9463603496551514,
      "learning_rate": 4.118766653977922e-06,
      "loss": 1.6178,
      "step": 770
    },
    {
      "epoch": 0.7909604519774012,
      "eval_loss": 1.9529738426208496,
      "eval_runtime": 45.0922,
      "eval_samples_per_second": 22.177,
      "eval_steps_per_second": 0.931,
      "step": 770
    },
    {
      "epoch": 0.7919876733436055,
      "grad_norm": 2.47148060798645,
      "learning_rate": 4.116863342215455e-06,
      "loss": 1.5991,
      "step": 771
    },
    {
      "epoch": 0.79301489470981,
      "grad_norm": 2.515717029571533,
      "learning_rate": 4.114960030452989e-06,
      "loss": 1.5725,
      "step": 772
    },
    {
      "epoch": 0.7940421160760144,
      "grad_norm": 2.5231993198394775,
      "learning_rate": 4.113056718690522e-06,
      "loss": 1.6145,
      "step": 773
    },
    {
      "epoch": 0.7950693374422187,
      "grad_norm": 2.479978084564209,
      "learning_rate": 4.111153406928055e-06,
      "loss": 1.5537,
      "step": 774
    },
    {
      "epoch": 0.7960965588084232,
      "grad_norm": 2.5341250896453857,
      "learning_rate": 4.109250095165589e-06,
      "loss": 1.6419,
      "step": 775
    },
    {
      "epoch": 0.7971237801746276,
      "grad_norm": 2.6277401447296143,
      "learning_rate": 4.107346783403122e-06,
      "loss": 1.6505,
      "step": 776
    },
    {
      "epoch": 0.7981510015408321,
      "grad_norm": 2.4688351154327393,
      "learning_rate": 4.105443471640655e-06,
      "loss": 1.5896,
      "step": 777
    },
    {
      "epoch": 0.7991782229070364,
      "grad_norm": 2.4134581089019775,
      "learning_rate": 4.103540159878188e-06,
      "loss": 1.5788,
      "step": 778
    },
    {
      "epoch": 0.8002054442732409,
      "grad_norm": 2.5090012550354004,
      "learning_rate": 4.101636848115721e-06,
      "loss": 1.6162,
      "step": 779
    },
    {
      "epoch": 0.8012326656394453,
      "grad_norm": 2.5123302936553955,
      "learning_rate": 4.099733536353255e-06,
      "loss": 1.5868,
      "step": 780
    },
    {
      "epoch": 0.8012326656394453,
      "eval_loss": 1.9545625448226929,
      "eval_runtime": 45.9316,
      "eval_samples_per_second": 21.772,
      "eval_steps_per_second": 0.914,
      "step": 780
    },
    {
      "epoch": 0.8022598870056498,
      "grad_norm": 2.787181854248047,
      "learning_rate": 4.097830224590788e-06,
      "loss": 1.612,
      "step": 781
    },
    {
      "epoch": 0.8032871083718541,
      "grad_norm": 2.414426326751709,
      "learning_rate": 4.095926912828322e-06,
      "loss": 1.6908,
      "step": 782
    },
    {
      "epoch": 0.8043143297380585,
      "grad_norm": 2.452697992324829,
      "learning_rate": 4.094023601065855e-06,
      "loss": 1.6575,
      "step": 783
    },
    {
      "epoch": 0.805341551104263,
      "grad_norm": 2.4377169609069824,
      "learning_rate": 4.092120289303388e-06,
      "loss": 1.6016,
      "step": 784
    },
    {
      "epoch": 0.8063687724704673,
      "grad_norm": 2.369154214859009,
      "learning_rate": 4.0902169775409215e-06,
      "loss": 1.6133,
      "step": 785
    },
    {
      "epoch": 0.8073959938366718,
      "grad_norm": 2.261869430541992,
      "learning_rate": 4.088313665778455e-06,
      "loss": 1.5555,
      "step": 786
    },
    {
      "epoch": 0.8084232152028762,
      "grad_norm": 2.7058632373809814,
      "learning_rate": 4.086410354015988e-06,
      "loss": 1.5957,
      "step": 787
    },
    {
      "epoch": 0.8094504365690807,
      "grad_norm": 2.3907508850097656,
      "learning_rate": 4.0845070422535216e-06,
      "loss": 1.5676,
      "step": 788
    },
    {
      "epoch": 0.810477657935285,
      "grad_norm": 2.510284900665283,
      "learning_rate": 4.082603730491055e-06,
      "loss": 1.5872,
      "step": 789
    },
    {
      "epoch": 0.8115048793014895,
      "grad_norm": 2.590810775756836,
      "learning_rate": 4.080700418728588e-06,
      "loss": 1.6315,
      "step": 790
    },
    {
      "epoch": 0.8115048793014895,
      "eval_loss": 1.9573780298233032,
      "eval_runtime": 44.5677,
      "eval_samples_per_second": 22.438,
      "eval_steps_per_second": 0.942,
      "step": 790
    },
    {
      "epoch": 0.8125321006676939,
      "grad_norm": 2.466134786605835,
      "learning_rate": 4.078797106966122e-06,
      "loss": 1.6018,
      "step": 791
    },
    {
      "epoch": 0.8135593220338984,
      "grad_norm": 2.474653482437134,
      "learning_rate": 4.076893795203655e-06,
      "loss": 1.5101,
      "step": 792
    },
    {
      "epoch": 0.8145865434001027,
      "grad_norm": 2.4463517665863037,
      "learning_rate": 4.074990483441188e-06,
      "loss": 1.5686,
      "step": 793
    },
    {
      "epoch": 0.8156137647663071,
      "grad_norm": 2.5148985385894775,
      "learning_rate": 4.073087171678721e-06,
      "loss": 1.6544,
      "step": 794
    },
    {
      "epoch": 0.8166409861325116,
      "grad_norm": 2.6183927059173584,
      "learning_rate": 4.071183859916254e-06,
      "loss": 1.6277,
      "step": 795
    },
    {
      "epoch": 0.8176682074987159,
      "grad_norm": 2.438427686691284,
      "learning_rate": 4.069280548153788e-06,
      "loss": 1.6002,
      "step": 796
    },
    {
      "epoch": 0.8186954288649204,
      "grad_norm": 2.662855863571167,
      "learning_rate": 4.067377236391321e-06,
      "loss": 1.5594,
      "step": 797
    },
    {
      "epoch": 0.8197226502311248,
      "grad_norm": 2.64789080619812,
      "learning_rate": 4.065473924628854e-06,
      "loss": 1.6007,
      "step": 798
    },
    {
      "epoch": 0.8207498715973293,
