{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.362609142270159,
  "eval_steps": 10,
  "global_step": 2300,
  "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": 2.7522,
      "step": 1
    },
    {
      "epoch": 0.002054442732408834,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7541,
      "step": 2
    },
    {
      "epoch": 0.0030816640986132513,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8031,
      "step": 3
    },
    {
      "epoch": 0.004108885464817668,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8062,
      "step": 4
    },
    {
      "epoch": 0.005136106831022085,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7733,
      "step": 5
    },
    {
      "epoch": 0.0061633281972265025,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7949,
      "step": 6
    },
    {
      "epoch": 0.007190549563430919,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7657,
      "step": 7
    },
    {
      "epoch": 0.008217770929635337,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.803,
      "step": 8
    },
    {
      "epoch": 0.009244992295839754,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7793,
      "step": 9
    },
    {
      "epoch": 0.01027221366204417,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7935,
      "step": 10
    },
    {
      "epoch": 0.01027221366204417,
      "eval_loss": 2.8114004135131836,
      "eval_runtime": 37.8463,
      "eval_samples_per_second": 26.423,
      "eval_steps_per_second": 1.11,
      "step": 10
    },
    {
      "epoch": 0.011299435028248588,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7546,
      "step": 11
    },
    {
      "epoch": 0.012326656394453005,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7423,
      "step": 12
    },
    {
      "epoch": 0.01335387776065742,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8057,
      "step": 13
    },
    {
      "epoch": 0.014381099126861838,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7663,
      "step": 14
    },
    {
      "epoch": 0.015408320493066256,
      "grad_norm": 4.694424152374268,
      "learning_rate": 1.7123287671232876e-08,
      "loss": 2.7715,
      "step": 15
    },
    {
      "epoch": 0.016435541859270673,
      "grad_norm": 4.905329704284668,
      "learning_rate": 3.424657534246575e-08,
      "loss": 2.751,
      "step": 16
    },
    {
      "epoch": 0.01746276322547509,
      "grad_norm": 4.705279350280762,
      "learning_rate": 5.136986301369863e-08,
      "loss": 2.7186,
      "step": 17
    },
    {
      "epoch": 0.01848998459167951,
      "grad_norm": 5.154810905456543,
      "learning_rate": 6.84931506849315e-08,
      "loss": 2.8073,
      "step": 18
    },
    {
      "epoch": 0.019517205957883924,
      "grad_norm": 5.070403099060059,
      "learning_rate": 8.561643835616439e-08,
      "loss": 2.7237,
      "step": 19
    },
    {
      "epoch": 0.02054442732408834,
      "grad_norm": 4.901423454284668,
      "learning_rate": 1.0273972602739726e-07,
      "loss": 2.7882,
      "step": 20
    },
    {
      "epoch": 0.02054442732408834,
      "eval_loss": 2.810307502746582,
      "eval_runtime": 37.65,
      "eval_samples_per_second": 26.56,
      "eval_steps_per_second": 1.116,
      "step": 20
    },
    {
      "epoch": 0.02157164869029276,
      "grad_norm": 4.914156436920166,
      "learning_rate": 1.1986301369863014e-07,
      "loss": 2.8005,
      "step": 21
    },
    {
      "epoch": 0.022598870056497175,
      "grad_norm": 4.937129974365234,
      "learning_rate": 1.36986301369863e-07,
      "loss": 2.7902,
      "step": 22
    },
    {
      "epoch": 0.02362609142270159,
      "grad_norm": 5.0690202713012695,
      "learning_rate": 1.541095890410959e-07,
      "loss": 2.7474,
      "step": 23
    },
    {
      "epoch": 0.02465331278890601,
      "grad_norm": 4.980844974517822,
      "learning_rate": 1.7123287671232878e-07,
      "loss": 2.7695,
      "step": 24
    },
    {
      "epoch": 0.025680534155110426,
      "grad_norm": 4.8092451095581055,
      "learning_rate": 1.8835616438356165e-07,
      "loss": 2.7611,
      "step": 25
    },
    {
      "epoch": 0.02670775552131484,
      "grad_norm": 4.765528202056885,
      "learning_rate": 2.0547945205479452e-07,
      "loss": 2.7823,
      "step": 26
    },
    {
      "epoch": 0.02773497688751926,
      "grad_norm": 3.9759938716888428,
      "learning_rate": 2.226027397260274e-07,
      "loss": 2.7445,
      "step": 27
    },
    {
      "epoch": 0.028762198253723677,
      "grad_norm": 4.023419380187988,
      "learning_rate": 2.397260273972603e-07,
      "loss": 2.7679,
      "step": 28
    },
    {
      "epoch": 0.029789419619928096,
      "grad_norm": 4.161378383636475,
      "learning_rate": 2.568493150684932e-07,
      "loss": 2.7704,
      "step": 29
    },
    {
      "epoch": 0.030816640986132512,
      "grad_norm": 3.8363842964172363,
      "learning_rate": 2.73972602739726e-07,
      "loss": 2.7598,
      "step": 30
    },
    {
      "epoch": 0.030816640986132512,
      "eval_loss": 2.772540330886841,
      "eval_runtime": 37.2624,
      "eval_samples_per_second": 26.837,
      "eval_steps_per_second": 1.127,
      "step": 30
    },
    {
      "epoch": 0.03184386235233693,
      "grad_norm": 3.718747615814209,
      "learning_rate": 2.910958904109589e-07,
      "loss": 2.7241,
      "step": 31
    },
    {
      "epoch": 0.03287108371854135,
      "grad_norm": 3.555985927581787,
      "learning_rate": 3.082191780821918e-07,
      "loss": 2.7751,
      "step": 32
    },
    {
      "epoch": 0.03389830508474576,
      "grad_norm": 3.6041972637176514,
      "learning_rate": 3.2534246575342466e-07,
      "loss": 2.7322,
      "step": 33
    },
    {
      "epoch": 0.03492552645095018,
      "grad_norm": 3.4408438205718994,
      "learning_rate": 3.4246575342465755e-07,
      "loss": 2.727,
      "step": 34
    },
    {
      "epoch": 0.035952747817154594,
      "grad_norm": 3.4169280529022217,
      "learning_rate": 3.595890410958904e-07,
      "loss": 2.7062,
      "step": 35
    },
    {
      "epoch": 0.03697996918335902,
      "grad_norm": 3.3882925510406494,
      "learning_rate": 3.767123287671233e-07,
      "loss": 2.6271,
      "step": 36
    },
    {
      "epoch": 0.03800719054956343,
      "grad_norm": 3.0491855144500732,
      "learning_rate": 3.938356164383562e-07,
      "loss": 2.7309,
      "step": 37
    },
    {
      "epoch": 0.03903441191576785,
      "grad_norm": 2.8444902896881104,
      "learning_rate": 4.1095890410958903e-07,
      "loss": 2.647,
      "step": 38
    },
    {
      "epoch": 0.040061633281972264,
      "grad_norm": 2.7438602447509766,
      "learning_rate": 4.2808219178082193e-07,
      "loss": 2.6478,
      "step": 39
    },
    {
      "epoch": 0.04108885464817668,
      "grad_norm": 2.8886020183563232,
      "learning_rate": 4.452054794520548e-07,
      "loss": 2.7065,
      "step": 40
    },
    {
      "epoch": 0.04108885464817668,
      "eval_loss": 2.7021424770355225,
      "eval_runtime": 37.3227,
      "eval_samples_per_second": 26.793,
      "eval_steps_per_second": 1.125,
      "step": 40
    },
    {
      "epoch": 0.042116076014381096,
      "grad_norm": 2.8825268745422363,
      "learning_rate": 4.6232876712328767e-07,
      "loss": 2.6666,
      "step": 41
    },
    {
      "epoch": 0.04314329738058552,
      "grad_norm": 2.8848748207092285,
      "learning_rate": 4.794520547945206e-07,
      "loss": 2.597,
      "step": 42
    },
    {
      "epoch": 0.044170518746789934,
      "grad_norm": 2.9176979064941406,
      "learning_rate": 4.965753424657534e-07,
      "loss": 2.6638,
      "step": 43
    },
    {
      "epoch": 0.04519774011299435,
      "grad_norm": 2.6903059482574463,
      "learning_rate": 5.136986301369864e-07,
      "loss": 2.6202,
      "step": 44
    },
    {
      "epoch": 0.046224961479198766,
      "grad_norm": 2.5349855422973633,
      "learning_rate": 5.308219178082192e-07,
      "loss": 2.6466,
      "step": 45
    },
    {
      "epoch": 0.04725218284540318,
      "grad_norm": 2.405383825302124,
      "learning_rate": 5.47945205479452e-07,
      "loss": 2.6291,
      "step": 46
    },
    {
      "epoch": 0.048279404211607604,
      "grad_norm": 2.526106595993042,
      "learning_rate": 5.65068493150685e-07,
      "loss": 2.6718,
      "step": 47
    },
    {
      "epoch": 0.04930662557781202,
      "grad_norm": 2.3981597423553467,
      "learning_rate": 5.821917808219178e-07,
      "loss": 2.592,
      "step": 48
    },
    {
      "epoch": 0.050333846944016436,
      "grad_norm": 2.61842679977417,
      "learning_rate": 5.993150684931507e-07,
      "loss": 2.6039,
      "step": 49
    },
    {
      "epoch": 0.05136106831022085,
      "grad_norm": 2.3001420497894287,
      "learning_rate": 6.164383561643836e-07,
      "loss": 2.5481,
      "step": 50
    },
    {
      "epoch": 0.05136106831022085,
      "eval_loss": 2.6244444847106934,
      "eval_runtime": 37.3248,
      "eval_samples_per_second": 26.792,
      "eval_steps_per_second": 1.125,
      "step": 50
    },
    {
      "epoch": 0.05238828967642527,
      "grad_norm": 2.12744140625,
      "learning_rate": 6.335616438356165e-07,
      "loss": 2.5336,
      "step": 51
    },
    {
      "epoch": 0.05341551104262968,
      "grad_norm": 1.974413275718689,
      "learning_rate": 6.506849315068493e-07,
      "loss": 2.5882,
      "step": 52
    },
    {
      "epoch": 0.054442732408834106,
      "grad_norm": 2.098527669906616,
      "learning_rate": 6.678082191780823e-07,
      "loss": 2.5073,
      "step": 53
    },
    {
      "epoch": 0.05546995377503852,
      "grad_norm": 2.056288719177246,
      "learning_rate": 6.849315068493151e-07,
      "loss": 2.5612,
      "step": 54
    },
    {
      "epoch": 0.05649717514124294,
      "grad_norm": 2.2291629314422607,
      "learning_rate": 7.020547945205481e-07,
      "loss": 2.527,
      "step": 55
    },
    {
      "epoch": 0.05752439650744735,
      "grad_norm": 2.140164375305176,
      "learning_rate": 7.191780821917808e-07,
      "loss": 2.4972,
      "step": 56
    },
    {
      "epoch": 0.05855161787365177,
      "grad_norm": 1.8682485818862915,
      "learning_rate": 7.363013698630137e-07,
      "loss": 2.5231,
      "step": 57
    },
    {
      "epoch": 0.05957883923985619,
      "grad_norm": 1.9373308420181274,
      "learning_rate": 7.534246575342466e-07,
      "loss": 2.4509,
      "step": 58
    },
    {
      "epoch": 0.06060606060606061,
      "grad_norm": 1.8535712957382202,
      "learning_rate": 7.705479452054795e-07,
      "loss": 2.5164,
      "step": 59
    },
    {
      "epoch": 0.061633281972265024,
      "grad_norm": 1.7497329711914062,
      "learning_rate": 7.876712328767124e-07,
      "loss": 2.5177,
      "step": 60
    },
    {
      "epoch": 0.061633281972265024,
      "eval_loss": 2.5608646869659424,
      "eval_runtime": 37.3426,
      "eval_samples_per_second": 26.779,
      "eval_steps_per_second": 1.125,
      "step": 60
    },
    {
      "epoch": 0.06266050333846944,
      "grad_norm": 1.8476629257202148,
      "learning_rate": 8.047945205479453e-07,
      "loss": 2.4838,
      "step": 61
    },
    {
      "epoch": 0.06368772470467386,
      "grad_norm": 1.7130696773529053,
      "learning_rate": 8.219178082191781e-07,
      "loss": 2.473,
      "step": 62
    },
    {
      "epoch": 0.06471494607087827,
      "grad_norm": 1.751081109046936,
      "learning_rate": 8.39041095890411e-07,
      "loss": 2.5152,
      "step": 63
    },
    {
      "epoch": 0.0657421674370827,
      "grad_norm": 1.7758578062057495,
      "learning_rate": 8.561643835616439e-07,
      "loss": 2.5031,
      "step": 64
    },
    {
      "epoch": 0.0667693888032871,
      "grad_norm": 1.9316363334655762,
      "learning_rate": 8.732876712328768e-07,
      "loss": 2.3695,
      "step": 65
    },
    {
      "epoch": 0.06779661016949153,
      "grad_norm": 1.7765878438949585,
      "learning_rate": 8.904109589041097e-07,
      "loss": 2.4669,
      "step": 66
    },
    {
      "epoch": 0.06882383153569595,
      "grad_norm": 1.7339810132980347,
      "learning_rate": 9.075342465753426e-07,
      "loss": 2.4112,
      "step": 67
    },
    {
      "epoch": 0.06985105290190036,
      "grad_norm": 1.717781662940979,
      "learning_rate": 9.246575342465753e-07,
      "loss": 2.4677,
      "step": 68
    },
    {
      "epoch": 0.07087827426810478,
      "grad_norm": 1.7711046934127808,
      "learning_rate": 9.417808219178083e-07,
      "loss": 2.4131,
      "step": 69
    },
    {
      "epoch": 0.07190549563430919,
      "grad_norm": 1.708680510520935,
      "learning_rate": 9.589041095890411e-07,
      "loss": 2.5191,
      "step": 70
    },
    {
      "epoch": 0.07190549563430919,
      "eval_loss": 2.4920735359191895,
      "eval_runtime": 37.3501,
      "eval_samples_per_second": 26.774,
      "eval_steps_per_second": 1.124,
      "step": 70
    },
    {
      "epoch": 0.07293271700051361,
      "grad_norm": 1.9429680109024048,
      "learning_rate": 9.76027397260274e-07,
      "loss": 2.4099,
      "step": 71
    },
    {
      "epoch": 0.07395993836671803,
      "grad_norm": 1.7411247491836548,
      "learning_rate": 9.931506849315068e-07,
      "loss": 2.3632,
      "step": 72
    },
    {
      "epoch": 0.07498715973292244,
      "grad_norm": 1.7577983140945435,
      "learning_rate": 1.0102739726027399e-06,
      "loss": 2.4214,
      "step": 73
    },
    {
      "epoch": 0.07601438109912687,
      "grad_norm": 2.025926351547241,
      "learning_rate": 1.0273972602739727e-06,
      "loss": 2.394,
      "step": 74
    },
    {
      "epoch": 0.07704160246533127,
      "grad_norm": 2.4366860389709473,
      "learning_rate": 1.0445205479452056e-06,
      "loss": 2.3943,
      "step": 75
    },
    {
      "epoch": 0.0780688238315357,
      "grad_norm": 2.437108278274536,
      "learning_rate": 1.0616438356164384e-06,
      "loss": 2.3844,
      "step": 76
    },
    {
      "epoch": 0.07909604519774012,
      "grad_norm": 2.1352434158325195,
      "learning_rate": 1.0787671232876712e-06,
      "loss": 2.3871,
      "step": 77
    },
    {
      "epoch": 0.08012326656394453,
      "grad_norm": 1.8895397186279297,
      "learning_rate": 1.095890410958904e-06,
      "loss": 2.3927,
      "step": 78
    },
    {
      "epoch": 0.08115048793014895,
      "grad_norm": 2.2960093021392822,
      "learning_rate": 1.1130136986301371e-06,
      "loss": 2.3343,
      "step": 79
    },
    {
      "epoch": 0.08217770929635336,
      "grad_norm": 2.0891849994659424,
      "learning_rate": 1.13013698630137e-06,
      "loss": 2.3732,
      "step": 80
    },
    {
      "epoch": 0.08217770929635336,
      "eval_loss": 2.4189229011535645,
      "eval_runtime": 37.3645,
      "eval_samples_per_second": 26.763,
      "eval_steps_per_second": 1.124,
      "step": 80
    },
    {
      "epoch": 0.08320493066255778,
      "grad_norm": 2.450587034225464,
      "learning_rate": 1.1472602739726028e-06,
      "loss": 2.3163,
      "step": 81
    },
    {
      "epoch": 0.08423215202876219,
      "grad_norm": 2.079949378967285,
      "learning_rate": 1.1643835616438357e-06,
      "loss": 2.3381,
      "step": 82
    },
    {
      "epoch": 0.08525937339496661,
      "grad_norm": 2.160737991333008,
      "learning_rate": 1.1815068493150685e-06,
      "loss": 2.3376,
      "step": 83
    },
    {
      "epoch": 0.08628659476117104,
      "grad_norm": 2.463890552520752,
      "learning_rate": 1.1986301369863014e-06,
      "loss": 2.308,
      "step": 84
    },
    {
      "epoch": 0.08731381612737545,
      "grad_norm": 2.3923065662384033,
      "learning_rate": 1.2157534246575344e-06,
      "loss": 2.2682,
      "step": 85
    },
    {
      "epoch": 0.08834103749357987,
      "grad_norm": 2.2096004486083984,
      "learning_rate": 1.2328767123287673e-06,
      "loss": 2.3256,
      "step": 86
    },
    {
      "epoch": 0.08936825885978428,
      "grad_norm": 2.1402556896209717,
      "learning_rate": 1.25e-06,
      "loss": 2.2767,
      "step": 87
    },
    {
      "epoch": 0.0903954802259887,
      "grad_norm": 2.500629186630249,
      "learning_rate": 1.267123287671233e-06,
      "loss": 2.2576,
      "step": 88
    },
    {
      "epoch": 0.09142270159219312,
      "grad_norm": 2.0257675647735596,