      "grad_norm": 2.5152997970581055,
      "learning_rate": 4.063570612866388e-06,
      "loss": 1.6366,
      "step": 799
    },
    {
      "epoch": 0.8217770929635336,
      "grad_norm": 2.468672037124634,
      "learning_rate": 4.061667301103921e-06,
      "loss": 1.6313,
      "step": 800
    },
    {
      "epoch": 0.8217770929635336,
      "eval_loss": 1.95821213722229,
      "eval_runtime": 45.9362,
      "eval_samples_per_second": 21.769,
      "eval_steps_per_second": 0.914,
      "step": 800
    },
    {
      "epoch": 0.8228043143297381,
      "grad_norm": 2.440307855606079,
      "learning_rate": 4.0597639893414544e-06,
      "loss": 1.5993,
      "step": 801
    },
    {
      "epoch": 0.8238315356959425,
      "grad_norm": 2.6519880294799805,
      "learning_rate": 4.057860677578988e-06,
      "loss": 1.6106,
      "step": 802
    },
    {
      "epoch": 0.8248587570621468,
      "grad_norm": 2.4224767684936523,
      "learning_rate": 4.055957365816521e-06,
      "loss": 1.6247,
      "step": 803
    },
    {
      "epoch": 0.8258859784283513,
      "grad_norm": 2.346858263015747,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 1.5699,
      "step": 804
    },
    {
      "epoch": 0.8269131997945557,
      "grad_norm": 2.577711582183838,
      "learning_rate": 4.052150742291588e-06,
      "loss": 1.5665,
      "step": 805
    },
    {
      "epoch": 0.8279404211607602,
      "grad_norm": 2.604501724243164,
      "learning_rate": 4.050247430529121e-06,
      "loss": 1.561,
      "step": 806
    },
    {
      "epoch": 0.8289676425269645,
      "grad_norm": 2.534820795059204,
      "learning_rate": 4.048344118766655e-06,
      "loss": 1.6391,
      "step": 807
    },
    {
      "epoch": 0.829994863893169,
      "grad_norm": 2.496765613555908,
      "learning_rate": 4.046440807004188e-06,
      "loss": 1.6338,
      "step": 808
    },
    {
      "epoch": 0.8310220852593734,
      "grad_norm": 2.587185859680176,
      "learning_rate": 4.044537495241721e-06,
      "loss": 1.6009,
      "step": 809
    },
    {
      "epoch": 0.8320493066255779,
      "grad_norm": 2.7336575984954834,
      "learning_rate": 4.042634183479255e-06,
      "loss": 1.5911,
      "step": 810
    },
    {
      "epoch": 0.8320493066255779,
      "eval_loss": 1.951606273651123,
      "eval_runtime": 42.4435,
      "eval_samples_per_second": 23.561,
      "eval_steps_per_second": 0.99,
      "step": 810
    },
    {
      "epoch": 0.8330765279917822,
      "grad_norm": 2.5524845123291016,
      "learning_rate": 4.040730871716787e-06,
      "loss": 1.6406,
      "step": 811
    },
    {
      "epoch": 0.8341037493579867,
      "grad_norm": 2.4861226081848145,
      "learning_rate": 4.038827559954321e-06,
      "loss": 1.6092,
      "step": 812
    },
    {
      "epoch": 0.8351309707241911,
      "grad_norm": 2.668187141418457,
      "learning_rate": 4.036924248191854e-06,
      "loss": 1.6529,
      "step": 813
    },
    {
      "epoch": 0.8361581920903954,
      "grad_norm": 2.5616471767425537,
      "learning_rate": 4.035020936429387e-06,
      "loss": 1.6253,
      "step": 814
    },
    {
      "epoch": 0.8371854134565999,
      "grad_norm": 2.4261975288391113,
      "learning_rate": 4.033117624666921e-06,
      "loss": 1.5955,
      "step": 815
    },
    {
      "epoch": 0.8382126348228043,
      "grad_norm": 2.4947080612182617,
      "learning_rate": 4.031214312904454e-06,
      "loss": 1.5686,
      "step": 816
    },
    {
      "epoch": 0.8392398561890088,
      "grad_norm": 2.747781276702881,
      "learning_rate": 4.029311001141987e-06,
      "loss": 1.5642,
      "step": 817
    },
    {
      "epoch": 0.8402670775552131,
      "grad_norm": 2.4762001037597656,
      "learning_rate": 4.027407689379521e-06,
      "loss": 1.5301,
      "step": 818
    },
    {
      "epoch": 0.8412942989214176,
      "grad_norm": 2.6253550052642822,
      "learning_rate": 4.025504377617054e-06,
      "loss": 1.5885,
      "step": 819
    },
    {
      "epoch": 0.842321520287622,
      "grad_norm": 2.6974875926971436,
      "learning_rate": 4.0236010658545875e-06,
      "loss": 1.5656,
      "step": 820
    },
    {
      "epoch": 0.842321520287622,
      "eval_loss": 1.9420732259750366,
      "eval_runtime": 46.9525,
      "eval_samples_per_second": 21.298,
      "eval_steps_per_second": 0.895,
      "step": 820
    },
    {
      "epoch": 0.8433487416538265,
      "grad_norm": 2.638587474822998,
      "learning_rate": 4.02169775409212e-06,
      "loss": 1.6081,
      "step": 821
    },
    {
      "epoch": 0.8443759630200308,
      "grad_norm": 2.963974714279175,
      "learning_rate": 4.019794442329653e-06,
      "loss": 1.6851,
      "step": 822
    },
    {
      "epoch": 0.8454031843862352,
      "grad_norm": 2.4657533168792725,
      "learning_rate": 4.0178911305671876e-06,
      "loss": 1.5396,
      "step": 823
    },
    {
      "epoch": 0.8464304057524397,
      "grad_norm": 2.5373876094818115,
      "learning_rate": 4.015987818804721e-06,
      "loss": 1.6186,
      "step": 824
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 2.5873284339904785,
      "learning_rate": 4.014084507042254e-06,
      "loss": 1.5618,
      "step": 825
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 2.4308598041534424,
      "learning_rate": 4.012181195279788e-06,
      "loss": 1.5399,
      "step": 826
    },
    {
      "epoch": 0.8495120698510529,
      "grad_norm": 2.5274946689605713,
      "learning_rate": 4.010277883517321e-06,
      "loss": 1.6375,
      "step": 827
    },
    {
      "epoch": 0.8505392912172574,
      "grad_norm": 2.4370181560516357,
      "learning_rate": 4.0083745717548535e-06,
      "loss": 1.6313,
      "step": 828
    },
    {
      "epoch": 0.8515665125834617,
      "grad_norm": 2.4524898529052734,
      "learning_rate": 4.006471259992387e-06,
      "loss": 1.5865,
      "step": 829
    },
    {
      "epoch": 0.8525937339496662,
      "grad_norm": 2.5587961673736572,
      "learning_rate": 4.00456794822992e-06,
      "loss": 1.608,
      "step": 830
    },
    {
      "epoch": 0.8525937339496662,
      "eval_loss": 1.94539475440979,
      "eval_runtime": 49.4543,
      "eval_samples_per_second": 20.221,
      "eval_steps_per_second": 0.849,
      "step": 830
    },
    {
      "epoch": 0.8536209553158706,
      "grad_norm": 2.512253761291504,
      "learning_rate": 4.002664636467454e-06,
      "loss": 1.6607,
      "step": 831
    },
    {
      "epoch": 0.8546481766820749,
      "grad_norm": 2.438451051712036,
      "learning_rate": 4.000761324704987e-06,
      "loss": 1.6153,
      "step": 832
    },
    {
      "epoch": 0.8556753980482794,
      "grad_norm": 2.5550496578216553,
      "learning_rate": 3.99885801294252e-06,
      "loss": 1.5489,
      "step": 833
    },
    {
      "epoch": 0.8567026194144838,
      "grad_norm": 2.559732437133789,
      "learning_rate": 3.996954701180054e-06,
      "loss": 1.6045,
      "step": 834
    },
    {
      "epoch": 0.8577298407806883,
      "grad_norm": 2.317692995071411,
      "learning_rate": 3.995051389417587e-06,
      "loss": 1.5316,
      "step": 835
    },
    {
      "epoch": 0.8587570621468926,
      "grad_norm": 2.3742363452911377,
      "learning_rate": 3.9931480776551204e-06,
      "loss": 1.553,
      "step": 836
    },
    {
      "epoch": 0.8597842835130971,
      "grad_norm": 2.5645551681518555,
      "learning_rate": 3.991244765892654e-06,
      "loss": 1.6558,
      "step": 837
    },
    {
      "epoch": 0.8608115048793015,
      "grad_norm": 2.494830846786499,
      "learning_rate": 3.989341454130186e-06,
      "loss": 1.5713,
      "step": 838
    },
    {
      "epoch": 0.861838726245506,
      "grad_norm": 2.416020393371582,
      "learning_rate": 3.98743814236772e-06,
      "loss": 1.5383,
      "step": 839
    },
    {
      "epoch": 0.8628659476117103,
      "grad_norm": 2.6063015460968018,
      "learning_rate": 3.985534830605253e-06,
      "loss": 1.5831,
      "step": 840
    },
    {
      "epoch": 0.8628659476117103,
      "eval_loss": 1.9441509246826172,
      "eval_runtime": 46.0083,
      "eval_samples_per_second": 21.735,
      "eval_steps_per_second": 0.913,
      "step": 840
    },
    {
      "epoch": 0.8638931689779148,
      "grad_norm": 2.5771663188934326,
      "learning_rate": 3.983631518842786e-06,
      "loss": 1.6077,
      "step": 841
    },
    {
      "epoch": 0.8649203903441192,
      "grad_norm": 2.573119878768921,
      "learning_rate": 3.98172820708032e-06,
      "loss": 1.6301,
      "step": 842
    },
    {
      "epoch": 0.8659476117103235,
      "grad_norm": 2.41396164894104,
      "learning_rate": 3.979824895317854e-06,
      "loss": 1.5758,
      "step": 843
    },
    {
      "epoch": 0.866974833076528,
      "grad_norm": 2.532883882522583,
      "learning_rate": 3.977921583555387e-06,
      "loss": 1.5525,
      "step": 844
    },
    {
      "epoch": 0.8680020544427324,
      "grad_norm": 2.669311285018921,
      "learning_rate": 3.97601827179292e-06,
      "loss": 1.6183,
      "step": 845
    },
    {
      "epoch": 0.8690292758089369,
      "grad_norm": 2.366861343383789,
      "learning_rate": 3.974114960030453e-06,
      "loss": 1.5551,
      "step": 846
    },
    {
      "epoch": 0.8700564971751412,
      "grad_norm": 2.3896048069000244,
      "learning_rate": 3.9722116482679866e-06,
      "loss": 1.5894,
      "step": 847
    },
    {
      "epoch": 0.8710837185413457,
      "grad_norm": 2.6509673595428467,
      "learning_rate": 3.97030833650552e-06,
      "loss": 1.5813,
      "step": 848
    },
    {
      "epoch": 0.8721109399075501,
      "grad_norm": 2.531726121902466,
      "learning_rate": 3.968405024743053e-06,
      "loss": 1.5899,
      "step": 849
    },
    {
      "epoch": 0.8731381612737545,
      "grad_norm": 2.4117019176483154,
      "learning_rate": 3.966501712980587e-06,
      "loss": 1.5117,
      "step": 850
    },
    {
      "epoch": 0.8731381612737545,
      "eval_loss": 1.9224250316619873,
      "eval_runtime": 47.7854,
      "eval_samples_per_second": 20.927,
      "eval_steps_per_second": 0.879,
      "step": 850
    },
    {
      "epoch": 0.8741653826399589,
      "grad_norm": 2.3789749145507812,
      "learning_rate": 3.96459840121812e-06,
      "loss": 1.5708,
      "step": 851
    },
    {
      "epoch": 0.8751926040061633,
      "grad_norm": 2.386664867401123,
      "learning_rate": 3.962695089455653e-06,
      "loss": 1.5734,
      "step": 852
    },
    {
      "epoch": 0.8762198253723678,
      "grad_norm": 2.4788997173309326,
      "learning_rate": 3.960791777693187e-06,
      "loss": 1.6315,
      "step": 853
    },
    {
      "epoch": 0.8772470467385721,
      "grad_norm": 2.630183219909668,
      "learning_rate": 3.95888846593072e-06,
      "loss": 1.614,
      "step": 854
    },
    {
      "epoch": 0.8782742681047766,
      "grad_norm": 2.3957736492156982,
      "learning_rate": 3.956985154168253e-06,
      "loss": 1.61,
      "step": 855
    },
    {
      "epoch": 0.879301489470981,
      "grad_norm": 2.652036428451538,
      "learning_rate": 3.955081842405786e-06,
      "loss": 1.6463,
      "step": 856
    },
    {
      "epoch": 0.8803287108371854,