      "learning_rate": 1.284246575342466e-06,
      "loss": 2.2722,
      "step": 89
    },
    {
      "epoch": 0.09244992295839753,
      "grad_norm": 2.148428440093994,
      "learning_rate": 1.3013698630136986e-06,
      "loss": 2.2309,
      "step": 90
    },
    {
      "epoch": 0.09244992295839753,
      "eval_loss": 2.3439993858337402,
      "eval_runtime": 37.3803,
      "eval_samples_per_second": 26.752,
      "eval_steps_per_second": 1.124,
      "step": 90
    },
    {
      "epoch": 0.09347714432460195,
      "grad_norm": 2.3541946411132812,
      "learning_rate": 1.3184931506849317e-06,
      "loss": 2.2347,
      "step": 91
    },
    {
      "epoch": 0.09450436569080636,
      "grad_norm": 2.6239383220672607,
      "learning_rate": 1.3356164383561645e-06,
      "loss": 2.2589,
      "step": 92
    },
    {
      "epoch": 0.09553158705701079,
      "grad_norm": 2.1981804370880127,
      "learning_rate": 1.3527397260273976e-06,
      "loss": 2.2372,
      "step": 93
    },
    {
      "epoch": 0.09655880842321521,
      "grad_norm": 2.3652358055114746,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 2.2169,
      "step": 94
    },
    {
      "epoch": 0.09758602978941962,
      "grad_norm": 2.2078158855438232,
      "learning_rate": 1.386986301369863e-06,
      "loss": 2.2761,
      "step": 95
    },
    {
      "epoch": 0.09861325115562404,
      "grad_norm": 2.494685649871826,
      "learning_rate": 1.4041095890410961e-06,
      "loss": 2.1691,
      "step": 96
    },
    {
      "epoch": 0.09964047252182845,
      "grad_norm": 2.3293051719665527,
      "learning_rate": 1.421232876712329e-06,
      "loss": 2.2246,
      "step": 97
    },
    {
      "epoch": 0.10066769388803287,
      "grad_norm": 2.410687208175659,
      "learning_rate": 1.4383561643835616e-06,
      "loss": 2.1166,
      "step": 98
    },
    {
      "epoch": 0.1016949152542373,
      "grad_norm": 2.5168960094451904,
      "learning_rate": 1.4554794520547946e-06,
      "loss": 2.1417,
      "step": 99
    },
    {
      "epoch": 0.1027221366204417,
      "grad_norm": 2.178412914276123,
      "learning_rate": 1.4726027397260275e-06,
      "loss": 2.1638,
      "step": 100
    },
    {
      "epoch": 0.1027221366204417,
      "eval_loss": 2.273622751235962,
      "eval_runtime": 37.3892,
      "eval_samples_per_second": 26.746,
      "eval_steps_per_second": 1.123,
      "step": 100
    },
    {
      "epoch": 0.10374935798664613,
      "grad_norm": 2.8951706886291504,
      "learning_rate": 1.4897260273972605e-06,
      "loss": 2.1631,
      "step": 101
    },
    {
      "epoch": 0.10477657935285054,
      "grad_norm": 2.103206157684326,
      "learning_rate": 1.5068493150684932e-06,
      "loss": 2.1419,
      "step": 102
    },
    {
      "epoch": 0.10580380071905496,
      "grad_norm": 2.495969772338867,
      "learning_rate": 1.5239726027397262e-06,
      "loss": 2.1029,
      "step": 103
    },
    {
      "epoch": 0.10683102208525937,
      "grad_norm": 2.5686826705932617,
      "learning_rate": 1.541095890410959e-06,
      "loss": 2.1235,
      "step": 104
    },
    {
      "epoch": 0.10785824345146379,
      "grad_norm": 3.4710471630096436,
      "learning_rate": 1.5582191780821921e-06,
      "loss": 2.1832,
      "step": 105
    },
    {
      "epoch": 0.10888546481766821,
      "grad_norm": 2.2563316822052,
      "learning_rate": 1.5753424657534248e-06,
      "loss": 2.1441,
      "step": 106
    },
    {
      "epoch": 0.10991268618387262,
      "grad_norm": 3.860506534576416,
      "learning_rate": 1.5924657534246576e-06,
      "loss": 2.1764,
      "step": 107
    },
    {
      "epoch": 0.11093990755007704,
      "grad_norm": 2.550135850906372,
      "learning_rate": 1.6095890410958907e-06,
      "loss": 2.1455,
      "step": 108
    },
    {
      "epoch": 0.11196712891628145,
      "grad_norm": 3.168283700942993,
      "learning_rate": 1.6267123287671235e-06,
      "loss": 2.0246,
      "step": 109
    },
    {
      "epoch": 0.11299435028248588,
      "grad_norm": 2.577069044113159,
      "learning_rate": 1.6438356164383561e-06,
      "loss": 2.0921,
      "step": 110
    },
    {
      "epoch": 0.11299435028248588,
      "eval_loss": 2.2218997478485107,
      "eval_runtime": 37.246,
      "eval_samples_per_second": 26.849,
      "eval_steps_per_second": 1.128,
      "step": 110
    },
    {
      "epoch": 0.1140215716486903,
      "grad_norm": 2.3515374660491943,
      "learning_rate": 1.6609589041095892e-06,
      "loss": 2.0209,
      "step": 111
    },
    {
      "epoch": 0.1150487930148947,
      "grad_norm": 2.7660279273986816,
      "learning_rate": 1.678082191780822e-06,
      "loss": 2.1165,
      "step": 112
    },
    {
      "epoch": 0.11607601438109913,
      "grad_norm": 2.567336082458496,
      "learning_rate": 1.695205479452055e-06,
      "loss": 2.0529,
      "step": 113
    },
    {
      "epoch": 0.11710323574730354,
      "grad_norm": 2.4126152992248535,
      "learning_rate": 1.7123287671232877e-06,
      "loss": 2.1028,
      "step": 114
    },
    {
      "epoch": 0.11813045711350796,
      "grad_norm": 3.577528476715088,
      "learning_rate": 1.7294520547945206e-06,
      "loss": 2.0317,
      "step": 115
    },
    {
      "epoch": 0.11915767847971238,
      "grad_norm": 2.9624195098876953,
      "learning_rate": 1.7465753424657536e-06,
      "loss": 2.1318,
      "step": 116
    },
    {
      "epoch": 0.12018489984591679,
      "grad_norm": 2.789854049682617,
      "learning_rate": 1.7636986301369865e-06,
      "loss": 2.0822,
      "step": 117
    },
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 2.8168365955352783,
      "learning_rate": 1.7808219178082193e-06,
      "loss": 2.08,
      "step": 118
    },
    {
      "epoch": 0.12223934257832562,
      "grad_norm": 2.7616257667541504,
      "learning_rate": 1.7979452054794521e-06,
      "loss": 2.0668,
      "step": 119
    },
    {
      "epoch": 0.12326656394453005,
      "grad_norm": 2.5610315799713135,
      "learning_rate": 1.8150684931506852e-06,
      "loss": 2.1518,
      "step": 120
    },
    {
      "epoch": 0.12326656394453005,
      "eval_loss": 2.184033155441284,
      "eval_runtime": 37.3784,
      "eval_samples_per_second": 26.753,
      "eval_steps_per_second": 1.124,
      "step": 120
    },
    {
      "epoch": 0.12429378531073447,
      "grad_norm": 2.817063093185425,
      "learning_rate": 1.832191780821918e-06,
      "loss": 2.0813,
      "step": 121
    },
    {
      "epoch": 0.12532100667693888,
      "grad_norm": 2.987205982208252,
      "learning_rate": 1.8493150684931507e-06,
      "loss": 2.0421,
      "step": 122
    },
    {
      "epoch": 0.1263482280431433,
      "grad_norm": 2.8480496406555176,
      "learning_rate": 1.8664383561643837e-06,
      "loss": 2.0018,
      "step": 123
    },
    {
      "epoch": 0.12737544940934772,
      "grad_norm": 2.994274616241455,
      "learning_rate": 1.8835616438356166e-06,
      "loss": 2.0266,
      "step": 124
    },
    {
      "epoch": 0.12840267077555212,
      "grad_norm": 2.64136004447937,
      "learning_rate": 1.9006849315068496e-06,
      "loss": 1.9873,
      "step": 125
    },
    {
      "epoch": 0.12942989214175654,
      "grad_norm": 2.9426915645599365,
      "learning_rate": 1.9178082191780823e-06,
      "loss": 2.0566,
      "step": 126
    },
    {
      "epoch": 0.13045711350796096,
      "grad_norm": 2.9018619060516357,
      "learning_rate": 1.9349315068493153e-06,
      "loss": 1.9674,
      "step": 127
    },
    {
      "epoch": 0.1314843348741654,
      "grad_norm": 3.6921045780181885,
      "learning_rate": 1.952054794520548e-06,
      "loss": 2.0295,
      "step": 128
    },
    {
      "epoch": 0.1325115562403698,
      "grad_norm": 2.4194657802581787,
      "learning_rate": 1.969178082191781e-06,
      "loss": 2.0639,
      "step": 129
    },
    {
      "epoch": 0.1335387776065742,
      "grad_norm": 3.252349615097046,
      "learning_rate": 1.9863013698630136e-06,
      "loss": 2.0613,
      "step": 130
    },
    {
      "epoch": 0.1335387776065742,
      "eval_loss": 2.153157949447632,
      "eval_runtime": 37.394,
      "eval_samples_per_second": 26.742,
      "eval_steps_per_second": 1.123,
      "step": 130
    },
    {
      "epoch": 0.13456599897277863,
      "grad_norm": 2.9361023902893066,
      "learning_rate": 2.0034246575342467e-06,
      "loss": 1.9817,
      "step": 131
    },
    {
      "epoch": 0.13559322033898305,
      "grad_norm": 2.834803581237793,
      "learning_rate": 2.0205479452054797e-06,
      "loss": 2.0539,
      "step": 132
    },
    {
      "epoch": 0.13662044170518747,
      "grad_norm": 2.8814785480499268,
      "learning_rate": 2.037671232876713e-06,
      "loss": 2.0296,
      "step": 133
    },
    {
      "epoch": 0.1376476630713919,
      "grad_norm": 2.647125482559204,
      "learning_rate": 2.0547945205479454e-06,
      "loss": 1.9739,
      "step": 134
    },
    {
      "epoch": 0.1386748844375963,
      "grad_norm": 2.9608750343322754,
      "learning_rate": 2.071917808219178e-06,
      "loss": 2.0083,
      "step": 135
    },
    {
      "epoch": 0.1397021058038007,
      "grad_norm": 2.8570032119750977,
      "learning_rate": 2.089041095890411e-06,
      "loss": 2.0216,
      "step": 136
    },
    {
      "epoch": 0.14072932717000514,
      "grad_norm": 2.806358575820923,
      "learning_rate": 2.106164383561644e-06,
      "loss": 1.9457,
      "step": 137
    },
    {
      "epoch": 0.14175654853620956,
      "grad_norm": 3.1003353595733643,
      "learning_rate": 2.123287671232877e-06,
      "loss": 2.0324,
      "step": 138
    },
    {
      "epoch": 0.14278376990241398,
      "grad_norm": 2.679588556289673,
      "learning_rate": 2.14041095890411e-06,
      "loss": 1.9699,
      "step": 139
    },
    {
      "epoch": 0.14381099126861838,
      "grad_norm": 2.9436194896698,
      "learning_rate": 2.1575342465753425e-06,
      "loss": 1.9824,
      "step": 140
    },
    {
      "epoch": 0.14381099126861838,
      "eval_loss": 2.123953104019165,
      "eval_runtime": 37.4214,
      "eval_samples_per_second": 26.723,
      "eval_steps_per_second": 1.122,
      "step": 140
    },
    {
      "epoch": 0.1448382126348228,
      "grad_norm": 3.088301420211792,
      "learning_rate": 2.1746575342465755e-06,
      "loss": 1.9861,
      "step": 141
    },
    {
      "epoch": 0.14586543400102722,
      "grad_norm": 3.011859178543091,
      "learning_rate": 2.191780821917808e-06,
      "loss": 1.9383,
      "step": 142
    },
    {
      "epoch": 0.14689265536723164,
      "grad_norm": 2.930264472961426,
      "learning_rate": 2.2089041095890412e-06,
      "loss": 2.0674,
      "step": 143
    },
    {
      "epoch": 0.14791987673343607,
      "grad_norm": 3.361328601837158,
      "learning_rate": 2.2260273972602743e-06,
      "loss": 1.9495,
      "step": 144
    },
    {
      "epoch": 0.14894709809964046,
      "grad_norm": 2.7825112342834473,
      "learning_rate": 2.243150684931507e-06,
      "loss": 1.9335,
      "step": 145
    },
    {
      "epoch": 0.14997431946584489,
      "grad_norm": 3.7215847969055176,
      "learning_rate": 2.26027397260274e-06,
      "loss": 1.9403,
      "step": 146
    },
    {
      "epoch": 0.1510015408320493,
      "grad_norm": 2.5125460624694824,
      "learning_rate": 2.2773972602739726e-06,
      "loss": 1.9437,
      "step": 147
    },
    {
      "epoch": 0.15202876219825373,
      "grad_norm": 3.0963492393493652,
      "learning_rate": 2.2945205479452057e-06,
      "loss": 1.9651,
      "step": 148
    },
    {
      "epoch": 0.15305598356445815,
      "grad_norm": 2.9147121906280518,
      "learning_rate": 2.3116438356164387e-06,
      "loss": 2.0567,
      "step": 149
    },
    {
      "epoch": 0.15408320493066255,
      "grad_norm": 3.5275802612304688,
      "learning_rate": 2.3287671232876713e-06,
      "loss": 1.9773,
      "step": 150
    },
    {
      "epoch": 0.15408320493066255,
      "eval_loss": 2.112344264984131,
      "eval_runtime": 37.4142,
      "eval_samples_per_second": 26.728,
      "eval_steps_per_second": 1.123,
      "step": 150
    },
    {
      "epoch": 0.15511042629686697,
      "grad_norm": 3.013786554336548,
      "learning_rate": 2.3458904109589044e-06,
      "loss": 1.8753,
      "step": 151
    },
    {
      "epoch": 0.1561376476630714,
      "grad_norm": 3.433615207672119,
      "learning_rate": 2.363013698630137e-06,
      "loss": 1.9256,
      "step": 152
    },
    {
      "epoch": 0.15716486902927582,
      "grad_norm": 3.1526033878326416,
      "learning_rate": 2.38013698630137e-06,
      "loss": 1.9815,
      "step": 153
    },
    {
      "epoch": 0.15819209039548024,
      "grad_norm": 3.2099974155426025,
      "learning_rate": 2.3972602739726027e-06,
      "loss": 1.9628,
      "step": 154
    },
    {
      "epoch": 0.15921931176168463,
      "grad_norm": 3.2713584899902344,
      "learning_rate": 2.4143835616438358e-06,
      "loss": 1.9128,
      "step": 155
    },
    {
      "epoch": 0.16024653312788906,
      "grad_norm": 3.287649393081665,
      "learning_rate": 2.431506849315069e-06,
      "loss": 1.8764,
      "step": 156
    },
    {
      "epoch": 0.16127375449409348,
      "grad_norm": 3.1439101696014404,
      "learning_rate": 2.4486301369863015e-06,
      "loss": 1.9428,
      "step": 157
    },
    {
      "epoch": 0.1623009758602979,
      "grad_norm": 3.006457805633545,
      "learning_rate": 2.4657534246575345e-06,
      "loss": 1.9516,
      "step": 158
    },
    {
      "epoch": 0.1633281972265023,
      "grad_norm": 2.9567787647247314,
      "learning_rate": 2.482876712328767e-06,
      "loss": 1.915,
      "step": 159
    },
    {
      "epoch": 0.16435541859270672,
      "grad_norm": 3.2475216388702393,
      "learning_rate": 2.5e-06,
      "loss": 1.9451,
      "step": 160
    },
    {
      "epoch": 0.16435541859270672,
      "eval_loss": 2.086299419403076,
      "eval_runtime": 37.3939,
      "eval_samples_per_second": 26.742,
      "eval_steps_per_second": 1.123,
      "step": 160
    },
    {
      "epoch": 0.16538263995891114,
      "grad_norm": 2.9586498737335205,
      "learning_rate": 2.5171232876712333e-06,
      "loss": 1.9298,
      "step": 161
    },
    {
      "epoch": 0.16640986132511557,
      "grad_norm": 3.066910982131958,
      "learning_rate": 2.534246575342466e-06,
      "loss": 1.9103,
      "step": 162
    },
    {
      "epoch": 0.16743708269132,
      "grad_norm": 2.814741611480713,
      "learning_rate": 2.551369863013699e-06,
      "loss": 1.8678,
      "step": 163
    },
    {
      "epoch": 0.16846430405752438,
      "grad_norm": 3.7239339351654053,
      "learning_rate": 2.568493150684932e-06,
      "loss": 1.9321,
      "step": 164
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 2.829899787902832,
      "learning_rate": 2.585616438356164e-06,
      "loss": 1.9587,
      "step": 165
    },
    {
      "epoch": 0.17051874678993323,
      "grad_norm": 3.826289176940918,
      "learning_rate": 2.6027397260273973e-06,
      "loss": 1.9288,
      "step": 166
    },
    {
      "epoch": 0.17154596815613765,
      "grad_norm": 3.1028499603271484,
      "learning_rate": 2.6198630136986303e-06,
      "loss": 1.8685,
      "step": 167
    },
    {
      "epoch": 0.17257318952234207,
      "grad_norm": 3.6923508644104004,
      "learning_rate": 2.6369863013698634e-06,
      "loss": 1.9276,
      "step": 168
    },
    {
      "epoch": 0.17360041088854647,
      "grad_norm": 3.0577642917633057,
      "learning_rate": 2.654109589041096e-06,
      "loss": 1.9653,
      "step": 169
    },
    {
      "epoch": 0.1746276322547509,
      "grad_norm": 3.363762617111206,
      "learning_rate": 2.671232876712329e-06,
      "loss": 1.9465,
      "step": 170
    },
    {
      "epoch": 0.1746276322547509,
      "eval_loss": 2.0753531455993652,
      "eval_runtime": 37.3955,
      "eval_samples_per_second": 26.741,
      "eval_steps_per_second": 1.123,
      "step": 170
    },
    {
      "epoch": 0.17565485362095531,
      "grad_norm": 2.9063165187835693,
      "learning_rate": 2.688356164383562e-06,
      "loss": 1.8827,
      "step": 171
    },
    {
      "epoch": 0.17668207498715974,
      "grad_norm": 3.4709181785583496,
      "learning_rate": 2.705479452054795e-06,
      "loss": 1.882,
      "step": 172