      "grad_norm": 2.579867124557495,
      "learning_rate": 3.953178530643319e-06,
      "loss": 1.6624,
      "step": 857
    },
    {
      "epoch": 0.8813559322033898,
      "grad_norm": 2.5118343830108643,
      "learning_rate": 3.951275218880853e-06,
      "loss": 1.6816,
      "step": 858
    },
    {
      "epoch": 0.8823831535695943,
      "grad_norm": 2.5963029861450195,
      "learning_rate": 3.949371907118386e-06,
      "loss": 1.5296,
      "step": 859
    },
    {
      "epoch": 0.8834103749357987,
      "grad_norm": 2.571241855621338,
      "learning_rate": 3.9474685953559194e-06,
      "loss": 1.6362,
      "step": 860
    },
    {
      "epoch": 0.8834103749357987,
      "eval_loss": 1.931492567062378,
      "eval_runtime": 52.2417,
      "eval_samples_per_second": 19.142,
      "eval_steps_per_second": 0.804,
      "step": 860
    },
    {
      "epoch": 0.884437596302003,
      "grad_norm": 2.518036365509033,
      "learning_rate": 3.945565283593453e-06,
      "loss": 1.5865,
      "step": 861
    },
    {
      "epoch": 0.8854648176682075,
      "grad_norm": 3.14587664604187,
      "learning_rate": 3.943661971830986e-06,
      "loss": 1.586,
      "step": 862
    },
    {
      "epoch": 0.8864920390344119,
      "grad_norm": 2.686350107192993,
      "learning_rate": 3.9417586600685195e-06,
      "loss": 1.6385,
      "step": 863
    },
    {
      "epoch": 0.8875192604006163,
      "grad_norm": 3.0324997901916504,
      "learning_rate": 3.939855348306053e-06,
      "loss": 1.6199,
      "step": 864
    },
    {
      "epoch": 0.8885464817668207,
      "grad_norm": 2.6129438877105713,
      "learning_rate": 3.937952036543586e-06,
      "loss": 1.6049,
      "step": 865
    },
    {
      "epoch": 0.8895737031330252,
      "grad_norm": 2.795729160308838,
      "learning_rate": 3.93604872478112e-06,
      "loss": 1.6027,
      "step": 866
    },
    {
      "epoch": 0.8906009244992296,
      "grad_norm": 3.2352302074432373,
      "learning_rate": 3.934145413018653e-06,
      "loss": 1.6063,
      "step": 867
    },
    {
      "epoch": 0.891628145865434,
      "grad_norm": 2.5050623416900635,
      "learning_rate": 3.932242101256186e-06,
      "loss": 1.5622,
      "step": 868
    },
    {
      "epoch": 0.8926553672316384,
      "grad_norm": 2.8998777866363525,
      "learning_rate": 3.93033878949372e-06,
      "loss": 1.5729,
      "step": 869
    },
    {
      "epoch": 0.8936825885978429,
      "grad_norm": 2.60239577293396,
      "learning_rate": 3.928435477731253e-06,
      "loss": 1.5377,
      "step": 870
    },
    {
      "epoch": 0.8936825885978429,
      "eval_loss": 1.9311563968658447,
      "eval_runtime": 46.8098,
      "eval_samples_per_second": 21.363,
      "eval_steps_per_second": 0.897,
      "step": 870
    },
    {
      "epoch": 0.8947098099640473,
      "grad_norm": 2.5061111450195312,
      "learning_rate": 3.9265321659687864e-06,
      "loss": 1.6216,
      "step": 871
    },
    {
      "epoch": 0.8957370313302516,
      "grad_norm": 2.7476489543914795,
      "learning_rate": 3.924628854206319e-06,
      "loss": 1.6357,
      "step": 872
    },
    {
      "epoch": 0.8967642526964561,
      "grad_norm": 2.605536937713623,
      "learning_rate": 3.922725542443852e-06,
      "loss": 1.5983,
      "step": 873
    },
    {
      "epoch": 0.8977914740626605,
      "grad_norm": 2.6024532318115234,
      "learning_rate": 3.920822230681386e-06,
      "loss": 1.5571,
      "step": 874
    },
    {
      "epoch": 0.8988186954288649,
      "grad_norm": 2.7357826232910156,
      "learning_rate": 3.918918918918919e-06,
      "loss": 1.5802,
      "step": 875
    },
    {
      "epoch": 0.8998459167950693,
      "grad_norm": 2.6406939029693604,
      "learning_rate": 3.917015607156452e-06,
      "loss": 1.6046,
      "step": 876
    },
    {
      "epoch": 0.9008731381612738,
      "grad_norm": 2.4611780643463135,
      "learning_rate": 3.915112295393986e-06,
      "loss": 1.5803,
      "step": 877
    },
    {
      "epoch": 0.9019003595274782,
      "grad_norm": 3.0608181953430176,
      "learning_rate": 3.913208983631519e-06,
      "loss": 1.6655,
      "step": 878
    },
    {
      "epoch": 0.9029275808936826,
      "grad_norm": 2.482142448425293,
      "learning_rate": 3.9113056718690525e-06,
      "loss": 1.6444,
      "step": 879
    },
    {
      "epoch": 0.903954802259887,
      "grad_norm": 2.4936728477478027,
      "learning_rate": 3.909402360106586e-06,
      "loss": 1.5596,
      "step": 880
    },
    {
      "epoch": 0.903954802259887,
      "eval_loss": 1.933276653289795,
      "eval_runtime": 48.0836,
      "eval_samples_per_second": 20.797,
      "eval_steps_per_second": 0.873,
      "step": 880
    },
    {
      "epoch": 0.9049820236260914,
      "grad_norm": 2.7802248001098633,
      "learning_rate": 3.907499048344119e-06,
      "loss": 1.6133,
      "step": 881
    },
    {
      "epoch": 0.9060092449922958,
      "grad_norm": 2.486586093902588,
      "learning_rate": 3.905595736581652e-06,
      "loss": 1.623,
      "step": 882
    },
    {
      "epoch": 0.9070364663585002,
      "grad_norm": 2.477555751800537,
      "learning_rate": 3.903692424819186e-06,
      "loss": 1.6219,
      "step": 883
    },
    {
      "epoch": 0.9080636877247047,
      "grad_norm": 2.5033347606658936,
      "learning_rate": 3.901789113056719e-06,
      "loss": 1.5612,
      "step": 884
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.410464286804199,
      "learning_rate": 3.899885801294253e-06,
      "loss": 1.6243,
      "step": 885
    },
    {
      "epoch": 0.9101181304571135,