    },
    {
      "epoch": 0.17770929635336416,
      "grad_norm": 3.395397663116455,
      "learning_rate": 2.7226027397260274e-06,
      "loss": 1.8711,
      "step": 173
    },
    {
      "epoch": 0.17873651771956856,
      "grad_norm": 3.9973363876342773,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 1.8223,
      "step": 174
    },
    {
      "epoch": 0.17976373908577298,
      "grad_norm": 3.1841156482696533,
      "learning_rate": 2.7568493150684935e-06,
      "loss": 1.947,
      "step": 175
    },
    {
      "epoch": 0.1807909604519774,
      "grad_norm": 3.1841156482696533,
      "learning_rate": 2.7568493150684935e-06,
      "loss": 1.9192,
      "step": 176
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 3.84712290763855,
      "learning_rate": 2.773972602739726e-06,
      "loss": 1.8344,
      "step": 177
    },
    {
      "epoch": 0.18284540318438625,
      "grad_norm": 3.487525463104248,
      "learning_rate": 2.791095890410959e-06,
      "loss": 1.9242,
      "step": 178
    },
    {
      "epoch": 0.18387262455059064,
      "grad_norm": 3.1980319023132324,
      "learning_rate": 2.8082191780821922e-06,
      "loss": 1.7959,
      "step": 179
    },
    {
      "epoch": 0.18489984591679506,
      "grad_norm": 3.56462025642395,
      "learning_rate": 2.8253424657534253e-06,
      "loss": 1.9668,
      "step": 180
    },
    {
      "epoch": 0.18489984591679506,
      "eval_loss": 2.0606908798217773,
      "eval_runtime": 37.3994,
      "eval_samples_per_second": 26.738,
      "eval_steps_per_second": 1.123,
      "step": 180
    },
    {
      "epoch": 0.1859270672829995,
      "grad_norm": 3.613842248916626,
      "learning_rate": 2.842465753424658e-06,
      "loss": 1.925,
      "step": 181
    },
    {
      "epoch": 0.1869542886492039,
      "grad_norm": 2.9755642414093018,
      "learning_rate": 2.8595890410958905e-06,
      "loss": 1.9251,
      "step": 182
    },
    {
      "epoch": 0.18798151001540833,
      "grad_norm": 2.9678146839141846,
      "learning_rate": 2.876712328767123e-06,
      "loss": 1.9406,
      "step": 183
    },
    {
      "epoch": 0.18900873138161273,
      "grad_norm": 3.3536393642425537,
      "learning_rate": 2.8938356164383562e-06,
      "loss": 1.8041,
      "step": 184
    },
    {
      "epoch": 0.19003595274781715,
      "grad_norm": 2.714902877807617,
      "learning_rate": 2.9109589041095893e-06,
      "loss": 1.8213,
      "step": 185
    },
    {
      "epoch": 0.19106317411402157,
      "grad_norm": 3.0686237812042236,
      "learning_rate": 2.9280821917808223e-06,
      "loss": 1.817,
      "step": 186
    },
    {
      "epoch": 0.192090395480226,
      "grad_norm": 3.510448932647705,
      "learning_rate": 2.945205479452055e-06,
      "loss": 1.8317,
      "step": 187
    },
    {
      "epoch": 0.19311761684643042,
      "grad_norm": 3.2060787677764893,
      "learning_rate": 2.962328767123288e-06,
      "loss": 1.894,
      "step": 188
    },
    {
      "epoch": 0.1941448382126348,
      "grad_norm": 3.1756906509399414,
      "learning_rate": 2.979452054794521e-06,
      "loss": 1.8788,
      "step": 189
    },
    {
      "epoch": 0.19517205957883924,
      "grad_norm": 3.059863805770874,
      "learning_rate": 2.9965753424657533e-06,
      "loss": 1.8198,
      "step": 190
    },
    {
      "epoch": 0.19517205957883924,
      "eval_loss": 2.0506513118743896,
      "eval_runtime": 37.4014,
      "eval_samples_per_second": 26.737,
      "eval_steps_per_second": 1.123,
      "step": 190
    },
    {
      "epoch": 0.19619928094504366,
      "grad_norm": 3.193685531616211,
      "learning_rate": 3.0136986301369864e-06,
      "loss": 1.8818,
      "step": 191
    },
    {
      "epoch": 0.19722650231124808,
      "grad_norm": 2.905402183532715,
      "learning_rate": 3.0308219178082194e-06,
      "loss": 1.8546,
      "step": 192
    },
    {
      "epoch": 0.1982537236774525,
      "grad_norm": 3.423109531402588,
      "learning_rate": 3.0479452054794525e-06,
      "loss": 1.8142,
      "step": 193
    },
    {
      "epoch": 0.1992809450436569,
      "grad_norm": 3.136094331741333,
      "learning_rate": 3.065068493150685e-06,
      "loss": 1.8553,
      "step": 194
    },
    {
      "epoch": 0.20030816640986132,
      "grad_norm": 3.0746538639068604,
      "learning_rate": 3.082191780821918e-06,
      "loss": 1.8153,
      "step": 195
    },
    {
      "epoch": 0.20133538777606574,
      "grad_norm": 2.9354681968688965,
      "learning_rate": 3.099315068493151e-06,
      "loss": 1.8301,
      "step": 196
    },
    {
      "epoch": 0.20236260914227017,
      "grad_norm": 3.00426983833313,
      "learning_rate": 3.1164383561643843e-06,
      "loss": 1.8233,
      "step": 197
    },
    {
      "epoch": 0.2033898305084746,
      "grad_norm": 3.2754054069519043,
      "learning_rate": 3.1335616438356165e-06,
      "loss": 1.787,
      "step": 198
    },
    {
      "epoch": 0.20441705187467898,
      "grad_norm": 2.9854509830474854,
      "learning_rate": 3.1506849315068495e-06,
      "loss": 1.9341,
      "step": 199
    },
    {
      "epoch": 0.2054442732408834,
      "grad_norm": 3.5655088424682617,
      "learning_rate": 3.167808219178082e-06,
      "loss": 1.7964,
      "step": 200
    },
    {
      "epoch": 0.2054442732408834,
      "eval_loss": 2.0440263748168945,
      "eval_runtime": 37.4054,
      "eval_samples_per_second": 26.734,
      "eval_steps_per_second": 1.123,
      "step": 200
    },
    {
      "epoch": 0.20647149460708783,
      "grad_norm": 3.132373571395874,
      "learning_rate": 3.184931506849315e-06,
      "loss": 1.8019,
      "step": 201
    },
    {
      "epoch": 0.20749871597329225,
      "grad_norm": 3.3017780780792236,
      "learning_rate": 3.2020547945205483e-06,
      "loss": 1.7786,
      "step": 202
    },
    {
      "epoch": 0.20852593733949668,
      "grad_norm": 3.009382486343384,
      "learning_rate": 3.2191780821917813e-06,
      "loss": 1.7639,
      "step": 203
    },
    {
      "epoch": 0.20955315870570107,
      "grad_norm": 3.5784854888916016,
      "learning_rate": 3.236301369863014e-06,
      "loss": 1.76,
      "step": 204
    },
    {
      "epoch": 0.2105803800719055,
      "grad_norm": 3.236858367919922,
      "learning_rate": 3.253424657534247e-06,
      "loss": 1.7295,
      "step": 205
    },
    {
      "epoch": 0.21160760143810992,
      "grad_norm": 3.6157479286193848,
      "learning_rate": 3.2705479452054796e-06,
      "loss": 1.8287,
      "step": 206
    },
    {
      "epoch": 0.21263482280431434,
      "grad_norm": 3.51790452003479,
      "learning_rate": 3.2876712328767123e-06,
      "loss": 1.8004,
      "step": 207
    },
    {
      "epoch": 0.21366204417051873,
      "grad_norm": 3.0700764656066895,
      "learning_rate": 3.3047945205479453e-06,
      "loss": 1.7581,
      "step": 208
    },
    {
      "epoch": 0.21468926553672316,
      "grad_norm": 3.2651591300964355,
      "learning_rate": 3.3219178082191784e-06,
      "loss": 1.7724,
      "step": 209
    },
    {
      "epoch": 0.21571648690292758,
      "grad_norm": 2.9267404079437256,
      "learning_rate": 3.3390410958904114e-06,
      "loss": 1.8154,
      "step": 210
    },
    {
      "epoch": 0.21571648690292758,
      "eval_loss": 2.0323731899261475,
      "eval_runtime": 37.2156,
      "eval_samples_per_second": 26.87,
      "eval_steps_per_second": 1.129,
      "step": 210
    },
    {
      "epoch": 0.216743708269132,
      "grad_norm": 3.463411808013916,
      "learning_rate": 3.356164383561644e-06,
      "loss": 1.8309,
      "step": 211
    },
    {
      "epoch": 0.21777092963533642,
      "grad_norm": 3.123573064804077,
      "learning_rate": 3.373287671232877e-06,
      "loss": 1.8793,
      "step": 212
    },
    {
      "epoch": 0.21879815100154082,
      "grad_norm": 3.427051305770874,
      "learning_rate": 3.39041095890411e-06,
      "loss": 1.8259,
      "step": 213
    },
    {
      "epoch": 0.21982537236774524,
      "grad_norm": 3.306366443634033,
      "learning_rate": 3.4075342465753424e-06,
      "loss": 1.8565,
      "step": 214
    },
    {
      "epoch": 0.22085259373394966,
      "grad_norm": 3.557340383529663,
      "learning_rate": 3.4246575342465754e-06,
      "loss": 1.7531,
      "step": 215
    },
    {
      "epoch": 0.2218798151001541,
      "grad_norm": 3.20525860786438,
      "learning_rate": 3.4417808219178085e-06,
      "loss": 1.8032,
      "step": 216
    },
    {
      "epoch": 0.2229070364663585,
      "grad_norm": 3.2657511234283447,
      "learning_rate": 3.458904109589041e-06,
      "loss": 1.8026,
      "step": 217
    },
    {
      "epoch": 0.2239342578325629,
      "grad_norm": 3.1869699954986572,
      "learning_rate": 3.476027397260274e-06,
      "loss": 1.814,
      "step": 218
    },
    {
      "epoch": 0.22496147919876733,
      "grad_norm": 2.661536931991577,
      "learning_rate": 3.4931506849315072e-06,
      "loss": 1.8005,
      "step": 219
    },
    {
      "epoch": 0.22598870056497175,
      "grad_norm": 3.1108739376068115,
      "learning_rate": 3.5102739726027403e-06,
      "loss": 1.8095,
      "step": 220
    },
    {
      "epoch": 0.22598870056497175,
      "eval_loss": 2.015268325805664,
      "eval_runtime": 37.2867,
      "eval_samples_per_second": 26.819,
      "eval_steps_per_second": 1.126,
      "step": 220
    },
    {
      "epoch": 0.22701592193117617,
      "grad_norm": 2.8078837394714355,
      "learning_rate": 3.527397260273973e-06,
      "loss": 1.8016,
      "step": 221
    },
    {
      "epoch": 0.2280431432973806,
      "grad_norm": 3.131399631500244,
      "learning_rate": 3.5445205479452056e-06,
      "loss": 1.7787,
      "step": 222
    },
    {
      "epoch": 0.229070364663585,
      "grad_norm": 3.114422559738159,
      "learning_rate": 3.5616438356164386e-06,
      "loss": 1.7899,
      "step": 223
    },
    {
      "epoch": 0.2300975860297894,
      "grad_norm": 3.040816068649292,
      "learning_rate": 3.5787671232876712e-06,
      "loss": 1.8601,
      "step": 224
    },
    {
      "epoch": 0.23112480739599384,
      "grad_norm": 3.043100357055664,
      "learning_rate": 3.5958904109589043e-06,
      "loss": 1.8244,
      "step": 225
    },
    {
      "epoch": 0.23215202876219826,
      "grad_norm": 3.4232981204986572,
      "learning_rate": 3.6130136986301373e-06,
      "loss": 1.7634,
      "step": 226
    },
    {
      "epoch": 0.23317925012840268,
      "grad_norm": 2.8170571327209473,
      "learning_rate": 3.6301369863013704e-06,
      "loss": 1.7793,
      "step": 227
    },
    {
      "epoch": 0.23420647149460708,
      "grad_norm": 3.9133572578430176,
      "learning_rate": 3.647260273972603e-06,
      "loss": 1.8239,
      "step": 228
    },
    {
      "epoch": 0.2352336928608115,
      "grad_norm": 2.7053382396698,
      "learning_rate": 3.664383561643836e-06,
      "loss": 1.8209,
      "step": 229
    },
    {
      "epoch": 0.23626091422701592,
      "grad_norm": 3.4366891384124756,
      "learning_rate": 3.6815068493150687e-06,
      "loss": 1.7539,
      "step": 230
    },
    {
      "epoch": 0.23626091422701592,
      "eval_loss": 2.0176751613616943,
      "eval_runtime": 37.3064,
      "eval_samples_per_second": 26.805,
      "eval_steps_per_second": 1.126,
      "step": 230
    },
    {
      "epoch": 0.23728813559322035,
      "grad_norm": 2.592280149459839,
      "learning_rate": 3.6986301369863014e-06,
      "loss": 1.7858,
      "step": 231
    },
    {
      "epoch": 0.23831535695942477,
      "grad_norm": 2.8370869159698486,
      "learning_rate": 3.7157534246575344e-06,
      "loss": 1.7414,
      "step": 232
    },
    {
      "epoch": 0.23934257832562916,
      "grad_norm": 3.2120535373687744,
      "learning_rate": 3.7328767123287675e-06,
      "loss": 1.8467,
      "step": 233
    },
    {
      "epoch": 0.24036979969183359,
      "grad_norm": 2.9314255714416504,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 1.7602,
      "step": 234
    },
    {
      "epoch": 0.241397021058038,
      "grad_norm": 3.056293249130249,
      "learning_rate": 3.767123287671233e-06,
      "loss": 1.7972,
      "step": 235
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 2.5918917655944824,
      "learning_rate": 3.784246575342466e-06,
      "loss": 1.8274,
      "step": 236
    },
    {
      "epoch": 0.24345146379044685,
      "grad_norm": 2.7911906242370605,
      "learning_rate": 3.8013698630136993e-06,
      "loss": 1.753,
      "step": 237
    },
    {
      "epoch": 0.24447868515665125,
      "grad_norm": 2.810612916946411,
      "learning_rate": 3.818493150684932e-06,
      "loss": 1.7822,
      "step": 238
    },
    {
      "epoch": 0.24550590652285567,
      "grad_norm": 2.8863744735717773,
      "learning_rate": 3.8356164383561645e-06,
      "loss": 1.7258,
      "step": 239
    },
    {
      "epoch": 0.2465331278890601,
      "grad_norm": 3.002087116241455,
      "learning_rate": 3.852739726027397e-06,
      "loss": 1.8423,
      "step": 240
    },
    {
      "epoch": 0.2465331278890601,
      "eval_loss": 2.020223617553711,
      "eval_runtime": 37.3197,
      "eval_samples_per_second": 26.796,
      "eval_steps_per_second": 1.125,
      "step": 240
    },
    {
      "epoch": 0.24756034925526452,
      "grad_norm": 2.892317056655884,
      "learning_rate": 3.869863013698631e-06,
      "loss": 1.8181,
      "step": 241
    },
    {
      "epoch": 0.24858757062146894,
      "grad_norm": 2.895620584487915,
      "learning_rate": 3.886986301369863e-06,
      "loss": 1.7538,
      "step": 242
    },
    {
      "epoch": 0.24961479198767333,
      "grad_norm": 3.015660524368286,
      "learning_rate": 3.904109589041096e-06,
      "loss": 1.766,
      "step": 243
    },
    {
      "epoch": 0.25064201335387776,
      "grad_norm": 2.7516112327575684,
      "learning_rate": 3.921232876712329e-06,
      "loss": 1.7324,
      "step": 244
    },
    {
      "epoch": 0.2516692347200822,
      "grad_norm": 2.937777519226074,
      "learning_rate": 3.938356164383562e-06,
      "loss": 1.7468,
      "step": 245
    },
    {
      "epoch": 0.2526964560862866,
      "grad_norm": 3.1706550121307373,
      "learning_rate": 3.9554794520547955e-06,
      "loss": 1.7035,
      "step": 246
    },
    {
      "epoch": 0.253723677452491,
      "grad_norm": 2.7795372009277344,
      "learning_rate": 3.972602739726027e-06,
      "loss": 1.7912,
      "step": 247
    },
    {
      "epoch": 0.25475089881869545,
      "grad_norm": 3.0375940799713135,
      "learning_rate": 3.989726027397261e-06,
      "loss": 1.7208,
      "step": 248
    },
    {
      "epoch": 0.25577812018489987,
      "grad_norm": 2.863913059234619,
      "learning_rate": 4.006849315068493e-06,
      "loss": 1.7891,
      "step": 249
    },
    {
      "epoch": 0.25680534155110424,
      "grad_norm": 3.02720046043396,
      "learning_rate": 4.023972602739726e-06,
      "loss": 1.7864,
      "step": 250
    },
    {
      "epoch": 0.25680534155110424,
      "eval_loss": 1.9981943368911743,
      "eval_runtime": 37.3019,
      "eval_samples_per_second": 26.808,
      "eval_steps_per_second": 1.126,
      "step": 250
    },
    {
      "epoch": 0.25783256291730866,
      "grad_norm": 2.90144681930542,
      "learning_rate": 4.0410958904109595e-06,
      "loss": 1.7508,
      "step": 251
    },
    {
      "epoch": 0.2588597842835131,
      "grad_norm": 2.7670342922210693,
      "learning_rate": 4.058219178082192e-06,
      "loss": 1.7668,
      "step": 252
    },
    {
      "epoch": 0.2598870056497175,
      "grad_norm": 3.0309600830078125,
      "learning_rate": 4.075342465753426e-06,
      "loss": 1.7853,
      "step": 253
    },
    {
      "epoch": 0.26091422701592193,
      "grad_norm": 2.995190143585205,
      "learning_rate": 4.092465753424658e-06,
      "loss": 1.7409,
      "step": 254
    },
    {
      "epoch": 0.26194144838212635,
      "grad_norm": 3.127767324447632,
      "learning_rate": 4.109589041095891e-06,
      "loss": 1.743,
      "step": 255
    },
    {
      "epoch": 0.2629686697483308,
      "grad_norm": 3.7104549407958984,
      "learning_rate": 4.1267123287671235e-06,
      "loss": 1.8149,
      "step": 256
    },
    {
      "epoch": 0.2639958911145352,
      "grad_norm": 3.042569160461426,
      "learning_rate": 4.143835616438356e-06,
      "loss": 1.7544,