      "grad_norm": 2.451402425765991,
      "learning_rate": 3.897982489531786e-06,
      "loss": 1.5961,
      "step": 886
    },
    {
      "epoch": 0.9111453518233179,
      "grad_norm": 2.439730167388916,
      "learning_rate": 3.896079177769319e-06,
      "loss": 1.6471,
      "step": 887
    },
    {
      "epoch": 0.9121725731895224,
      "grad_norm": 2.527719259262085,
      "learning_rate": 3.894175866006853e-06,
      "loss": 1.6262,
      "step": 888
    },
    {
      "epoch": 0.9131997945557268,
      "grad_norm": 2.612992763519287,
      "learning_rate": 3.892272554244385e-06,
      "loss": 1.6533,
      "step": 889
    },
    {
      "epoch": 0.9142270159219312,
      "grad_norm": 2.509989023208618,
      "learning_rate": 3.890369242481919e-06,
      "loss": 1.5596,
      "step": 890
    },
    {
      "epoch": 0.9142270159219312,
      "eval_loss": 1.9224061965942383,
      "eval_runtime": 45.5462,
      "eval_samples_per_second": 21.956,
      "eval_steps_per_second": 0.922,
      "step": 890
    },
    {
      "epoch": 0.9152542372881356,
      "grad_norm": 2.3849823474884033,
      "learning_rate": 3.888465930719452e-06,
      "loss": 1.6336,
      "step": 891
    },
    {
      "epoch": 0.91628145865434,
      "grad_norm": 2.4989709854125977,
      "learning_rate": 3.886562618956985e-06,
      "loss": 1.5733,
      "step": 892
    },
    {
      "epoch": 0.9173086800205444,
      "grad_norm": 2.4675447940826416,
      "learning_rate": 3.884659307194519e-06,
      "loss": 1.5818,
      "step": 893
    },
    {
      "epoch": 0.9183359013867488,
      "grad_norm": 2.3105645179748535,
      "learning_rate": 3.882755995432052e-06,
      "loss": 1.5251,
      "step": 894
    },
    {
      "epoch": 0.9193631227529533,
      "grad_norm": 2.5410287380218506,
      "learning_rate": 3.8808526836695854e-06,
      "loss": 1.6089,
      "step": 895
    },
    {
      "epoch": 0.9203903441191577,
      "grad_norm": 2.617091655731201,
      "learning_rate": 3.878949371907119e-06,
      "loss": 1.5631,
      "step": 896
    },
    {
      "epoch": 0.9214175654853621,
      "grad_norm": 2.4979820251464844,
      "learning_rate": 3.877046060144652e-06,
      "loss": 1.5577,
      "step": 897
    },
    {
      "epoch": 0.9224447868515665,
      "grad_norm": 2.4857399463653564,
      "learning_rate": 3.8751427483821855e-06,
      "loss": 1.56,
      "step": 898
    },
    {
      "epoch": 0.923472008217771,
      "grad_norm": 2.456709146499634,
      "learning_rate": 3.873239436619718e-06,
      "loss": 1.594,
      "step": 899
    },
    {
      "epoch": 0.9244992295839753,
      "grad_norm": 2.355168342590332,
      "learning_rate": 3.871336124857251e-06,
      "loss": 1.5371,
      "step": 900
    },
    {
      "epoch": 0.9244992295839753,
      "eval_loss": 1.9357926845550537,
      "eval_runtime": 44.4044,
      "eval_samples_per_second": 22.52,
      "eval_steps_per_second": 0.946,
      "step": 900
    },
    {
      "epoch": 0.9255264509501797,
      "grad_norm": 2.5965569019317627,
      "learning_rate": 3.869432813094785e-06,
      "loss": 1.5966,
      "step": 901
    },
    {
      "epoch": 0.9265536723163842,
      "grad_norm": 2.443662405014038,
      "learning_rate": 3.867529501332318e-06,
      "loss": 1.6053,
      "step": 902
    },
    {
      "epoch": 0.9275808936825886,
      "grad_norm": 2.3888511657714844,
      "learning_rate": 3.8656261895698515e-06,
      "loss": 1.5921,
      "step": 903
    },
    {
      "epoch": 0.928608115048793,
      "grad_norm": 2.4648637771606445,
      "learning_rate": 3.863722877807386e-06,
      "loss": 1.5879,
      "step": 904
    },
    {
      "epoch": 0.9296353364149974,
      "grad_norm": 2.4636101722717285,
      "learning_rate": 3.861819566044919e-06,
      "loss": 1.6286,
      "step": 905
    },
    {
      "epoch": 0.9306625577812019,
      "grad_norm": 2.722627878189087,
      "learning_rate": 3.859916254282452e-06,
      "loss": 1.6666,
      "step": 906
    },
    {
      "epoch": 0.9316897791474062,
      "grad_norm": 2.550255298614502,
      "learning_rate": 3.858012942519985e-06,
      "loss": 1.5711,
      "step": 907
    },
    {
      "epoch": 0.9327170005136107,
      "grad_norm": 2.510209798812866,
      "learning_rate": 3.856109630757518e-06,
      "loss": 1.5987,
      "step": 908
    },
    {
      "epoch": 0.9337442218798151,
      "grad_norm": 2.6028876304626465,
      "learning_rate": 3.854206318995052e-06,
      "loss": 1.5682,
      "step": 909
    },
    {
      "epoch": 0.9347714432460195,
      "grad_norm": 2.487908124923706,
      "learning_rate": 3.852303007232585e-06,
      "loss": 1.608,
      "step": 910
    },
    {
      "epoch": 0.9347714432460195,
      "eval_loss": 1.93544602394104,
      "eval_runtime": 41.5201,
      "eval_samples_per_second": 24.085,
      "eval_steps_per_second": 1.012,
      "step": 910
    },
    {
      "epoch": 0.9357986646122239,
      "grad_norm": 2.4569835662841797,
      "learning_rate": 3.850399695470118e-06,
      "loss": 1.5492,
      "step": 911
    },
    {
      "epoch": 0.9368258859784283,
      "grad_norm": 2.572781562805176,
      "learning_rate": 3.848496383707652e-06,
      "loss": 1.5633,
      "step": 912
    },
    {
      "epoch": 0.9378531073446328,
      "grad_norm": 2.6213486194610596,
      "learning_rate": 3.846593071945185e-06,
      "loss": 1.5488,
      "step": 913
    },
    {
      "epoch": 0.9388803287108372,
      "grad_norm": 2.6079821586608887,
      "learning_rate": 3.8446897601827185e-06,
      "loss": 1.5311,
      "step": 914
    },
    {
      "epoch": 0.9399075500770416,
      "grad_norm": 2.6774539947509766,
      "learning_rate": 3.842786448420252e-06,
      "loss": 1.6564,
      "step": 915
    },
    {
      "epoch": 0.940934771443246,
      "grad_norm": 2.5028746128082275,
      "learning_rate": 3.840883136657785e-06,
      "loss": 1.6551,
      "step": 916
    },
    {
      "epoch": 0.9419619928094505,
      "grad_norm": 2.5254387855529785,
      "learning_rate": 3.838979824895318e-06,
      "loss": 1.6492,
      "step": 917
    },
    {
      "epoch": 0.9429892141756548,
      "grad_norm": 2.5451207160949707,
      "learning_rate": 3.837076513132851e-06,
      "loss": 1.5306,
      "step": 918
    },
    {
      "epoch": 0.9440164355418593,
      "grad_norm": 2.526526689529419,
      "learning_rate": 3.8351732013703844e-06,
      "loss": 1.6131,
      "step": 919
    },
    {
      "epoch": 0.9450436569080637,
      "grad_norm": 2.454895257949829,
      "learning_rate": 3.833269889607918e-06,
      "loss": 1.5592,
      "step": 920
    },
    {
      "epoch": 0.9450436569080637,
      "eval_loss": 1.9236185550689697,
      "eval_runtime": 44.149,
      "eval_samples_per_second": 22.651,
      "eval_steps_per_second": 0.951,
      "step": 920
    },
    {
      "epoch": 0.9460708782742681,
      "grad_norm": 2.6773581504821777,
      "learning_rate": 3.831366577845451e-06,
      "loss": 1.5498,
      "step": 921
    },
    {
      "epoch": 0.9470980996404725,
      "grad_norm": 2.63554310798645,
      "learning_rate": 3.8294632660829845e-06,
      "loss": 1.5413,
      "step": 922
    },
    {
      "epoch": 0.9481253210066769,
      "grad_norm": 2.7118968963623047,
      "learning_rate": 3.827559954320518e-06,
      "loss": 1.572,
      "step": 923
    },
    {
      "epoch": 0.9491525423728814,
      "grad_norm": 2.4681396484375,
      "learning_rate": 3.825656642558051e-06,
      "loss": 1.6003,
      "step": 924
    },
    {
      "epoch": 0.9501797637390857,
      "grad_norm": 2.387608051300049,
      "learning_rate": 3.823753330795585e-06,
      "loss": 1.6057,
      "step": 925
    },
    {
      "epoch": 0.9512069851052902,
      "grad_norm": 2.426020383834839,
      "learning_rate": 3.821850019033118e-06,
      "loss": 1.5976,
      "step": 926
    },
    {
      "epoch": 0.9522342064714946,
      "grad_norm": 2.446044445037842,
      "learning_rate": 3.819946707270651e-06,
      "loss": 1.5363,
      "step": 927
    },
    {
      "epoch": 0.9532614278376991,
      "grad_norm": 2.4415299892425537,
      "learning_rate": 3.818043395508185e-06,
      "loss": 1.5685,
      "step": 928
    },
    {
      "epoch": 0.9542886492039034,
      "grad_norm": 2.4006199836730957,
      "learning_rate": 3.816140083745718e-06,
      "loss": 1.5823,
      "step": 929
    },
    {
      "epoch": 0.9553158705701078,
      "grad_norm": 2.6479294300079346,
      "learning_rate": 3.8142367719832514e-06,
      "loss": 1.6658,
      "step": 930
    },
    {
      "epoch": 0.9553158705701078,
      "eval_loss": 1.9434653520584106,
      "eval_runtime": 45.5063,
      "eval_samples_per_second": 21.975,
      "eval_steps_per_second": 0.923,
      "step": 930
    },
    {
      "epoch": 0.9563430919363123,
      "grad_norm": 2.530367851257324,
      "learning_rate": 3.8123334602207844e-06,
      "loss": 1.544,
      "step": 931
    },
    {
      "epoch": 0.9573703133025167,
      "grad_norm": 2.641665458679199,
      "learning_rate": 3.8104301484583177e-06,
      "loss": 1.5658,
      "step": 932
    },
    {
      "epoch": 0.9583975346687211,
      "grad_norm": 2.3992276191711426,
      "learning_rate": 3.808526836695851e-06,
      "loss": 1.5524,
      "step": 933
    },
    {
      "epoch": 0.9594247560349255,
      "grad_norm": 2.640503168106079,
      "learning_rate": 3.8066235249333845e-06,
      "loss": 1.6127,
      "step": 934
    },
    {
      "epoch": 0.96045197740113,
      "grad_norm": 2.513735771179199,
      "learning_rate": 3.804720213170918e-06,
      "loss": 1.5901,
      "step": 935
    },
    {
      "epoch": 0.9614791987673343,
      "grad_norm": 2.74481463432312,
      "learning_rate": 3.8028169014084508e-06,
      "loss": 1.5792,
      "step": 936
    },
    {
      "epoch": 0.9625064201335388,
      "grad_norm": 2.5238802433013916,
      "learning_rate": 3.800913589645984e-06,
      "loss": 1.602,
      "step": 937
    },
    {
      "epoch": 0.9635336414997432,
      "grad_norm": 2.5111966133117676,
      "learning_rate": 3.7990102778835175e-06,
      "loss": 1.607,
      "step": 938
    },
    {
      "epoch": 0.9645608628659477,
      "grad_norm": 2.658449649810791,
      "learning_rate": 3.797106966121051e-06,
      "loss": 1.593,
      "step": 939
    },
    {
      "epoch": 0.965588084232152,
      "grad_norm": 2.6884148120880127,
      "learning_rate": 3.7952036543585842e-06,
      "loss": 1.6576,
      "step": 940
    },
    {
      "epoch": 0.965588084232152,
      "eval_loss": 1.9191738367080688,
      "eval_runtime": 45.1928,
      "eval_samples_per_second": 22.127,
      "eval_steps_per_second": 0.929,
      "step": 940
    },
    {
      "epoch": 0.9666153055983564,
      "grad_norm": 2.4238054752349854,
      "learning_rate": 3.793300342596117e-06,
      "loss": 1.6079,
      "step": 941
    },
    {
      "epoch": 0.9676425269645609,
      "grad_norm": 2.505571126937866,
      "learning_rate": 3.7913970308336505e-06,
      "loss": 1.6299,
      "step": 942
    },
    {
      "epoch": 0.9686697483307652,
      "grad_norm": 2.4143226146698,
      "learning_rate": 3.789493719071184e-06,
      "loss": 1.6158,
      "step": 943