      "step": 257
    },
    {
      "epoch": 0.2650231124807396,
      "grad_norm": 2.9337198734283447,
      "learning_rate": 4.16095890410959e-06,
      "loss": 1.823,
      "step": 258
    },
    {
      "epoch": 0.26605033384694404,
      "grad_norm": 3.1853089332580566,
      "learning_rate": 4.178082191780822e-06,
      "loss": 1.8002,
      "step": 259
    },
    {
      "epoch": 0.2670775552131484,
      "grad_norm": 2.9620680809020996,
      "learning_rate": 4.195205479452055e-06,
      "loss": 1.7401,
      "step": 260
    },
    {
      "epoch": 0.2670775552131484,
      "eval_loss": 1.9844669103622437,
      "eval_runtime": 37.3162,
      "eval_samples_per_second": 26.798,
      "eval_steps_per_second": 1.126,
      "step": 260
    },
    {
      "epoch": 0.26810477657935283,
      "grad_norm": 3.477985382080078,
      "learning_rate": 4.212328767123288e-06,
      "loss": 1.7456,
      "step": 261
    },
    {
      "epoch": 0.26913199794555726,
      "grad_norm": 2.881277084350586,
      "learning_rate": 4.229452054794521e-06,
      "loss": 1.7427,
      "step": 262
    },
    {
      "epoch": 0.2701592193117617,
      "grad_norm": 3.3313355445861816,
      "learning_rate": 4.246575342465754e-06,
      "loss": 1.7537,
      "step": 263
    },
    {
      "epoch": 0.2711864406779661,
      "grad_norm": 2.947119951248169,
      "learning_rate": 4.263698630136986e-06,
      "loss": 1.7349,
      "step": 264
    },
    {
      "epoch": 0.2722136620441705,
      "grad_norm": 3.0780889987945557,
      "learning_rate": 4.28082191780822e-06,
      "loss": 1.694,
      "step": 265
    },
    {
      "epoch": 0.27324088341037495,
      "grad_norm": 3.3371143341064453,
      "learning_rate": 4.297945205479452e-06,
      "loss": 1.8489,
      "step": 266
    },
    {
      "epoch": 0.27426810477657937,
      "grad_norm": 3.01694917678833,
      "learning_rate": 4.315068493150685e-06,
      "loss": 1.7478,
      "step": 267
    },
    {
      "epoch": 0.2752953261427838,
      "grad_norm": 3.001850128173828,
      "learning_rate": 4.3321917808219185e-06,
      "loss": 1.7298,
      "step": 268
    },
    {
      "epoch": 0.2763225475089882,
      "grad_norm": 2.9096696376800537,
      "learning_rate": 4.349315068493151e-06,
      "loss": 1.7542,
      "step": 269
    },
    {
      "epoch": 0.2773497688751926,
      "grad_norm": 3.209601402282715,
      "learning_rate": 4.3664383561643846e-06,
      "loss": 1.7248,
      "step": 270
    },
    {
      "epoch": 0.2773497688751926,
      "eval_loss": 1.9732507467269897,
      "eval_runtime": 37.3151,
      "eval_samples_per_second": 26.799,
      "eval_steps_per_second": 1.126,
      "step": 270
    },
    {
      "epoch": 0.278376990241397,
      "grad_norm": 2.9635558128356934,
      "learning_rate": 4.383561643835616e-06,
      "loss": 1.727,
      "step": 271
    },
    {
      "epoch": 0.2794042116076014,
      "grad_norm": 3.143821954727173,
      "learning_rate": 4.40068493150685e-06,
      "loss": 1.8045,
      "step": 272
    },
    {
      "epoch": 0.28043143297380585,
      "grad_norm": 3.4248714447021484,
      "learning_rate": 4.4178082191780825e-06,
      "loss": 1.7649,
      "step": 273
    },
    {
      "epoch": 0.2814586543400103,
      "grad_norm": 2.902162790298462,
      "learning_rate": 4.434931506849315e-06,
      "loss": 1.7324,
      "step": 274
    },
    {
      "epoch": 0.2824858757062147,
      "grad_norm": 2.8899712562561035,
      "learning_rate": 4.4520547945205486e-06,
      "loss": 1.6886,
      "step": 275
    },
    {
      "epoch": 0.2835130970724191,
      "grad_norm": 3.2267258167266846,
      "learning_rate": 4.469178082191781e-06,
      "loss": 1.7442,
      "step": 276
    },
    {
      "epoch": 0.28454031843862354,
      "grad_norm": 3.275735378265381,
      "learning_rate": 4.486301369863014e-06,
      "loss": 1.7334,
      "step": 277
    },
    {
      "epoch": 0.28556753980482796,
      "grad_norm": 2.991509199142456,
      "learning_rate": 4.503424657534247e-06,
      "loss": 1.7853,
      "step": 278
    },
    {
      "epoch": 0.28659476117103233,
      "grad_norm": 2.8380391597747803,
      "learning_rate": 4.52054794520548e-06,
      "loss": 1.6885,
      "step": 279
    },
    {
      "epoch": 0.28762198253723675,
      "grad_norm": 2.909980297088623,
      "learning_rate": 4.537671232876713e-06,
      "loss": 1.7146,
      "step": 280
    },
    {
      "epoch": 0.28762198253723675,
      "eval_loss": 1.9755980968475342,
      "eval_runtime": 37.3524,
      "eval_samples_per_second": 26.772,
      "eval_steps_per_second": 1.124,
      "step": 280
    },
    {
      "epoch": 0.2886492039034412,
      "grad_norm": 3.2689175605773926,
      "learning_rate": 4.554794520547945e-06,
      "loss": 1.7747,
      "step": 281
    },
    {
      "epoch": 0.2896764252696456,
      "grad_norm": 2.8242290019989014,
      "learning_rate": 4.571917808219179e-06,
      "loss": 1.7657,
      "step": 282
    },
    {
      "epoch": 0.29070364663585,
      "grad_norm": 3.03828501701355,
      "learning_rate": 4.589041095890411e-06,
      "loss": 1.7555,
      "step": 283
    },
    {
      "epoch": 0.29173086800205444,
      "grad_norm": 2.714351177215576,
      "learning_rate": 4.606164383561644e-06,
      "loss": 1.7055,
      "step": 284
    },
    {
      "epoch": 0.29275808936825887,
      "grad_norm": 2.9540348052978516,
      "learning_rate": 4.6232876712328774e-06,
      "loss": 1.7258,
      "step": 285
    },
    {
      "epoch": 0.2937853107344633,
      "grad_norm": 3.107060670852661,
      "learning_rate": 4.64041095890411e-06,
      "loss": 1.7376,
      "step": 286
    },
    {
      "epoch": 0.2948125321006677,
      "grad_norm": 3.3266639709472656,
      "learning_rate": 4.657534246575343e-06,
      "loss": 1.7063,
      "step": 287
    },
    {
      "epoch": 0.29583975346687214,
      "grad_norm": 3.3113174438476562,
      "learning_rate": 4.674657534246575e-06,
      "loss": 1.7514,
      "step": 288
    },
    {
      "epoch": 0.2968669748330765,
      "grad_norm": 3.313321352005005,
      "learning_rate": 4.691780821917809e-06,
      "loss": 1.73,
      "step": 289
    },
    {
      "epoch": 0.2978941961992809,
      "grad_norm": 2.901522159576416,
      "learning_rate": 4.7089041095890414e-06,
      "loss": 1.7709,
      "step": 290
    },
    {
      "epoch": 0.2978941961992809,
      "eval_loss": 1.9608837366104126,
      "eval_runtime": 37.3687,
      "eval_samples_per_second": 26.76,
      "eval_steps_per_second": 1.124,
      "step": 290
    },
    {
      "epoch": 0.29892141756548535,
      "grad_norm": 3.4208264350891113,
      "learning_rate": 4.726027397260274e-06,
      "loss": 1.73,
      "step": 291
    },
    {
      "epoch": 0.29994863893168977,
      "grad_norm": 2.8982441425323486,
      "learning_rate": 4.7431506849315075e-06,
      "loss": 1.7416,
      "step": 292
    },
    {
      "epoch": 0.3009758602978942,
      "grad_norm": 3.2743732929229736,
      "learning_rate": 4.76027397260274e-06,
      "loss": 1.7443,
      "step": 293
    },
    {
      "epoch": 0.3020030816640986,
      "grad_norm": 3.028764486312866,
      "learning_rate": 4.777397260273973e-06,
      "loss": 1.6879,
      "step": 294
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 3.3531785011291504,
      "learning_rate": 4.7945205479452054e-06,
      "loss": 1.7096,
      "step": 295
    },
    {
      "epoch": 0.30405752439650746,
      "grad_norm": 2.5602540969848633,
      "learning_rate": 4.811643835616439e-06,
      "loss": 1.7125,
      "step": 296
    },
    {
      "epoch": 0.3050847457627119,
      "grad_norm": 3.1823129653930664,
      "learning_rate": 4.8287671232876716e-06,
      "loss": 1.6531,
      "step": 297
    },
    {
      "epoch": 0.3061119671289163,
      "grad_norm": 2.6700282096862793,
      "learning_rate": 4.845890410958904e-06,
      "loss": 1.6983,
      "step": 298
    },
    {
      "epoch": 0.3071391884951207,
      "grad_norm": 2.7118732929229736,
      "learning_rate": 4.863013698630138e-06,
      "loss": 1.7056,
      "step": 299
    },
    {
      "epoch": 0.3081664098613251,
      "grad_norm": 3.3493311405181885,
      "learning_rate": 4.88013698630137e-06,
      "loss": 1.6944,
      "step": 300
    },
    {
      "epoch": 0.3081664098613251,
      "eval_loss": 1.9489911794662476,
      "eval_runtime": 37.35,
      "eval_samples_per_second": 26.774,
      "eval_steps_per_second": 1.124,
      "step": 300
    },
    {
      "epoch": 0.3091936312275295,
      "grad_norm": 2.9436538219451904,
      "learning_rate": 4.897260273972603e-06,
      "loss": 1.6717,
      "step": 301
    },
    {
      "epoch": 0.31022085259373394,
      "grad_norm": 2.894737482070923,
      "learning_rate": 4.914383561643836e-06,
      "loss": 1.7895,
      "step": 302
    },
    {
      "epoch": 0.31124807395993837,
      "grad_norm": 2.8645665645599365,
      "learning_rate": 4.931506849315069e-06,
      "loss": 1.721,
      "step": 303
    },
    {
      "epoch": 0.3122752953261428,
      "grad_norm": 3.050203323364258,
      "learning_rate": 4.948630136986302e-06,
      "loss": 1.7564,
      "step": 304
    },
    {
      "epoch": 0.3133025166923472,
      "grad_norm": 3.146395206451416,
      "learning_rate": 4.965753424657534e-06,
      "loss": 1.7938,
      "step": 305
    },
    {
      "epoch": 0.31432973805855163,
      "grad_norm": 3.1021251678466797,
      "learning_rate": 4.982876712328768e-06,
      "loss": 1.6682,
      "step": 306
    },
    {
      "epoch": 0.31535695942475606,
      "grad_norm": 2.7301125526428223,
      "learning_rate": 5e-06,
      "loss": 1.6974,
      "step": 307
    },
    {
      "epoch": 0.3163841807909605,
      "grad_norm": 2.7602250576019287,
      "learning_rate": 4.998096688237534e-06,
      "loss": 1.6523,
      "step": 308
    },
    {
      "epoch": 0.31741140215716485,
      "grad_norm": 2.8149876594543457,
      "learning_rate": 4.996193376475067e-06,
      "loss": 1.7535,
      "step": 309
    },
    {
      "epoch": 0.31843862352336927,
      "grad_norm": 2.54823637008667,
      "learning_rate": 4.9942900647126005e-06,
      "loss": 1.63,
      "step": 310
    },
    {
      "epoch": 0.31843862352336927,
      "eval_loss": 1.928893804550171,
      "eval_runtime": 37.2236,
      "eval_samples_per_second": 26.865,
      "eval_steps_per_second": 1.128,
      "step": 310
    },
    {
      "epoch": 0.3194658448895737,
      "grad_norm": 2.7819766998291016,
      "learning_rate": 4.992386752950134e-06,
      "loss": 1.6461,
      "step": 311
    },
    {
      "epoch": 0.3204930662557781,
      "grad_norm": 2.999209403991699,
      "learning_rate": 4.990483441187667e-06,
      "loss": 1.6861,
      "step": 312
    },
    {
      "epoch": 0.32152028762198254,
      "grad_norm": 2.7288851737976074,
      "learning_rate": 4.9885801294252e-06,
      "loss": 1.7076,
      "step": 313
    },
    {
      "epoch": 0.32254750898818696,
      "grad_norm": 2.8722457885742188,
      "learning_rate": 4.986676817662733e-06,
      "loss": 1.7079,
      "step": 314
    },
    {
      "epoch": 0.3235747303543914,
      "grad_norm": 2.853344678878784,
      "learning_rate": 4.9847735059002665e-06,
      "loss": 1.6699,
      "step": 315
    },
    {
      "epoch": 0.3246019517205958,
      "grad_norm": 2.7404720783233643,
      "learning_rate": 4.9828701941378e-06,
      "loss": 1.7423,
      "step": 316
    },
    {
      "epoch": 0.3256291730868002,
      "grad_norm": 2.6062633991241455,
      "learning_rate": 4.980966882375333e-06,
      "loss": 1.6811,
      "step": 317
    },
    {
      "epoch": 0.3266563944530046,
      "grad_norm": 2.834162712097168,
      "learning_rate": 4.979063570612867e-06,
      "loss": 1.6575,
      "step": 318
    },
    {
      "epoch": 0.327683615819209,
      "grad_norm": 2.7741048336029053,
      "learning_rate": 4.9771602588504e-06,
      "loss": 1.6158,
      "step": 319
    },
    {
      "epoch": 0.32871083718541344,
      "grad_norm": 2.6216888427734375,
      "learning_rate": 4.975256947087933e-06,
      "loss": 1.7295,
      "step": 320
    },
    {
      "epoch": 0.32871083718541344,
      "eval_loss": 1.9309543371200562,
      "eval_runtime": 37.3489,
      "eval_samples_per_second": 26.775,
      "eval_steps_per_second": 1.125,
      "step": 320
    },
    {
      "epoch": 0.32973805855161786,
      "grad_norm": 2.6282548904418945,
      "learning_rate": 4.973353635325467e-06,
      "loss": 1.6266,
      "step": 321
    },
    {
      "epoch": 0.3307652799178223,
      "grad_norm": 2.632124662399292,
      "learning_rate": 4.971450323563e-06,
      "loss": 1.6887,
      "step": 322
    },
    {
      "epoch": 0.3317925012840267,
      "grad_norm": 2.5844857692718506,
      "learning_rate": 4.969547011800533e-06,
      "loss": 1.6287,
      "step": 323
    },
    {
      "epoch": 0.33281972265023113,
      "grad_norm": 2.597620964050293,
      "learning_rate": 4.967643700038067e-06,
      "loss": 1.6876,
      "step": 324
    },
    {
      "epoch": 0.33384694401643555,
      "grad_norm": 2.6529343128204346,
      "learning_rate": 4.9657403882756e-06,
      "loss": 1.6883,
      "step": 325
    },
    {
      "epoch": 0.33487416538264,
      "grad_norm": 2.592874526977539,
      "learning_rate": 4.9638370765131335e-06,
      "loss": 1.653,
      "step": 326
    },
    {
      "epoch": 0.3359013867488444,
      "grad_norm": 2.523489236831665,
      "learning_rate": 4.961933764750667e-06,
      "loss": 1.6501,
      "step": 327
    },
    {
      "epoch": 0.33692860811504877,
      "grad_norm": 2.6721699237823486,
      "learning_rate": 4.9600304529882e-06,
      "loss": 1.7126,
      "step": 328
    },
    {
      "epoch": 0.3379558294812532,
      "grad_norm": 2.513122797012329,
      "learning_rate": 4.9581271412257335e-06,
      "loss": 1.7268,
      "step": 329
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 2.571118116378784,
      "learning_rate": 4.956223829463266e-06,
      "loss": 1.6862,
      "step": 330
    },
    {
      "epoch": 0.3389830508474576,
      "eval_loss": 1.9501315355300903,
      "eval_runtime": 37.3287,
      "eval_samples_per_second": 26.789,
      "eval_steps_per_second": 1.125,
      "step": 330
    },
    {
      "epoch": 0.34001027221366203,
      "grad_norm": 2.5783371925354004,
      "learning_rate": 4.954320517700799e-06,
      "loss": 1.6041,
      "step": 331
    },
    {
      "epoch": 0.34103749357986646,
      "grad_norm": 2.6829233169555664,
      "learning_rate": 4.952417205938333e-06,
      "loss": 1.7111,
      "step": 332
    },
    {
      "epoch": 0.3420647149460709,
      "grad_norm": 2.6643247604370117,
      "learning_rate": 4.950513894175866e-06,
      "loss": 1.6797,
      "step": 333
    },
    {
      "epoch": 0.3430919363122753,
      "grad_norm": 2.776308298110962,
      "learning_rate": 4.9486105824133995e-06,
      "loss": 1.6766,
      "step": 334
    },
    {
      "epoch": 0.3441191576784797,
      "grad_norm": 2.722667932510376,
      "learning_rate": 4.946707270650933e-06,
      "loss": 1.6865,
      "step": 335
    },
    {
      "epoch": 0.34514637904468415,
      "grad_norm": 2.7050845623016357,
      "learning_rate": 4.944803958888466e-06,
      "loss": 1.6967,
      "step": 336
    },
    {
      "epoch": 0.34617360041088857,
      "grad_norm": 3.0009098052978516,
      "learning_rate": 4.942900647126e-06,
      "loss": 1.714,
      "step": 337
    },
    {
      "epoch": 0.34720082177709294,
      "grad_norm": 2.793621301651001,
      "learning_rate": 4.940997335363533e-06,
      "loss": 1.6591,
      "step": 338
    },
    {
      "epoch": 0.34822804314329736,
      "grad_norm": 2.844383478164673,
      "learning_rate": 4.939094023601066e-06,
      "loss": 1.611,
      "step": 339
    },
    {
      "epoch": 0.3492552645095018,
      "grad_norm": 2.829378366470337,
      "learning_rate": 4.9371907118386e-06,
      "loss": 1.666,
      "step": 340
    },
    {
      "epoch": 0.3492552645095018,
      "eval_loss": 1.96309494972229,
      "eval_runtime": 37.3346,
      "eval_samples_per_second": 26.785,
      "eval_steps_per_second": 1.125,
      "step": 340
    },
    {
      "epoch": 0.3502824858757062,
      "grad_norm": 2.9702041149139404,
      "learning_rate": 4.935287400076133e-06,
      "loss": 1.6703,
      "step": 341
    },
    {
      "epoch": 0.35130970724191063,
      "grad_norm": 2.7555935382843018,
      "learning_rate": 4.933384088313666e-06,
      "loss": 1.6879,
      "step": 342
    },
    {
      "epoch": 0.35233692860811505,
      "grad_norm": 2.8968169689178467,
      "learning_rate": 4.9314807765512e-06,
      "loss": 1.6342,
      "step": 343
    },
    {
      "epoch": 0.3533641499743195,
      "grad_norm": 3.010115385055542,
      "learning_rate": 4.929577464788733e-06,
      "loss": 1.5883,
      "step": 344
    },
    {
      "epoch": 0.3543913713405239,
      "grad_norm": 2.6449785232543945,
      "learning_rate": 4.9276741530262665e-06,
      "loss": 1.654,
      "step": 345
    },
    {
      "epoch": 0.3554185927067283,
      "grad_norm": 2.976109266281128,
      "learning_rate": 4.9257708412638e-06,
      "loss": 1.6483,
      "step": 346
    },
    {
      "epoch": 0.35644581407293274,
      "grad_norm": 2.856144666671753,
      "learning_rate": 4.923867529501332e-06,
      "loss": 1.6536,
      "step": 347
    },
    {
      "epoch": 0.3574730354391371,
      "grad_norm": 2.8142166137695312,
      "learning_rate": 4.921964217738866e-06,
      "loss": 1.6415,
      "step": 348
    },
    {
      "epoch": 0.35850025680534153,
      "grad_norm": 2.70807147026062,
      "learning_rate": 4.920060905976399e-06,
      "loss": 1.7099,
      "step": 349
    },
    {
      "epoch": 0.35952747817154596,
      "grad_norm": 2.7715506553649902,
      "learning_rate": 4.9181575942139325e-06,
      "loss": 1.6947,
      "step": 350
    },
    {
      "epoch": 0.35952747817154596,
      "eval_loss": 1.9507063627243042,
      "eval_runtime": 37.3094,
      "eval_samples_per_second": 26.803,
      "eval_steps_per_second": 1.126,
      "step": 350
    },
    {
      "epoch": 0.3605546995377504,
      "grad_norm": 2.8816769123077393,
      "learning_rate": 4.916254282451466e-06,
      "loss": 1.6906,
      "step": 351
    },
    {
      "epoch": 0.3615819209039548,
      "grad_norm": 3.08797025680542,
      "learning_rate": 4.914350970688999e-06,
      "loss": 1.7387,
      "step": 352
    },
    {
      "epoch": 0.3626091422701592,
      "grad_norm": 2.6018118858337402,
      "learning_rate": 4.9124476589265325e-06,
      "loss": 1.6319,
      "step": 353
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.7790255546569824,
      "learning_rate": 4.910544347164066e-06,
      "loss": 1.6357,
      "step": 354
    },
    {
      "epoch": 0.36466358500256807,
      "grad_norm": 2.7360289096832275,
      "learning_rate": 4.908641035401599e-06,
      "loss": 1.7127,
      "step": 355
    },
    {
      "epoch": 0.3656908063687725,
      "grad_norm": 2.6693854331970215,
      "learning_rate": 4.906737723639133e-06,
      "loss": 1.7747,
      "step": 356
    },
    {
      "epoch": 0.3667180277349769,
      "grad_norm": 2.642868995666504,
      "learning_rate": 4.904834411876665e-06,
      "loss": 1.6401,
      "step": 357
    },
    {
      "epoch": 0.3677452491011813,
      "grad_norm": 2.7230374813079834,
      "learning_rate": 4.902931100114199e-06,
      "loss": 1.6833,
      "step": 358
    },
    {
      "epoch": 0.3687724704673857,
      "grad_norm": 2.499457359313965,
      "learning_rate": 4.901027788351733e-06,
      "loss": 1.6007,
      "step": 359
    },
    {
      "epoch": 0.3697996918335901,
      "grad_norm": 2.7695276737213135,
      "learning_rate": 4.899124476589266e-06,
      "loss": 1.6486,
      "step": 360
    },
    {
      "epoch": 0.3697996918335901,
      "eval_loss": 1.9434194564819336,
      "eval_runtime": 37.3126,
      "eval_samples_per_second": 26.801,
      "eval_steps_per_second": 1.126,
      "step": 360
    },
    {
      "epoch": 0.37082691319979455,
      "grad_norm": 2.5715794563293457,
      "learning_rate": 4.8972211648267994e-06,
      "loss": 1.6183,
      "step": 361
    },
    {
      "epoch": 0.371854134565999,
      "grad_norm": 2.637859582901001,
      "learning_rate": 4.895317853064333e-06,
      "loss": 1.6648,
      "step": 362
    },
    {
      "epoch": 0.3728813559322034,
      "grad_norm": 2.5601117610931396,
      "learning_rate": 4.893414541301866e-06,
      "loss": 1.6475,
      "step": 363
    },
    {
      "epoch": 0.3739085772984078,
      "grad_norm": 2.429337739944458,
      "learning_rate": 4.891511229539399e-06,
      "loss": 1.6435,
      "step": 364
    },
    {
      "epoch": 0.37493579866461224,
      "grad_norm": 2.6890885829925537,
      "learning_rate": 4.889607917776932e-06,
      "loss": 1.6577,
      "step": 365
    },
    {
      "epoch": 0.37596302003081666,
      "grad_norm": 2.55492901802063,
      "learning_rate": 4.887704606014465e-06,
      "loss": 1.6523,
      "step": 366
    },
    {
      "epoch": 0.37699024139702103,
      "grad_norm": 2.841362476348877,
      "learning_rate": 4.885801294251999e-06,
      "loss": 1.6742,
      "step": 367
    },
    {
      "epoch": 0.37801746276322545,
      "grad_norm": 2.8041367530822754,
      "learning_rate": 4.883897982489532e-06,
      "loss": 1.5946,
      "step": 368
    },
    {
      "epoch": 0.3790446841294299,
      "grad_norm": 2.7135822772979736,
      "learning_rate": 4.8819946707270655e-06,
      "loss": 1.6909,
      "step": 369
    },
    {
      "epoch": 0.3800719054956343,
      "grad_norm": 2.9020211696624756,
      "learning_rate": 4.880091358964599e-06,
      "loss": 1.6804,
      "step": 370
    },
    {
      "epoch": 0.3800719054956343,
      "eval_loss": 1.9353317022323608,
      "eval_runtime": 37.3188,
      "eval_samples_per_second": 26.796,
      "eval_steps_per_second": 1.125,
      "step": 370
    },
    {
      "epoch": 0.3810991268618387,
      "grad_norm": 2.567310333251953,
      "learning_rate": 4.878188047202132e-06,
      "loss": 1.5885,
      "step": 371
    },
    {
      "epoch": 0.38212634822804314,
      "grad_norm": 2.910834789276123,
      "learning_rate": 4.876284735439666e-06,
      "loss": 1.7681,
      "step": 372
    },
    {
      "epoch": 0.38315356959424757,
      "grad_norm": 2.7507543563842773,
      "learning_rate": 4.874381423677199e-06,
      "loss": 1.7228,
      "step": 373
    },
    {
      "epoch": 0.384180790960452,
      "grad_norm": 2.9230854511260986,
      "learning_rate": 4.8724781119147315e-06,
      "loss": 1.6491,
      "step": 374
    },
    {
      "epoch": 0.3852080123266564,
      "grad_norm": 2.608980178833008,
      "learning_rate": 4.870574800152265e-06,
      "loss": 1.6757,
      "step": 375
    },
    {
      "epoch": 0.38623523369286084,
      "grad_norm": 2.9557063579559326,
      "learning_rate": 4.868671488389798e-06,
      "loss": 1.607,
      "step": 376
    },
    {
      "epoch": 0.3872624550590652,
      "grad_norm": 2.982243537902832,
      "learning_rate": 4.8667681766273316e-06,
      "loss": 1.6169,
      "step": 377
    },
    {
      "epoch": 0.3882896764252696,
      "grad_norm": 2.775012254714966,
      "learning_rate": 4.864864864864866e-06,
      "loss": 1.5835,
      "step": 378
    },
    {
      "epoch": 0.38931689779147405,
      "grad_norm": 2.9500741958618164,
      "learning_rate": 4.862961553102399e-06,
      "loss": 1.6504,
      "step": 379
    },
    {
      "epoch": 0.39034411915767847,
      "grad_norm": 2.889702796936035,
      "learning_rate": 4.8610582413399325e-06,
      "loss": 1.6814,
      "step": 380
    },
    {
      "epoch": 0.39034411915767847,
      "eval_loss": 1.9426418542861938,
      "eval_runtime": 37.3276,
      "eval_samples_per_second": 26.79,
      "eval_steps_per_second": 1.125,
      "step": 380
    },
    {
      "epoch": 0.3913713405238829,
      "grad_norm": 2.5075154304504395,
      "learning_rate": 4.859154929577465e-06,
      "loss": 1.6368,
      "step": 381
    },
    {
      "epoch": 0.3923985618900873,
      "grad_norm": 2.6461148262023926,
      "learning_rate": 4.857251617814998e-06,
      "loss": 1.6526,
      "step": 382
    },
    {
      "epoch": 0.39342578325629174,
      "grad_norm": 2.681501626968384,
      "learning_rate": 4.855348306052532e-06,
      "loss": 1.6846,
      "step": 383
    },
    {
      "epoch": 0.39445300462249616,
      "grad_norm": 2.557311773300171,
      "learning_rate": 4.853444994290065e-06,
      "loss": 1.6645,
      "step": 384
    },
    {
      "epoch": 0.3954802259887006,
      "grad_norm": 2.6819186210632324,
      "learning_rate": 4.8515416825275985e-06,
      "loss": 1.6323,
      "step": 385
    },
    {
      "epoch": 0.396507447354905,
      "grad_norm": 2.8803794384002686,
      "learning_rate": 4.849638370765132e-06,
      "loss": 1.6991,
      "step": 386
    },
    {
      "epoch": 0.3975346687211094,
      "grad_norm": 2.7181358337402344,
      "learning_rate": 4.847735059002665e-06,
      "loss": 1.6256,
      "step": 387
    },
    {
      "epoch": 0.3985618900873138,
      "grad_norm": 2.5668489933013916,
      "learning_rate": 4.8458317472401985e-06,
      "loss": 1.58,
      "step": 388
    },
    {
      "epoch": 0.3995891114535182,
      "grad_norm": 2.708804130554199,
      "learning_rate": 4.843928435477732e-06,
      "loss": 1.67,
      "step": 389
    },
    {
      "epoch": 0.40061633281972264,
      "grad_norm": 2.6565728187561035,
      "learning_rate": 4.842025123715265e-06,
      "loss": 1.622,
      "step": 390
    },
    {
      "epoch": 0.40061633281972264,
      "eval_loss": 1.9529728889465332,
      "eval_runtime": 37.3489,
      "eval_samples_per_second": 26.775,
      "eval_steps_per_second": 1.125,
      "step": 390
    },
    {
      "epoch": 0.40164355418592707,
      "grad_norm": 2.8743324279785156,
      "learning_rate": 4.840121811952798e-06,
      "loss": 1.6618,
      "step": 391
    },
    {
      "epoch": 0.4026707755521315,
      "grad_norm": 2.86134672164917,
      "learning_rate": 4.838218500190331e-06,
      "loss": 1.6544,
      "step": 392
    },
    {
      "epoch": 0.4036979969183359,
      "grad_norm": 3.3492071628570557,
      "learning_rate": 4.8363151884278645e-06,
      "loss": 1.6683,
      "step": 393
    },
    {
      "epoch": 0.40472521828454033,
      "grad_norm": 2.6630825996398926,
      "learning_rate": 4.834411876665398e-06,
      "loss": 1.6447,
      "step": 394
    },
    {
      "epoch": 0.40575243965074476,
      "grad_norm": 2.820096492767334,
      "learning_rate": 4.832508564902931e-06,
      "loss": 1.6579,
      "step": 395
    },
    {
      "epoch": 0.4067796610169492,
      "grad_norm": 3.076234817504883,
      "learning_rate": 4.830605253140465e-06,
      "loss": 1.6039,
      "step": 396
    },
    {
      "epoch": 0.40780688238315355,
      "grad_norm": 2.667719841003418,
      "learning_rate": 4.828701941377998e-06,
      "loss": 1.6364,
      "step": 397
    },
    {
      "epoch": 0.40883410374935797,
      "grad_norm": 3.2132771015167236,
      "learning_rate": 4.826798629615531e-06,
      "loss": 1.5906,
      "step": 398
    },
    {
      "epoch": 0.4098613251155624,
      "grad_norm": 2.661484479904175,
      "learning_rate": 4.824895317853065e-06,
      "loss": 1.613,
      "step": 399
    },
    {
      "epoch": 0.4108885464817668,
      "grad_norm": 3.002636671066284,
      "learning_rate": 4.822992006090598e-06,
      "loss": 1.6305,
      "step": 400
    },
    {
      "epoch": 0.4108885464817668,
      "eval_loss": 1.9345510005950928,
      "eval_runtime": 37.3369,
      "eval_samples_per_second": 26.783,
      "eval_steps_per_second": 1.125,
      "step": 400
    },
    {
      "epoch": 0.41191576784797124,
      "grad_norm": 2.6405227184295654,
      "learning_rate": 4.821088694328131e-06,
      "loss": 1.5881,
      "step": 401
    },
    {
      "epoch": 0.41294298921417566,
      "grad_norm": 3.006169080734253,
      "learning_rate": 4.819185382565665e-06,
      "loss": 1.6126,
      "step": 402
    },
    {
      "epoch": 0.4139702105803801,
      "grad_norm": 2.7614858150482178,
      "learning_rate": 4.817282070803198e-06,
      "loss": 1.6695,
      "step": 403
    },
    {
      "epoch": 0.4149974319465845,
      "grad_norm": 2.797260046005249,
      "learning_rate": 4.8153787590407315e-06,
      "loss": 1.6539,
      "step": 404
    },
    {
      "epoch": 0.41602465331278893,
      "grad_norm": 2.8156301975250244,
      "learning_rate": 4.813475447278265e-06,
      "loss": 1.5845,
      "step": 405
    },
    {
      "epoch": 0.41705187467899335,
      "grad_norm": 2.74100399017334,
      "learning_rate": 4.811572135515798e-06,
      "loss": 1.6699,
      "step": 406
    },
    {
      "epoch": 0.4180790960451977,
      "grad_norm": 2.7785775661468506,
      "learning_rate": 4.809668823753332e-06,
      "loss": 1.5987,
      "step": 407
    },
    {
      "epoch": 0.41910631741140214,
      "grad_norm": 2.663504123687744,
      "learning_rate": 4.807765511990864e-06,
      "loss": 1.5987,
      "step": 408
    },
    {
      "epoch": 0.42013353877760656,
      "grad_norm": 2.903374433517456,
      "learning_rate": 4.8058622002283975e-06,
      "loss": 1.6302,
      "step": 409
    },
    {
      "epoch": 0.421160760143811,
      "grad_norm": 2.696049690246582,
      "learning_rate": 4.803958888465931e-06,
      "loss": 1.671,
      "step": 410
    },
    {
      "epoch": 0.421160760143811,
      "eval_loss": 1.9333878755569458,
      "eval_runtime": 37.2472,
      "eval_samples_per_second": 26.848,
      "eval_steps_per_second": 1.128,
      "step": 410
    },
    {
      "epoch": 0.4221879815100154,
      "grad_norm": 2.486461877822876,
      "learning_rate": 4.802055576703464e-06,
      "loss": 1.6282,
      "step": 411
    },
    {
      "epoch": 0.42321520287621983,
      "grad_norm": 2.6812057495117188,
      "learning_rate": 4.8001522649409976e-06,
      "loss": 1.7144,
      "step": 412
    },
    {
      "epoch": 0.42424242424242425,
      "grad_norm": 2.8370249271392822,
      "learning_rate": 4.798248953178531e-06,
      "loss": 1.5979,
      "step": 413
    },
    {
      "epoch": 0.4252696456086287,
      "grad_norm": 2.6073036193847656,
      "learning_rate": 4.796345641416064e-06,
      "loss": 1.5703,
      "step": 414
    },
    {
      "epoch": 0.4262968669748331,
      "grad_norm": 2.5891220569610596,
      "learning_rate": 4.794442329653598e-06,
      "loss": 1.6212,
      "step": 415
    },
    {
      "epoch": 0.42732408834103747,
      "grad_norm": 2.6296865940093994,
      "learning_rate": 4.792539017891131e-06,
      "loss": 1.6711,
      "step": 416
    },
    {
      "epoch": 0.4283513097072419,
      "grad_norm": 2.573472499847412,
      "learning_rate": 4.790635706128664e-06,
      "loss": 1.5994,
      "step": 417
    },
    {
      "epoch": 0.4293785310734463,
      "grad_norm": 2.734464168548584,
      "learning_rate": 4.788732394366197e-06,
      "loss": 1.6383,
      "step": 418
    },
    {
      "epoch": 0.43040575243965074,
      "grad_norm": 2.6412851810455322,
      "learning_rate": 4.786829082603731e-06,
      "loss": 1.5646,
      "step": 419
    },
    {
      "epoch": 0.43143297380585516,
      "grad_norm": 2.5065555572509766,
      "learning_rate": 4.7849257708412645e-06,
      "loss": 1.5636,
      "step": 420
    },
    {
      "epoch": 0.43143297380585516,
      "eval_loss": 1.9357348680496216,
      "eval_runtime": 37.372,
      "eval_samples_per_second": 26.758,
      "eval_steps_per_second": 1.124,
      "step": 420
    },
    {
      "epoch": 0.4324601951720596,
      "grad_norm": 2.4889843463897705,
      "learning_rate": 4.783022459078798e-06,
      "loss": 1.6457,
      "step": 421
    },
    {
      "epoch": 0.433487416538264,
      "grad_norm": 2.544801950454712,
      "learning_rate": 4.781119147316331e-06,
      "loss": 1.5897,
      "step": 422
    },
    {
      "epoch": 0.4345146379044684,
      "grad_norm": 2.7834064960479736,
      "learning_rate": 4.7792158355538645e-06,
      "loss": 1.6174,
      "step": 423
    },
    {
      "epoch": 0.43554185927067285,
      "grad_norm": 2.602952480316162,
      "learning_rate": 4.777312523791398e-06,
      "loss": 1.6152,
      "step": 424
    },
    {
      "epoch": 0.43656908063687727,
      "grad_norm": 2.629746198654175,
      "learning_rate": 4.77540921202893e-06,
      "loss": 1.5991,
      "step": 425
    },
    {
      "epoch": 0.43759630200308164,
      "grad_norm": 2.4987270832061768,
      "learning_rate": 4.773505900266464e-06,