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 2.4664456844329834,
      "learning_rate": 3.7875904073087177e-06,
      "loss": 1.5672,
      "step": 944
    },
    {
      "epoch": 0.9707241910631741,
      "grad_norm": 2.5686466693878174,
      "learning_rate": 3.785687095546251e-06,
      "loss": 1.5338,
      "step": 945
    },
    {
      "epoch": 0.9717514124293786,
      "grad_norm": 2.4666805267333984,
      "learning_rate": 3.7837837837837844e-06,
      "loss": 1.5773,
      "step": 946
    },
    {
      "epoch": 0.9727786337955829,
      "grad_norm": 2.6416635513305664,
      "learning_rate": 3.7818804720213178e-06,
      "loss": 1.5469,
      "step": 947
    },
    {
      "epoch": 0.9738058551617874,
      "grad_norm": 2.721205711364746,
      "learning_rate": 3.7799771602588507e-06,
      "loss": 1.5278,
      "step": 948
    },
    {
      "epoch": 0.9748330765279918,
      "grad_norm": 2.4834787845611572,
      "learning_rate": 3.778073848496384e-06,
      "loss": 1.52,
      "step": 949
    },
    {
      "epoch": 0.9758602978941961,
      "grad_norm": 2.42047381401062,
      "learning_rate": 3.7761705367339174e-06,
      "loss": 1.5814,
      "step": 950
    },
    {
      "epoch": 0.9758602978941961,
      "eval_loss": 1.9303041696548462,
      "eval_runtime": 47.3843,
      "eval_samples_per_second": 21.104,
      "eval_steps_per_second": 0.886,
      "step": 950
    },
    {
      "epoch": 0.9768875192604006,
      "grad_norm": 2.70369815826416,
      "learning_rate": 3.7742672249714508e-06,
      "loss": 1.6143,
      "step": 951
    },
    {
      "epoch": 0.977914740626605,
      "grad_norm": 2.393588066101074,
      "learning_rate": 3.772363913208984e-06,
      "loss": 1.6023,
      "step": 952
    },
    {
      "epoch": 0.9789419619928095,
      "grad_norm": 2.361304998397827,
      "learning_rate": 3.770460601446517e-06,
      "loss": 1.5679,
      "step": 953
    },
    {
      "epoch": 0.9799691833590138,
      "grad_norm": 2.4044971466064453,
      "learning_rate": 3.7685572896840504e-06,
      "loss": 1.6007,
      "step": 954
    },
    {
      "epoch": 0.9809964047252183,
      "grad_norm": 2.498539447784424,
      "learning_rate": 3.766653977921584e-06,
      "loss": 1.5429,
      "step": 955
    },
    {
      "epoch": 0.9820236260914227,
      "grad_norm": 2.457162857055664,
      "learning_rate": 3.764750666159117e-06,
      "loss": 1.4861,
      "step": 956
    },
    {
      "epoch": 0.9830508474576272,
      "grad_norm": 2.52756404876709,
      "learning_rate": 3.7628473543966505e-06,
      "loss": 1.5848,
      "step": 957
    },
    {
      "epoch": 0.9840780688238315,
      "grad_norm": 2.503998279571533,
      "learning_rate": 3.7609440426341835e-06,
      "loss": 1.5435,
      "step": 958
    },
    {
      "epoch": 0.9851052901900359,
      "grad_norm": 2.461087226867676,
      "learning_rate": 3.759040730871717e-06,
      "loss": 1.6344,
      "step": 959
    },
    {
      "epoch": 0.9861325115562404,
      "grad_norm": 2.4149672985076904,
      "learning_rate": 3.75713741910925e-06,
      "loss": 1.5273,
      "step": 960
    },
    {
      "epoch": 0.9861325115562404,
      "eval_loss": 1.928680658340454,
      "eval_runtime": 48.7892,
      "eval_samples_per_second": 20.496,
      "eval_steps_per_second": 0.861,
      "step": 960
    },
    {
      "epoch": 0.9871597329224447,
      "grad_norm": 2.5741734504699707,
      "learning_rate": 3.7552341073467836e-06,
      "loss": 1.5244,
      "step": 961
    },
    {
      "epoch": 0.9881869542886492,
      "grad_norm": 2.5517220497131348,
      "learning_rate": 3.753330795584317e-06,
      "loss": 1.5695,
      "step": 962
    },
    {
      "epoch": 0.9892141756548536,
      "grad_norm": 2.5910379886627197,
      "learning_rate": 3.75142748382185e-06,
      "loss": 1.6322,
      "step": 963
    },
    {
      "epoch": 0.9902413970210581,
      "grad_norm": 2.4780521392822266,
      "learning_rate": 3.749524172059384e-06,
      "loss": 1.5292,
      "step": 964
    },
    {
      "epoch": 0.9912686183872624,
      "grad_norm": 2.6141762733459473,
      "learning_rate": 3.747620860296917e-06,
      "loss": 1.5809,
      "step": 965
    },
    {
      "epoch": 0.9922958397534669,
      "grad_norm": 2.467606544494629,
      "learning_rate": 3.7457175485344504e-06,
      "loss": 1.6123,
      "step": 966
    },
    {
      "epoch": 0.9933230611196713,
      "grad_norm": 2.6112184524536133,
      "learning_rate": 3.7438142367719837e-06,
      "loss": 1.6226,
      "step": 967
    },
    {
      "epoch": 0.9943502824858758,
      "grad_norm": 2.352513313293457,
      "learning_rate": 3.741910925009517e-06,
      "loss": 1.5928,
      "step": 968
    },
    {
      "epoch": 0.9953775038520801,
      "grad_norm": 2.4357361793518066,
      "learning_rate": 3.7400076132470505e-06,
      "loss": 1.5484,
      "step": 969
    },
    {
      "epoch": 0.9964047252182845,
      "grad_norm": 2.5168254375457764,
      "learning_rate": 3.7381043014845834e-06,
      "loss": 1.5722,
      "step": 970
    },
    {
      "epoch": 0.9964047252182845,
      "eval_loss": 1.929540753364563,
      "eval_runtime": 49.7232,
      "eval_samples_per_second": 20.111,
      "eval_steps_per_second": 0.845,
      "step": 970
    },
    {
      "epoch": 0.997431946584489,
      "grad_norm": 2.475173234939575,
      "learning_rate": 3.7362009897221168e-06,
      "loss": 1.5952,
      "step": 971
    },
    {
      "epoch": 0.9984591679506933,
      "grad_norm": 2.5278515815734863,
      "learning_rate": 3.73429767795965e-06,
      "loss": 1.5769,