      "loss": 1.6342,
      "step": 426
    },
    {
      "epoch": 0.43862352336928606,
      "grad_norm": 2.607611894607544,
      "learning_rate": 4.771602588503997e-06,
      "loss": 1.6145,
      "step": 427
    },
    {
      "epoch": 0.4396507447354905,
      "grad_norm": 2.5654876232147217,
      "learning_rate": 4.7696992767415305e-06,
      "loss": 1.6162,
      "step": 428
    },
    {
      "epoch": 0.4406779661016949,
      "grad_norm": 2.5701870918273926,
      "learning_rate": 4.767795964979064e-06,
      "loss": 1.6516,
      "step": 429
    },
    {
      "epoch": 0.44170518746789933,
      "grad_norm": 2.8381104469299316,
      "learning_rate": 4.765892653216597e-06,
      "loss": 1.659,
      "step": 430
    },
    {
      "epoch": 0.44170518746789933,
      "eval_loss": 1.9259562492370605,
      "eval_runtime": 37.3906,
      "eval_samples_per_second": 26.745,
      "eval_steps_per_second": 1.123,
      "step": 430
    },
    {
      "epoch": 0.44273240883410375,
      "grad_norm": 2.5777862071990967,
      "learning_rate": 4.763989341454131e-06,
      "loss": 1.6571,
      "step": 431
    },
    {
      "epoch": 0.4437596302003082,
      "grad_norm": 2.6656861305236816,
      "learning_rate": 4.762086029691664e-06,
      "loss": 1.5786,
      "step": 432
    },
    {
      "epoch": 0.4447868515665126,
      "grad_norm": 2.724851131439209,
      "learning_rate": 4.760182717929197e-06,
      "loss": 1.5916,
      "step": 433
    },
    {
      "epoch": 0.445814072932717,
      "grad_norm": 2.532191514968872,
      "learning_rate": 4.758279406166731e-06,
      "loss": 1.608,
      "step": 434
    },
    {
      "epoch": 0.44684129429892144,
      "grad_norm": 2.6252050399780273,
      "learning_rate": 4.756376094404263e-06,
      "loss": 1.5297,
      "step": 435
    },
    {
      "epoch": 0.4478685156651258,
      "grad_norm": 2.674530029296875,
      "learning_rate": 4.7544727826417966e-06,
      "loss": 1.615,
      "step": 436
    },
    {
      "epoch": 0.44889573703133023,
      "grad_norm": 2.9209136962890625,
      "learning_rate": 4.75256947087933e-06,
      "loss": 1.6749,
      "step": 437
    },
    {
      "epoch": 0.44992295839753466,
      "grad_norm": 2.4959394931793213,
      "learning_rate": 4.750666159116863e-06,
      "loss": 1.6186,
      "step": 438
    },
    {
      "epoch": 0.4509501797637391,
      "grad_norm": 2.6317503452301025,
      "learning_rate": 4.7487628473543975e-06,
      "loss": 1.5873,
      "step": 439
    },
    {
      "epoch": 0.4519774011299435,
      "grad_norm": 2.6485610008239746,
      "learning_rate": 4.746859535591931e-06,
      "loss": 1.5451,
      "step": 440
    },
    {
      "epoch": 0.4519774011299435,
      "eval_loss": 1.9235283136367798,
      "eval_runtime": 37.367,
      "eval_samples_per_second": 26.762,
      "eval_steps_per_second": 1.124,
      "step": 440
    },
    {
      "epoch": 0.4530046224961479,
      "grad_norm": 2.491852045059204,
      "learning_rate": 4.744956223829464e-06,
      "loss": 1.6016,
      "step": 441
    },
    {
      "epoch": 0.45403184386235235,
      "grad_norm": 2.5354514122009277,
      "learning_rate": 4.743052912066997e-06,
      "loss": 1.5795,
      "step": 442
    },
    {
      "epoch": 0.45505906522855677,
      "grad_norm": 2.7756357192993164,
      "learning_rate": 4.74114960030453e-06,
      "loss": 1.6299,
      "step": 443
    },
    {
      "epoch": 0.4560862865947612,
      "grad_norm": 2.638906955718994,
      "learning_rate": 4.7392462885420635e-06,
      "loss": 1.6429,
      "step": 444
    },
    {
      "epoch": 0.4571135079609656,
      "grad_norm": 2.5858829021453857,
      "learning_rate": 4.737342976779597e-06,
      "loss": 1.5316,
      "step": 445
    },
    {
      "epoch": 0.45814072932717,
      "grad_norm": 2.798567295074463,
      "learning_rate": 4.73543966501713e-06,
      "loss": 1.6166,
      "step": 446
    },
    {
      "epoch": 0.4591679506933744,
      "grad_norm": 2.512143135070801,
      "learning_rate": 4.7335363532546636e-06,
      "loss": 1.5952,
      "step": 447
    },
    {
      "epoch": 0.4601951720595788,
      "grad_norm": 2.7063655853271484,
      "learning_rate": 4.731633041492197e-06,
      "loss": 1.6051,
      "step": 448
    },
    {
      "epoch": 0.46122239342578325,
      "grad_norm": 2.6358017921447754,
      "learning_rate": 4.72972972972973e-06,
      "loss": 1.6008,
      "step": 449
    },
    {
      "epoch": 0.4622496147919877,
      "grad_norm": 2.75992751121521,
      "learning_rate": 4.727826417967264e-06,
      "loss": 1.644,
      "step": 450
    },
    {
      "epoch": 0.4622496147919877,
      "eval_loss": 1.927345633506775,
      "eval_runtime": 37.3707,
      "eval_samples_per_second": 26.759,
      "eval_steps_per_second": 1.124,
      "step": 450
    },
    {
      "epoch": 0.4632768361581921,
      "grad_norm": 2.9103775024414062,
      "learning_rate": 4.725923106204797e-06,
      "loss": 1.6176,
      "step": 451
    },
    {
      "epoch": 0.4643040575243965,
      "grad_norm": 2.627481460571289,
      "learning_rate": 4.72401979444233e-06,
      "loss": 1.586,
      "step": 452
    },
    {
      "epoch": 0.46533127889060094,
      "grad_norm": 2.8432116508483887,
      "learning_rate": 4.722116482679863e-06,
      "loss": 1.6303,
      "step": 453
    },
    {
      "epoch": 0.46635850025680536,
      "grad_norm": 2.8088958263397217,
      "learning_rate": 4.720213170917396e-06,
      "loss": 1.6163,
      "step": 454
    },
    {
      "epoch": 0.46738572162300973,
      "grad_norm": 2.6040408611297607,
      "learning_rate": 4.71830985915493e-06,
      "loss": 1.6346,
      "step": 455
    },
    {
      "epoch": 0.46841294298921415,
      "grad_norm": 2.8951191902160645,
      "learning_rate": 4.716406547392463e-06,
      "loss": 1.5859,
      "step": 456
    },
    {
      "epoch": 0.4694401643554186,
      "grad_norm": 2.792473793029785,
      "learning_rate": 4.714503235629996e-06,
      "loss": 1.6441,
      "step": 457
    },
    {
      "epoch": 0.470467385721623,
      "grad_norm": 2.6793878078460693,
      "learning_rate": 4.71259992386753e-06,
      "loss": 1.5699,
      "step": 458
    },
    {
      "epoch": 0.4714946070878274,
      "grad_norm": 2.7223684787750244,
      "learning_rate": 4.710696612105063e-06,
      "loss": 1.6038,
      "step": 459
    },
    {
      "epoch": 0.47252182845403184,
      "grad_norm": 2.8043856620788574,
      "learning_rate": 4.708793300342596e-06,
      "loss": 1.6018,
      "step": 460
    },
    {
      "epoch": 0.47252182845403184,
      "eval_loss": 1.9213919639587402,
      "eval_runtime": 37.3451,
      "eval_samples_per_second": 26.777,
      "eval_steps_per_second": 1.125,
      "step": 460
    },
    {
      "epoch": 0.47354904982023627,
      "grad_norm": 2.7657470703125,
      "learning_rate": 4.70688998858013e-06,
      "loss": 1.5896,
      "step": 461
    },
    {
      "epoch": 0.4745762711864407,
      "grad_norm": 2.8417739868164062,
      "learning_rate": 4.704986676817663e-06,
      "loss": 1.5892,
      "step": 462
    },
    {
      "epoch": 0.4756034925526451,
      "grad_norm": 2.5660197734832764,
      "learning_rate": 4.7030833650551965e-06,
      "loss": 1.615,
      "step": 463
    },
    {
      "epoch": 0.47663071391884954,
      "grad_norm": 2.762880325317383,
      "learning_rate": 4.70118005329273e-06,
      "loss": 1.5747,
      "step": 464
    },
    {
      "epoch": 0.4776579352850539,
      "grad_norm": 2.7191319465637207,
      "learning_rate": 4.699276741530263e-06,
      "loss": 1.6529,
      "step": 465
    },
    {
      "epoch": 0.4786851566512583,
      "grad_norm": 2.4995453357696533,
      "learning_rate": 4.697373429767797e-06,
      "loss": 1.6285,
      "step": 466
    },
    {
      "epoch": 0.47971237801746275,
      "grad_norm": 2.8300375938415527,
      "learning_rate": 4.69547011800533e-06,
      "loss": 1.6171,
      "step": 467
    },
    {
      "epoch": 0.48073959938366717,
      "grad_norm": 2.6028811931610107,
      "learning_rate": 4.693566806242863e-06,
      "loss": 1.6014,
      "step": 468
    },
    {
      "epoch": 0.4817668207498716,
      "grad_norm": 3.080156087875366,
      "learning_rate": 4.691663494480397e-06,
      "loss": 1.6055,
      "step": 469
    },
    {
      "epoch": 0.482794042116076,
      "grad_norm": 2.6707916259765625,
      "learning_rate": 4.689760182717929e-06,
      "loss": 1.627,
      "step": 470
    },
    {
      "epoch": 0.482794042116076,
      "eval_loss": 1.9015462398529053,
      "eval_runtime": 37.3476,
      "eval_samples_per_second": 26.776,
      "eval_steps_per_second": 1.125,
      "step": 470
    },
    {
      "epoch": 0.48382126348228044,
      "grad_norm": 2.6820614337921143,
      "learning_rate": 4.6878568709554626e-06,
      "loss": 1.5261,
      "step": 471
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 2.841031789779663,
      "learning_rate": 4.685953559192996e-06,
      "loss": 1.5903,
      "step": 472
    },
    {
      "epoch": 0.4858757062146893,
      "grad_norm": 2.470867156982422,
      "learning_rate": 4.684050247430529e-06,
      "loss": 1.6091,
      "step": 473
    },
    {
      "epoch": 0.4869029275808937,
      "grad_norm": 2.4955153465270996,
      "learning_rate": 4.682146935668063e-06,
      "loss": 1.544,
      "step": 474
    },
    {
      "epoch": 0.4879301489470981,
      "grad_norm": 2.9290695190429688,
      "learning_rate": 4.680243623905596e-06,
      "loss": 1.5893,
      "step": 475
    },
    {
      "epoch": 0.4889573703133025,
      "grad_norm": 2.5902767181396484,
      "learning_rate": 4.678340312143129e-06,
      "loss": 1.6117,
      "step": 476
    },
    {
      "epoch": 0.4899845916795069,
      "grad_norm": 2.6067540645599365,
      "learning_rate": 4.676437000380663e-06,
      "loss": 1.5703,
      "step": 477
    },
    {
      "epoch": 0.49101181304571134,
      "grad_norm": 2.490105152130127,
      "learning_rate": 4.674533688618196e-06,
      "loss": 1.6931,
      "step": 478
    },
    {
      "epoch": 0.49203903441191577,
      "grad_norm": 2.7736282348632812,
      "learning_rate": 4.6726303768557295e-06,
      "loss": 1.6204,
      "step": 479
    },
    {
      "epoch": 0.4930662557781202,
      "grad_norm": 2.741196870803833,
      "learning_rate": 4.670727065093263e-06,
      "loss": 1.5331,
      "step": 480
    },
    {
      "epoch": 0.4930662557781202,
      "eval_loss": 1.9095442295074463,
      "eval_runtime": 37.3351,
      "eval_samples_per_second": 26.784,
      "eval_steps_per_second": 1.125,
      "step": 480
    },
    {
      "epoch": 0.4940934771443246,
      "grad_norm": 2.7624332904815674,
      "learning_rate": 4.668823753330796e-06,
      "loss": 1.5591,
      "step": 481
    },
    {
      "epoch": 0.49512069851052903,
      "grad_norm": 2.86407470703125,
      "learning_rate": 4.6669204415683295e-06,
      "loss": 1.5844,
      "step": 482
    },
    {
      "epoch": 0.49614791987673346,
      "grad_norm": 2.770030975341797,
      "learning_rate": 4.665017129805863e-06,
      "loss": 1.6515,
      "step": 483
    },
    {
      "epoch": 0.4971751412429379,
      "grad_norm": 2.569862127304077,
      "learning_rate": 4.663113818043396e-06,
      "loss": 1.6228,
      "step": 484
    },
    {
      "epoch": 0.49820236260914225,
      "grad_norm": 2.7176733016967773,
      "learning_rate": 4.66121050628093e-06,
      "loss": 1.5938,
      "step": 485
    },
    {
      "epoch": 0.49922958397534667,
      "grad_norm": 2.4514758586883545,
      "learning_rate": 4.659307194518463e-06,
      "loss": 1.6307,
      "step": 486
    },
    {
      "epoch": 0.5002568053415511,
      "grad_norm": 2.5439529418945312,
      "learning_rate": 4.6574038827559955e-06,
      "loss": 1.5855,
      "step": 487
    },
    {
      "epoch": 0.5012840267077555,
      "grad_norm": 2.6395790576934814,
      "learning_rate": 4.655500570993529e-06,
      "loss": 1.6124,
      "step": 488
    },
    {
      "epoch": 0.50231124807396,
      "grad_norm": 2.69413685798645,
      "learning_rate": 4.653597259231062e-06,
      "loss": 1.6358,
      "step": 489
    },
    {
      "epoch": 0.5033384694401644,
      "grad_norm": 2.693129777908325,
      "learning_rate": 4.651693947468596e-06,
      "loss": 1.5218,
      "step": 490
    },
    {
      "epoch": 0.5033384694401644,
      "eval_loss": 1.9201244115829468,
      "eval_runtime": 37.3324,
      "eval_samples_per_second": 26.786,
      "eval_steps_per_second": 1.125,
      "step": 490
    },
    {
      "epoch": 0.5043656908063687,
      "grad_norm": 2.630596876144409,
      "learning_rate": 4.649790635706129e-06,
      "loss": 1.6719,
      "step": 491
    },
    {
      "epoch": 0.5053929121725732,
      "grad_norm": 2.7230262756347656,
      "learning_rate": 4.647887323943662e-06,
      "loss": 1.5705,
      "step": 492
    },
    {
      "epoch": 0.5064201335387776,
      "grad_norm": 2.529944896697998,
      "learning_rate": 4.645984012181196e-06,
      "loss": 1.5767,
      "step": 493
    },
    {
      "epoch": 0.507447354904982,
      "grad_norm": 2.546849489212036,
      "learning_rate": 4.644080700418729e-06,
      "loss": 1.5039,
      "step": 494
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 2.6698246002197266,
      "learning_rate": 4.642177388656262e-06,
      "loss": 1.6229,
      "step": 495
    },
    {
      "epoch": 0.5095017976373909,
      "grad_norm": 2.7093148231506348,
      "learning_rate": 4.640274076893796e-06,
      "loss": 1.5797,
      "step": 496
    },
    {
      "epoch": 0.5105290190035953,
      "grad_norm": 2.48629093170166,
      "learning_rate": 4.638370765131328e-06,
      "loss": 1.6366,
      "step": 497
    },
    {
      "epoch": 0.5115562403697997,
      "grad_norm": 2.5818612575531006,
      "learning_rate": 4.636467453368862e-06,
      "loss": 1.5608,
      "step": 498
    },
    {
      "epoch": 0.5125834617360041,
      "grad_norm": 2.6263651847839355,
      "learning_rate": 4.634564141606396e-06,
      "loss": 1.5518,
      "step": 499
    },
    {
      "epoch": 0.5136106831022085,
      "grad_norm": 2.5724668502807617,
      "learning_rate": 4.632660829843929e-06,
      "loss": 1.6195,
      "step": 500
    },
    {
      "epoch": 0.5136106831022085,
      "eval_loss": 1.9094111919403076,
      "eval_runtime": 37.3483,
      "eval_samples_per_second": 26.775,
      "eval_steps_per_second": 1.125,
      "step": 500
    },
    {
      "epoch": 0.514637904468413,
      "grad_norm": 2.5028088092803955,
      "learning_rate": 4.630757518081463e-06,
      "loss": 1.5454,
      "step": 501
    },
    {
      "epoch": 0.5156651258346173,
      "grad_norm": 2.588115692138672,
      "learning_rate": 4.628854206318996e-06,
      "loss": 1.6676,
      "step": 502
    },
    {
      "epoch": 0.5166923472008218,
      "grad_norm": 2.9701285362243652,
      "learning_rate": 4.626950894556529e-06,
      "loss": 1.5822,
      "step": 503
    },
    {
      "epoch": 0.5177195685670262,
      "grad_norm": 2.593867301940918,
      "learning_rate": 4.625047582794062e-06,
      "loss": 1.6091,
      "step": 504
    },
    {
      "epoch": 0.5187467899332306,
      "grad_norm": 2.559642791748047,
      "learning_rate": 4.623144271031595e-06,
      "loss": 1.5359,
      "step": 505
    },
    {
      "epoch": 0.519774011299435,
      "grad_norm": 2.4158153533935547,
      "learning_rate": 4.6212409592691286e-06,
      "loss": 1.5476,
      "step": 506
    },
    {
      "epoch": 0.5208012326656395,
      "grad_norm": 2.666099786758423,
      "learning_rate": 4.619337647506662e-06,
      "loss": 1.6297,
      "step": 507
    },
    {
      "epoch": 0.5218284540318439,
      "grad_norm": 2.7026472091674805,
      "learning_rate": 4.617434335744195e-06,
      "loss": 1.5891,
      "step": 508
    },
    {
      "epoch": 0.5228556753980482,
      "grad_norm": 2.4591612815856934,
      "learning_rate": 4.615531023981729e-06,
      "loss": 1.5546,
      "step": 509
    },
    {
      "epoch": 0.5238828967642527,
      "grad_norm": 2.870255708694458,
      "learning_rate": 4.613627712219262e-06,
      "loss": 1.5817,
      "step": 510
    },
    {
      "epoch": 0.5238828967642527,
      "eval_loss": 1.910629153251648,
      "eval_runtime": 37.1517,
      "eval_samples_per_second": 26.917,
      "eval_steps_per_second": 1.131,
      "step": 510
    },
    {
      "epoch": 0.5249101181304571,
      "grad_norm": 2.653801679611206,
      "learning_rate": 4.611724400456795e-06,
      "loss": 1.586,
      "step": 511
    },
    {
      "epoch": 0.5259373394966615,
      "grad_norm": 2.4235377311706543,
      "learning_rate": 4.609821088694329e-06,
      "loss": 1.5886,
      "step": 512
    },
    {
      "epoch": 0.5269645608628659,
      "grad_norm": 2.8782477378845215,
      "learning_rate": 4.607917776931862e-06,
      "loss": 1.5634,
      "step": 513
    },
    {
      "epoch": 0.5279917822290704,
      "grad_norm": 2.792876958847046,
      "learning_rate": 4.606014465169395e-06,
      "loss": 1.6073,
      "step": 514
    },
    {
      "epoch": 0.5290190035952748,
      "grad_norm": 2.5579030513763428,
      "learning_rate": 4.604111153406928e-06,
      "loss": 1.6161,
      "step": 515
    },
    {
      "epoch": 0.5300462249614792,
      "grad_norm": 2.6486189365386963,
      "learning_rate": 4.602207841644461e-06,
      "loss": 1.5297,
      "step": 516
    },
    {
      "epoch": 0.5310734463276836,
      "grad_norm": 2.9047019481658936,
      "learning_rate": 4.600304529881995e-06,
      "loss": 1.626,
      "step": 517
    },
    {
      "epoch": 0.5321006676938881,
      "grad_norm": 2.9164223670959473,
      "learning_rate": 4.598401218119528e-06,
      "loss": 1.6051,
      "step": 518
    },
    {
      "epoch": 0.5331278890600925,
      "grad_norm": 2.544311046600342,
      "learning_rate": 4.5964979063570614e-06,
      "loss": 1.6247,
      "step": 519
    },
    {
      "epoch": 0.5341551104262968,
      "grad_norm": 2.8632311820983887,
      "learning_rate": 4.594594594594596e-06,
      "loss": 1.5615,
      "step": 520
    },
    {
      "epoch": 0.5341551104262968,
      "eval_loss": 1.9171669483184814,
      "eval_runtime": 37.2243,
      "eval_samples_per_second": 26.864,
      "eval_steps_per_second": 1.128,
      "step": 520
    },
    {
      "epoch": 0.5351823317925013,
      "grad_norm": 2.507697105407715,
      "learning_rate": 4.592691282832128e-06,
      "loss": 1.5929,
      "step": 521
    },
    {
      "epoch": 0.5362095531587057,
      "grad_norm": 2.657752752304077,
      "learning_rate": 4.5907879710696615e-06,
      "loss": 1.6159,
      "step": 522
    },
    {
      "epoch": 0.5372367745249101,
      "grad_norm": 2.32254695892334,
      "learning_rate": 4.588884659307195e-06,
      "loss": 1.5318,
      "step": 523
    },
    {
      "epoch": 0.5382639958911145,
      "grad_norm": 2.5979015827178955,
      "learning_rate": 4.586981347544728e-06,
      "loss": 1.5999,
      "step": 524
    },
    {
      "epoch": 0.539291217257319,
      "grad_norm": 2.839956521987915,
      "learning_rate": 4.585078035782262e-06,
      "loss": 1.547,
      "step": 525
    },
    {
      "epoch": 0.5403184386235234,
      "grad_norm": 2.700131893157959,
      "learning_rate": 4.583174724019795e-06,
      "loss": 1.5831,
      "step": 526
    },
    {
      "epoch": 0.5413456599897278,
      "grad_norm": 2.365183115005493,
      "learning_rate": 4.581271412257328e-06,
      "loss": 1.5357,
      "step": 527
    },
    {
      "epoch": 0.5423728813559322,
      "grad_norm": 2.675567865371704,
      "learning_rate": 4.579368100494862e-06,
      "loss": 1.5212,
      "step": 528
    },
    {
      "epoch": 0.5434001027221366,
      "grad_norm": 2.9096713066101074,
      "learning_rate": 4.577464788732395e-06,
      "loss": 1.6196,
      "step": 529
    },
    {
      "epoch": 0.544427324088341,
      "grad_norm": 2.6186270713806152,
      "learning_rate": 4.575561476969928e-06,
      "loss": 1.5895,
      "step": 530
    },
    {
      "epoch": 0.544427324088341,
      "eval_loss": 1.926998496055603,
      "eval_runtime": 37.2248,
      "eval_samples_per_second": 26.864,
      "eval_steps_per_second": 1.128,
      "step": 530
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 2.7291345596313477,
      "learning_rate": 4.573658165207461e-06,
      "loss": 1.6335,
      "step": 531
    },
    {
      "epoch": 0.5464817668207499,
      "grad_norm": 2.7084522247314453,
      "learning_rate": 4.571754853444994e-06,
      "loss": 1.5968,
      "step": 532
    },
    {
      "epoch": 0.5475089881869543,
      "grad_norm": 2.699721336364746,
      "learning_rate": 4.569851541682528e-06,
      "loss": 1.5775,
      "step": 533
    },
    {
      "epoch": 0.5485362095531587,
      "grad_norm": 2.602526903152466,
      "learning_rate": 4.567948229920061e-06,
      "loss": 1.5497,
      "step": 534
    },
    {
      "epoch": 0.5495634309193631,
      "grad_norm": 2.597362518310547,
      "learning_rate": 4.566044918157594e-06,
      "loss": 1.5218,
      "step": 535
    },
    {
      "epoch": 0.5505906522855676,
      "grad_norm": 2.6624414920806885,
      "learning_rate": 4.564141606395128e-06,
      "loss": 1.5553,
      "step": 536
    },
    {
      "epoch": 0.551617873651772,
      "grad_norm": 2.750807285308838,
      "learning_rate": 4.562238294632661e-06,
      "loss": 1.6078,
      "step": 537
    },
    {
      "epoch": 0.5526450950179764,
      "grad_norm": 2.7480952739715576,
      "learning_rate": 4.5603349828701945e-06,
      "loss": 1.5707,
      "step": 538
    },
    {
      "epoch": 0.5536723163841808,
      "grad_norm": 2.6287965774536133,
      "learning_rate": 4.558431671107728e-06,
      "loss": 1.5961,
      "step": 539
    },
    {
      "epoch": 0.5546995377503852,
      "grad_norm": 2.721311569213867,
      "learning_rate": 4.556528359345261e-06,
      "loss": 1.6335,
      "step": 540
    },
    {
      "epoch": 0.5546995377503852,
      "eval_loss": 1.9290324449539185,
      "eval_runtime": 37.2243,
      "eval_samples_per_second": 26.864,
      "eval_steps_per_second": 1.128,
      "step": 540
    },
    {
      "epoch": 0.5557267591165896,
      "grad_norm": 2.470325469970703,
      "learning_rate": 4.5546250475827946e-06,
      "loss": 1.5645,
      "step": 541
    },
    {
      "epoch": 0.556753980482794,
      "grad_norm": 2.6089508533477783,
      "learning_rate": 4.552721735820328e-06,
      "loss": 1.5256,
      "step": 542
    },
    {
      "epoch": 0.5577812018489985,
      "grad_norm": 2.541785478591919,
      "learning_rate": 4.550818424057861e-06,
      "loss": 1.5641,
      "step": 543
    },
    {
      "epoch": 0.5588084232152029,
      "grad_norm": 2.6722488403320312,
      "learning_rate": 4.548915112295395e-06,
      "loss": 1.5708,
      "step": 544
    },
    {
      "epoch": 0.5598356445814073,
      "grad_norm": 2.583695650100708,
      "learning_rate": 4.547011800532928e-06,
      "loss": 1.5799,
      "step": 545
    },
    {
      "epoch": 0.5608628659476117,
      "grad_norm": 2.956885576248169,
      "learning_rate": 4.545108488770461e-06,
      "loss": 1.5345,
      "step": 546
    },
    {
      "epoch": 0.5618900873138162,
      "grad_norm": 2.5534746646881104,
      "learning_rate": 4.543205177007995e-06,
      "loss": 1.636,
      "step": 547
    },
    {
      "epoch": 0.5629173086800205,
      "grad_norm": 2.8133325576782227,
      "learning_rate": 4.541301865245527e-06,
      "loss": 1.5678,
      "step": 548
    },
    {
      "epoch": 0.5639445300462249,
      "grad_norm": 2.5969228744506836,
      "learning_rate": 4.539398553483061e-06,
      "loss": 1.5975,
      "step": 549
    },
    {
      "epoch": 0.5649717514124294,
      "grad_norm": 2.7594687938690186,
      "learning_rate": 4.537495241720594e-06,
      "loss": 1.5622,
      "step": 550
    },
    {
      "epoch": 0.5649717514124294,
      "eval_loss": 1.8870681524276733,
      "eval_runtime": 37.2094,
      "eval_samples_per_second": 26.875,
      "eval_steps_per_second": 1.129,
      "step": 550
    },
    {
      "epoch": 0.5659989727786338,
      "grad_norm": 2.7413411140441895,
      "learning_rate": 4.535591929958127e-06,
      "loss": 1.6031,
      "step": 551
    },
    {
      "epoch": 0.5670261941448382,
      "grad_norm": 2.6182661056518555,
      "learning_rate": 4.533688618195661e-06,
      "loss": 1.5828,
      "step": 552
    },
    {
      "epoch": 0.5680534155110426,
      "grad_norm": 2.5016422271728516,
      "learning_rate": 4.531785306433194e-06,
      "loss": 1.5712,
      "step": 553
    },
    {
      "epoch": 0.5690806368772471,
      "grad_norm": 2.7788732051849365,
      "learning_rate": 4.529881994670727e-06,
      "loss": 1.5472,
      "step": 554
    },
    {
      "epoch": 0.5701078582434514,
      "grad_norm": 2.7602436542510986,
      "learning_rate": 4.527978682908261e-06,
      "loss": 1.5526,
      "step": 555
    },
    {
      "epoch": 0.5711350796096559,
      "grad_norm": 2.5898544788360596,
      "learning_rate": 4.526075371145794e-06,
      "loss": 1.5644,
      "step": 556
    },
    {
      "epoch": 0.5721623009758603,
      "grad_norm": 2.439344882965088,
      "learning_rate": 4.5241720593833275e-06,
      "loss": 1.5892,
      "step": 557
    },
    {
      "epoch": 0.5731895223420647,
      "grad_norm": 2.6324563026428223,
      "learning_rate": 4.52226874762086e-06,
      "loss": 1.5729,
      "step": 558
    },
    {
      "epoch": 0.5742167437082691,
      "grad_norm": 2.674042224884033,
      "learning_rate": 4.520365435858393e-06,
      "loss": 1.6016,
      "step": 559
    },
    {
      "epoch": 0.5752439650744735,
      "grad_norm": 2.692619800567627,
      "learning_rate": 4.518462124095928e-06,
      "loss": 1.6229,
      "step": 560
    },
    {
      "epoch": 0.5752439650744735,
      "eval_loss": 1.8689097166061401,
      "eval_runtime": 37.2177,
      "eval_samples_per_second": 26.869,
      "eval_steps_per_second": 1.128,
      "step": 560
    },
    {
      "epoch": 0.576271186440678,
      "grad_norm": 2.636924982070923,
      "learning_rate": 4.516558812333461e-06,
      "loss": 1.5748,
      "step": 561
    },
    {
      "epoch": 0.5772984078068824,
      "grad_norm": 2.549285888671875,
      "learning_rate": 4.514655500570994e-06,
      "loss": 1.5656,
      "step": 562
    },
    {
      "epoch": 0.5783256291730868,
      "grad_norm": 2.5873067378997803,
      "learning_rate": 4.512752188808528e-06,
      "loss": 1.631,
      "step": 563
    },
    {
      "epoch": 0.5793528505392912,
      "grad_norm": 2.552121162414551,
      "learning_rate": 4.510848877046061e-06,
      "loss": 1.6334,
      "step": 564
    },
    {
      "epoch": 0.5803800719054957,
      "grad_norm": 2.697615146636963,
      "learning_rate": 4.5089455652835936e-06,
      "loss": 1.5951,
      "step": 565
    },
    {
      "epoch": 0.5814072932717,
      "grad_norm": 2.64585280418396,
      "learning_rate": 4.507042253521127e-06,
      "loss": 1.556,
      "step": 566
    },
    {
      "epoch": 0.5824345146379045,
      "grad_norm": 2.6383707523345947,
      "learning_rate": 4.50513894175866e-06,
      "loss": 1.5184,
      "step": 567
    },
    {
      "epoch": 0.5834617360041089,
      "grad_norm": 2.7571511268615723,
      "learning_rate": 4.503235629996194e-06,
      "loss": 1.5984,
      "step": 568
    },
    {
      "epoch": 0.5844889573703133,
      "grad_norm": 2.6637237071990967,
      "learning_rate": 4.501332318233727e-06,
      "loss": 1.5989,
      "step": 569
    },
    {
      "epoch": 0.5855161787365177,
      "grad_norm": 2.725754976272583,
      "learning_rate": 4.49942900647126e-06,
      "loss": 1.5594,
      "step": 570
    },
    {
      "epoch": 0.5855161787365177,
      "eval_loss": 1.8722314834594727,
      "eval_runtime": 37.1859,
      "eval_samples_per_second": 26.892,
      "eval_steps_per_second": 1.129,
      "step": 570
    },
    {
      "epoch": 0.5865434001027221,
      "grad_norm": 2.6657378673553467,
      "learning_rate": 4.497525694708794e-06,
      "loss": 1.5954,
      "step": 571
    },
    {
      "epoch": 0.5875706214689266,
      "grad_norm": 2.78940486907959,
      "learning_rate": 4.495622382946327e-06,
      "loss": 1.592,
      "step": 572
    },
    {
      "epoch": 0.588597842835131,
      "grad_norm": 2.580956220626831,
      "learning_rate": 4.4937190711838605e-06,
      "loss": 1.5444,
      "step": 573
    },
    {
      "epoch": 0.5896250642013354,
      "grad_norm": 2.7783710956573486,
      "learning_rate": 4.491815759421394e-06,
      "loss": 1.6257,
      "step": 574
    },
    {
      "epoch": 0.5906522855675398,
      "grad_norm": 2.9381537437438965,
      "learning_rate": 4.489912447658926e-06,
      "loss": 1.5489,
      "step": 575
    },
    {
      "epoch": 0.5916795069337443,
      "grad_norm": 2.7915165424346924,
      "learning_rate": 4.48800913589646e-06,
      "loss": 1.5471,
      "step": 576
    },
    {
      "epoch": 0.5927067282999486,
      "grad_norm": 2.5928280353546143,
      "learning_rate": 4.486105824133993e-06,
      "loss": 1.5864,
      "step": 577
    },
    {
      "epoch": 0.593733949666153,
      "grad_norm": 2.844512462615967,
      "learning_rate": 4.4842025123715264e-06,
      "loss": 1.5841,
      "step": 578
    },
    {
      "epoch": 0.5947611710323575,
      "grad_norm": 2.562718391418457,
      "learning_rate": 4.48229920060906e-06,
      "loss": 1.5476,
      "step": 579
    },
    {
      "epoch": 0.5957883923985619,
      "grad_norm": 2.535759925842285,
      "learning_rate": 4.480395888846594e-06,
      "loss": 1.5274,
      "step": 580
    },
    {
      "epoch": 0.5957883923985619,
      "eval_loss": 1.8881579637527466,
      "eval_runtime": 37.2533,
      "eval_samples_per_second": 26.843,
      "eval_steps_per_second": 1.127,
      "step": 580
    },
    {
      "epoch": 0.5968156137647663,
      "grad_norm": 2.6940653324127197,
      "learning_rate": 4.478492577084127e-06,
      "loss": 1.6377,
      "step": 581
    },
    {
      "epoch": 0.5978428351309707,
      "grad_norm": 2.4839298725128174,
      "learning_rate": 4.47658926532166e-06,
      "loss": 1.534,
      "step": 582
    },
    {
      "epoch": 0.5988700564971752,
      "grad_norm": 2.5035688877105713,
      "learning_rate": 4.474685953559193e-06,
      "loss": 1.5764,
      "step": 583
    },
    {
      "epoch": 0.5998972778633795,
      "grad_norm": 2.791907548904419,
      "learning_rate": 4.472782641796727e-06,
      "loss": 1.704,
      "step": 584
    },
    {
      "epoch": 0.600924499229584,
      "grad_norm": 2.5384254455566406,
      "learning_rate": 4.47087933003426e-06,
      "loss": 1.5184,
      "step": 585
    },
    {
      "epoch": 0.6019517205957884,
      "grad_norm": 2.4990930557250977,
      "learning_rate": 4.468976018271793e-06,
      "loss": 1.5335,
      "step": 586
    },
    {
      "epoch": 0.6029789419619929,
      "grad_norm": 2.470872163772583,
      "learning_rate": 4.467072706509327e-06,
      "loss": 1.5464,
      "step": 587
    },
    {
      "epoch": 0.6040061633281972,
      "grad_norm": 2.539036989212036,
      "learning_rate": 4.46516939474686e-06,
      "loss": 1.5793,
      "step": 588
    },
    {
      "epoch": 0.6050333846944016,
      "grad_norm": 2.495309829711914,
      "learning_rate": 4.463266082984393e-06,
      "loss": 1.5312,
      "step": 589
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.5937912464141846,
      "learning_rate": 4.461362771221927e-06,
      "loss": 1.6508,
      "step": 590
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 1.8893816471099854,
      "eval_runtime": 37.2539,
      "eval_samples_per_second": 26.843,
      "eval_steps_per_second": 1.127,
      "step": 590
    },
    {
      "epoch": 0.6070878274268104,
      "grad_norm": 2.53300404548645,
      "learning_rate": 4.45945945945946e-06,
      "loss": 1.5408,
      "step": 591
    },
    {
      "epoch": 0.6081150487930149,
      "grad_norm": 2.5760276317596436,
      "learning_rate": 4.457556147696993e-06,
      "loss": 1.5575,
      "step": 592
    },
    {
      "epoch": 0.6091422701592193,
      "grad_norm": 2.638353109359741,
      "learning_rate": 4.455652835934526e-06,
      "loss": 1.6286,
      "step": 593
    },
    {
      "epoch": 0.6101694915254238,
      "grad_norm": 2.916550397872925,
      "learning_rate": 4.453749524172059e-06,
      "loss": 1.6334,
      "step": 594
    },
    {
      "epoch": 0.6111967128916281,
      "grad_norm": 2.560878038406372,
      "learning_rate": 4.451846212409593e-06,
      "loss": 1.5806,
      "step": 595
    },
    {