      "step": 972
    },
    {
      "epoch": 0.9994863893168978,
      "grad_norm": 2.4988675117492676,
      "learning_rate": 3.7323943661971835e-06,
      "loss": 1.5912,
      "step": 973
    },
    {
      "epoch": 1.0005136106831023,
      "grad_norm": 2.406353712081909,
      "learning_rate": 3.730491054434717e-06,
      "loss": 1.5022,
      "step": 974
    },
    {
      "epoch": 1.0015408320493067,
      "grad_norm": 2.611318349838257,
      "learning_rate": 3.72858774267225e-06,
      "loss": 1.4633,
      "step": 975
    },
    {
      "epoch": 1.002568053415511,
      "grad_norm": 2.504613161087036,
      "learning_rate": 3.726684430909783e-06,
      "loss": 1.5405,
      "step": 976
    },
    {
      "epoch": 1.0035952747817154,
      "grad_norm": 2.6135828495025635,
      "learning_rate": 3.7247811191473165e-06,
      "loss": 1.5756,
      "step": 977
    },
    {
      "epoch": 1.00462249614792,
      "grad_norm": 2.692592144012451,
      "learning_rate": 3.72287780738485e-06,
      "loss": 1.5777,
      "step": 978
    },
    {
      "epoch": 1.0056497175141244,
      "grad_norm": 2.660323143005371,
      "learning_rate": 3.7209744956223832e-06,
      "loss": 1.5488,
      "step": 979
    },
    {
      "epoch": 1.0066769388803287,
      "grad_norm": 2.647883415222168,
      "learning_rate": 3.719071183859916e-06,
      "loss": 1.5619,
      "step": 980
    },
    {
      "epoch": 1.0066769388803287,
      "eval_loss": 1.9224871397018433,
      "eval_runtime": 52.6149,
      "eval_samples_per_second": 19.006,
      "eval_steps_per_second": 0.798,
      "step": 980
    },
    {
      "epoch": 1.007704160246533,
      "grad_norm": 2.6186633110046387,
      "learning_rate": 3.7171678720974495e-06,
      "loss": 1.492,
      "step": 981
    },
    {
      "epoch": 1.0087313816127375,
      "grad_norm": 2.7562549114227295,
      "learning_rate": 3.715264560334983e-06,
      "loss": 1.5932,
      "step": 982
    },
    {
      "epoch": 1.009758602978942,
      "grad_norm": 2.618856430053711,
      "learning_rate": 3.7133612485725163e-06,
      "loss": 1.5514,
      "step": 983
    },
    {
      "epoch": 1.0107858243451464,
      "grad_norm": 2.522064685821533,
      "learning_rate": 3.71145793681005e-06,
      "loss": 1.5699,
      "step": 984
    },
    {
      "epoch": 1.0118130457113508,
      "grad_norm": 2.627108335494995,
      "learning_rate": 3.7095546250475834e-06,
      "loss": 1.5033,
      "step": 985
    },
    {
      "epoch": 1.0128402670775551,
      "grad_norm": 2.6393308639526367,
      "learning_rate": 3.7076513132851168e-06,
      "loss": 1.6064,
      "step": 986
    },
    {
      "epoch": 1.0138674884437597,
      "grad_norm": 2.560675621032715,
      "learning_rate": 3.7057480015226497e-06,
      "loss": 1.5167,
      "step": 987
    },
    {
      "epoch": 1.014894709809964,
      "grad_norm": 2.4677579402923584,
      "learning_rate": 3.703844689760183e-06,
      "loss": 1.4953,
      "step": 988
    },
    {
      "epoch": 1.0159219311761685,
      "grad_norm": 2.9904487133026123,
      "learning_rate": 3.7019413779977164e-06,
      "loss": 1.5786,
      "step": 989
    },
    {
      "epoch": 1.0169491525423728,
      "grad_norm": 2.7342119216918945,
      "learning_rate": 3.70003806623525e-06,
      "loss": 1.5459,
      "step": 990
    },
    {
      "epoch": 1.0169491525423728,
      "eval_loss": 1.929221510887146,
      "eval_runtime": 45.9704,
      "eval_samples_per_second": 21.753,
      "eval_steps_per_second": 0.914,
      "step": 990
    },
    {
      "epoch": 1.0179763739085772,
      "grad_norm": 2.723241090774536,
      "learning_rate": 3.698134754472783e-06,
      "loss": 1.5317,
      "step": 991
    },
    {
      "epoch": 1.0190035952747818,
      "grad_norm": 2.562356948852539,
      "learning_rate": 3.696231442710316e-06,
      "loss": 1.5957,
      "step": 992
    },
    {
      "epoch": 1.0200308166409862,
      "grad_norm": 2.635737419128418,
      "learning_rate": 3.6943281309478495e-06,
      "loss": 1.5404,
      "step": 993
    },
    {
      "epoch": 1.0210580380071905,
      "grad_norm": 2.588714361190796,
      "learning_rate": 3.692424819185383e-06,
      "loss": 1.5618,
      "step": 994
    },
    {
      "epoch": 1.022085259373395,
      "grad_norm": 2.61643385887146,
      "learning_rate": 3.690521507422916e-06,
      "loss": 1.5625,
      "step": 995
    },
    {
      "epoch": 1.0231124807395995,
      "grad_norm": 2.566077470779419,
      "learning_rate": 3.6886181956604496e-06,
      "loss": 1.5051,
      "step": 996
    },
    {
      "epoch": 1.0241397021058039,
      "grad_norm": 2.5667169094085693,
      "learning_rate": 3.686714883897983e-06,
      "loss": 1.5142,
      "step": 997
    },
    {
      "epoch": 1.0251669234720082,
      "grad_norm": 2.839078903198242,
      "learning_rate": 3.684811572135516e-06,
      "loss": 1.5833,
      "step": 998
    },
    {
      "epoch": 1.0261941448382126,
      "grad_norm": 2.651643753051758,
      "learning_rate": 3.6829082603730492e-06,
      "loss": 1.5735,
      "step": 999
    },
    {
      "epoch": 1.027221366204417,
      "grad_norm": 2.6310813426971436,
      "learning_rate": 3.6810049486105826e-06,
      "loss": 1.5412,
      "step": 1000
    },
    {
      "epoch": 1.027221366204417,
      "eval_loss": 1.9219006299972534,
      "eval_runtime": 53.0111,
      "eval_samples_per_second": 18.864,
      "eval_steps_per_second": 0.792,
      "step": 1000
    }
  ],
  "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": 3.11730940936192e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}