      "epoch": 0.6122239342578326,
      "grad_norm": 2.637882947921753,
      "learning_rate": 4.449942900647126e-06,
      "loss": 1.5791,
      "step": 596
    },
    {
      "epoch": 0.613251155624037,
      "grad_norm": 2.752870798110962,
      "learning_rate": 4.4480395888846595e-06,
      "loss": 1.5117,
      "step": 597
    },
    {
      "epoch": 0.6142783769902413,
      "grad_norm": 2.702087640762329,
      "learning_rate": 4.446136277122193e-06,
      "loss": 1.5783,
      "step": 598
    },
    {
      "epoch": 0.6153055983564458,
      "grad_norm": 2.504819631576538,
      "learning_rate": 4.444232965359726e-06,
      "loss": 1.5136,
      "step": 599
    },
    {
      "epoch": 0.6163328197226502,
      "grad_norm": 2.521670341491699,
      "learning_rate": 4.4423296535972596e-06,
      "loss": 1.5741,
      "step": 600
    },
    {
      "epoch": 0.6163328197226502,
      "eval_loss": 1.8973714113235474,
      "eval_runtime": 37.3167,
      "eval_samples_per_second": 26.798,
      "eval_steps_per_second": 1.126,
      "step": 600
    },
    {
      "epoch": 0.6173600410888547,
      "grad_norm": 2.927755355834961,
      "learning_rate": 4.440426341834793e-06,
      "loss": 1.5262,
      "step": 601
    },
    {
      "epoch": 0.618387262455059,
      "grad_norm": 2.477053642272949,
      "learning_rate": 4.438523030072326e-06,
      "loss": 1.572,
      "step": 602
    },
    {
      "epoch": 0.6194144838212635,
      "grad_norm": 2.6500253677368164,
      "learning_rate": 4.43661971830986e-06,
      "loss": 1.596,
      "step": 603
    },
    {
      "epoch": 0.6204417051874679,
      "grad_norm": 2.5134825706481934,
      "learning_rate": 4.434716406547393e-06,
      "loss": 1.5432,
      "step": 604
    },
    {
      "epoch": 0.6214689265536724,
      "grad_norm": 2.4142072200775146,
      "learning_rate": 4.432813094784926e-06,
      "loss": 1.5391,
      "step": 605
    },
    {
      "epoch": 0.6224961479198767,
      "grad_norm": 2.752556800842285,
      "learning_rate": 4.43090978302246e-06,
      "loss": 1.6055,
      "step": 606
    },
    {
      "epoch": 0.6235233692860811,
      "grad_norm": 2.5731008052825928,
      "learning_rate": 4.429006471259993e-06,
      "loss": 1.5648,
      "step": 607
    },
    {
      "epoch": 0.6245505906522856,
      "grad_norm": 2.531656503677368,
      "learning_rate": 4.4271031594975265e-06,
      "loss": 1.5492,
      "step": 608
    },
    {
      "epoch": 0.6255778120184899,
      "grad_norm": 2.6665475368499756,
      "learning_rate": 4.425199847735059e-06,
      "loss": 1.5194,
      "step": 609
    },
    {
      "epoch": 0.6266050333846944,
      "grad_norm": 2.437051773071289,
      "learning_rate": 4.423296535972592e-06,
      "loss": 1.5271,
      "step": 610
    },
    {
      "epoch": 0.6266050333846944,
      "eval_loss": 1.88953697681427,
      "eval_runtime": 36.8703,
      "eval_samples_per_second": 27.122,
      "eval_steps_per_second": 1.139,
      "step": 610
    },
    {
      "epoch": 0.6276322547508988,
      "grad_norm": 2.5932440757751465,
      "learning_rate": 4.421393224210126e-06,
      "loss": 1.5563,
      "step": 611
    },
    {
      "epoch": 0.6286594761171033,
      "grad_norm": 2.589447021484375,
      "learning_rate": 4.419489912447659e-06,
      "loss": 1.5207,
      "step": 612
    },
    {
      "epoch": 0.6296866974833076,
      "grad_norm": 2.570096731185913,
      "learning_rate": 4.4175866006851924e-06,
      "loss": 1.5995,
      "step": 613
    },
    {
      "epoch": 0.6307139188495121,
      "grad_norm": 2.5356791019439697,
      "learning_rate": 4.415683288922726e-06,
      "loss": 1.5703,
      "step": 614
    },
    {
      "epoch": 0.6317411402157165,
      "grad_norm": 2.705193281173706,
      "learning_rate": 4.413779977160259e-06,
      "loss": 1.5405,
      "step": 615
    },
    {
      "epoch": 0.632768361581921,
      "grad_norm": 2.555168390274048,
      "learning_rate": 4.4118766653977925e-06,
      "loss": 1.5487,
      "step": 616
    },
    {
      "epoch": 0.6337955829481253,
      "grad_norm": 2.804696559906006,
      "learning_rate": 4.409973353635326e-06,
      "loss": 1.6094,
      "step": 617
    },
    {
      "epoch": 0.6348228043143297,
      "grad_norm": 2.6889376640319824,
      "learning_rate": 4.408070041872859e-06,
      "loss": 1.5656,
      "step": 618
    },
    {
      "epoch": 0.6358500256805342,
      "grad_norm": 2.576746702194214,
      "learning_rate": 4.406166730110393e-06,
      "loss": 1.5244,
      "step": 619
    },
    {
      "epoch": 0.6368772470467385,
      "grad_norm": 2.811156749725342,
      "learning_rate": 4.404263418347925e-06,
      "loss": 1.5153,
      "step": 620
    },
    {
      "epoch": 0.6368772470467385,
      "eval_loss": 1.8949775695800781,
      "eval_runtime": 36.978,
      "eval_samples_per_second": 27.043,
      "eval_steps_per_second": 1.136,
      "step": 620
    },
    {
      "epoch": 0.637904468412943,
      "grad_norm": 2.6560428142547607,
      "learning_rate": 4.402360106585459e-06,
      "loss": 1.5347,
      "step": 621
    },
    {
      "epoch": 0.6389316897791474,
      "grad_norm": 2.668100118637085,
      "learning_rate": 4.400456794822993e-06,
      "loss": 1.5236,
      "step": 622
    },
    {
      "epoch": 0.6399589111453519,
      "grad_norm": 2.925978660583496,
      "learning_rate": 4.398553483060526e-06,
      "loss": 1.5494,
      "step": 623
    },
    {
      "epoch": 0.6409861325115562,
      "grad_norm": 2.5964415073394775,
      "learning_rate": 4.396650171298059e-06,
      "loss": 1.522,
      "step": 624
    },
    {
      "epoch": 0.6420133538777607,
      "grad_norm": 2.643888235092163,
      "learning_rate": 4.394746859535593e-06,
      "loss": 1.5332,
      "step": 625
    },
    {
      "epoch": 0.6430405752439651,
      "grad_norm": 2.805192708969116,
      "learning_rate": 4.392843547773125e-06,
      "loss": 1.5031,
      "step": 626
    },
    {
      "epoch": 0.6440677966101694,
      "grad_norm": 2.6862149238586426,
      "learning_rate": 4.390940236010659e-06,
      "loss": 1.5335,
      "step": 627
    },
    {
      "epoch": 0.6450950179763739,
      "grad_norm": 2.5579185485839844,
      "learning_rate": 4.389036924248192e-06,
      "loss": 1.5466,
      "step": 628
    },
    {
      "epoch": 0.6461222393425783,
      "grad_norm": 2.4550797939300537,
      "learning_rate": 4.387133612485725e-06,
      "loss": 1.5776,
      "step": 629
    },
    {
      "epoch": 0.6471494607087828,
      "grad_norm": 2.6051297187805176,
      "learning_rate": 4.385230300723259e-06,
      "loss": 1.5435,
      "step": 630
    },
    {
      "epoch": 0.6471494607087828,
      "eval_loss": 1.8860505819320679,
      "eval_runtime": 36.989,
      "eval_samples_per_second": 27.035,
      "eval_steps_per_second": 1.135,
      "step": 630
    },
    {
      "epoch": 0.6481766820749871,
      "grad_norm": 2.3723294734954834,
      "learning_rate": 4.383326988960792e-06,
      "loss": 1.5413,
      "step": 631
    },
    {
      "epoch": 0.6492039034411916,
      "grad_norm": 2.44370174407959,
      "learning_rate": 4.3814236771983255e-06,
      "loss": 1.5152,
      "step": 632
    },
    {
      "epoch": 0.650231124807396,
      "grad_norm": 2.474081039428711,
      "learning_rate": 4.379520365435859e-06,
      "loss": 1.579,
      "step": 633
    },
    {
      "epoch": 0.6512583461736005,
      "grad_norm": 2.837843179702759,
      "learning_rate": 4.377617053673392e-06,
      "loss": 1.559,
      "step": 634
    },
    {
      "epoch": 0.6522855675398048,
      "grad_norm": 2.593322515487671,
      "learning_rate": 4.3757137419109256e-06,
      "loss": 1.552,
      "step": 635
    },
    {
      "epoch": 0.6533127889060092,
      "grad_norm": 3.018051862716675,
      "learning_rate": 4.373810430148459e-06,
      "loss": 1.6231,
      "step": 636
    },
    {
      "epoch": 0.6543400102722137,
      "grad_norm": 2.6338863372802734,
      "learning_rate": 4.3719071183859914e-06,
      "loss": 1.517,
      "step": 637
    },
    {
      "epoch": 0.655367231638418,
      "grad_norm": 2.544140577316284,
      "learning_rate": 4.370003806623525e-06,
      "loss": 1.5763,
      "step": 638
    },
    {
      "epoch": 0.6563944530046225,
      "grad_norm": 2.6375792026519775,
      "learning_rate": 4.368100494861058e-06,
      "loss": 1.5098,
      "step": 639
    },
    {
      "epoch": 0.6574216743708269,
      "grad_norm": 2.5452051162719727,
      "learning_rate": 4.3661971830985915e-06,
      "loss": 1.5671,
      "step": 640
    },
    {
      "epoch": 0.6574216743708269,
      "eval_loss": 1.881069540977478,
      "eval_runtime": 36.9961,
      "eval_samples_per_second": 27.03,
      "eval_steps_per_second": 1.135,
      "step": 640
    },
    {
      "epoch": 0.6584488957370314,
      "grad_norm": 2.623411178588867,
      "learning_rate": 4.364293871336126e-06,
      "loss": 1.5215,
      "step": 641
    },
    {
      "epoch": 0.6594761171032357,
      "grad_norm": 2.6149892807006836,
      "learning_rate": 4.362390559573659e-06,
      "loss": 1.5507,
      "step": 642
    },
    {
      "epoch": 0.6605033384694402,
      "grad_norm": 2.5570266246795654,
      "learning_rate": 4.3604872478111925e-06,
      "loss": 1.5143,
      "step": 643
    },
    {
      "epoch": 0.6615305598356446,
      "grad_norm": 2.4813005924224854,
      "learning_rate": 4.358583936048725e-06,
      "loss": 1.4518,
      "step": 644
    },
    {
      "epoch": 0.662557781201849,
      "grad_norm": 3.0318472385406494,
      "learning_rate": 4.356680624286258e-06,
      "loss": 1.5812,
      "step": 645
    },
    {
      "epoch": 0.6635850025680534,
      "grad_norm": 2.46936297416687,
      "learning_rate": 4.354777312523792e-06,
      "loss": 1.5396,
      "step": 646
    },
    {
      "epoch": 0.6646122239342578,
      "grad_norm": 2.5295984745025635,
      "learning_rate": 4.352874000761325e-06,
      "loss": 1.586,
      "step": 647
    },
    {
      "epoch": 0.6656394453004623,
      "grad_norm": 2.7826883792877197,
      "learning_rate": 4.3509706889988584e-06,
      "loss": 1.5339,
      "step": 648
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.4273526668548584,
      "learning_rate": 4.349067377236392e-06,
      "loss": 1.598,
      "step": 649
    },
    {
      "epoch": 0.6676938880328711,
      "grad_norm": 2.644308090209961,
      "learning_rate": 4.347164065473925e-06,
      "loss": 1.5889,
      "step": 650
    },
    {
      "epoch": 0.6676938880328711,
      "eval_loss": 1.8861087560653687,
      "eval_runtime": 36.99,
      "eval_samples_per_second": 27.034,
      "eval_steps_per_second": 1.135,
      "step": 650
    },
    {
      "epoch": 0.6687211093990755,
      "grad_norm": 2.827376127243042,
      "learning_rate": 4.3452607537114585e-06,
      "loss": 1.606,
      "step": 651
    },
    {
      "epoch": 0.66974833076528,
      "grad_norm": 2.7463080883026123,
      "learning_rate": 4.343357441948992e-06,
      "loss": 1.5208,
      "step": 652
    },
    {
      "epoch": 0.6707755521314843,
      "grad_norm": 2.6082730293273926,
      "learning_rate": 4.341454130186525e-06,
      "loss": 1.5597,
      "step": 653
    },
    {
      "epoch": 0.6718027734976888,
      "grad_norm": 2.7131078243255615,
      "learning_rate": 4.339550818424058e-06,
      "loss": 1.526,
      "step": 654
    },
    {
      "epoch": 0.6728299948638932,
      "grad_norm": 2.6635708808898926,
      "learning_rate": 4.337647506661591e-06,
      "loss": 1.5052,
      "step": 655
    },
    {
      "epoch": 0.6738572162300975,
      "grad_norm": 2.5686581134796143,
      "learning_rate": 4.3357441948991245e-06,
      "loss": 1.5374,
      "step": 656
    },
    {
      "epoch": 0.674884437596302,
      "grad_norm": 3.0532894134521484,
      "learning_rate": 4.333840883136658e-06,
      "loss": 1.5787,
      "step": 657
    },
    {
      "epoch": 0.6759116589625064,
      "grad_norm": 2.6494507789611816,
      "learning_rate": 4.331937571374191e-06,
      "loss": 1.5121,
      "step": 658
    },
    {
      "epoch": 0.6769388803287109,
      "grad_norm": 2.748142719268799,
      "learning_rate": 4.3300342596117246e-06,
      "loss": 1.5294,
      "step": 659
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 2.837799310684204,
      "learning_rate": 4.328130947849258e-06,
      "loss": 1.5424,
      "step": 660
    },
    {
      "epoch": 0.6779661016949152,
      "eval_loss": 1.8644919395446777,
      "eval_runtime": 37.0096,
      "eval_samples_per_second": 27.02,
      "eval_steps_per_second": 1.135,
      "step": 660
    },
    {
      "epoch": 0.6789933230611197,
      "grad_norm": 2.7405757904052734,
      "learning_rate": 4.326227636086791e-06,
      "loss": 1.4949,
      "step": 661
    },
    {
      "epoch": 0.6800205444273241,
      "grad_norm": 2.545466661453247,
      "learning_rate": 4.324324324324325e-06,
      "loss": 1.5445,
      "step": 662
    },
    {
      "epoch": 0.6810477657935285,
      "grad_norm": 2.5715537071228027,
      "learning_rate": 4.322421012561858e-06,
      "loss": 1.5468,
      "step": 663
    },
    {
      "epoch": 0.6820749871597329,
      "grad_norm": 2.7073662281036377,
      "learning_rate": 4.320517700799391e-06,
      "loss": 1.5293,
      "step": 664
    },
    {
      "epoch": 0.6831022085259374,
      "grad_norm": 2.62264347076416,
      "learning_rate": 4.318614389036925e-06,
      "loss": 1.6258,
      "step": 665
    },
    {
      "epoch": 0.6841294298921418,
      "grad_norm": 2.4506373405456543,
      "learning_rate": 4.316711077274458e-06,
      "loss": 1.5845,
      "step": 666
    },
    {
      "epoch": 0.6851566512583461,
      "grad_norm": 2.849473237991333,
      "learning_rate": 4.3148077655119915e-06,
      "loss": 1.4959,
      "step": 667
    },
    {
      "epoch": 0.6861838726245506,
      "grad_norm": 2.5865421295166016,
      "learning_rate": 4.312904453749525e-06,
      "loss": 1.599,
      "step": 668
    },
    {
      "epoch": 0.687211093990755,
      "grad_norm": 2.4636387825012207,
      "learning_rate": 4.311001141987058e-06,
      "loss": 1.5495,
      "step": 669
    },
    {
      "epoch": 0.6882383153569595,
      "grad_norm": 2.638047218322754,
      "learning_rate": 4.3090978302245916e-06,
      "loss": 1.5507,
      "step": 670
    },
    {
      "epoch": 0.6882383153569595,
      "eval_loss": 1.8834826946258545,
      "eval_runtime": 37.0335,
      "eval_samples_per_second": 27.003,
      "eval_steps_per_second": 1.134,
      "step": 670
    },
    {
      "epoch": 0.6892655367231638,
      "grad_norm": 2.669201612472534,
      "learning_rate": 4.307194518462124e-06,
      "loss": 1.5494,
      "step": 671
    },
    {
      "epoch": 0.6902927580893683,
      "grad_norm": 2.634634494781494,
      "learning_rate": 4.3052912066996574e-06,
      "loss": 1.6177,
      "step": 672
    },
    {
      "epoch": 0.6913199794555727,
      "grad_norm": 2.5489139556884766,
      "learning_rate": 4.303387894937191e-06,
      "loss": 1.5876,
      "step": 673
    },
    {
      "epoch": 0.6923472008217771,
      "grad_norm": 2.580073833465576,
      "learning_rate": 4.301484583174724e-06,
      "loss": 1.5106,
      "step": 674
    },
    {
      "epoch": 0.6933744221879815,
      "grad_norm": 2.5126636028289795,
      "learning_rate": 4.2995812714122575e-06,
      "loss": 1.4988,
      "step": 675
    },
    {
      "epoch": 0.6944016435541859,
      "grad_norm": 2.4630191326141357,
      "learning_rate": 4.297677959649791e-06,
      "loss": 1.5218,
      "step": 676
    },
    {
      "epoch": 0.6954288649203904,
      "grad_norm": 2.8822433948516846,
      "learning_rate": 4.295774647887324e-06,
      "loss": 1.5592,
      "step": 677
    },
    {
      "epoch": 0.6964560862865947,
      "grad_norm": 2.5498058795928955,
      "learning_rate": 4.293871336124858e-06,
      "loss": 1.5713,
      "step": 678
    },
    {
      "epoch": 0.6974833076527992,
      "grad_norm": 2.6661157608032227,
      "learning_rate": 4.291968024362391e-06,
      "loss": 1.5431,
      "step": 679
    },
    {
      "epoch": 0.6985105290190036,
      "grad_norm": 2.617971897125244,
      "learning_rate": 4.290064712599924e-06,
      "loss": 1.5792,
      "step": 680
    },
    {
      "epoch": 0.6985105290190036,
      "eval_loss": 1.8702285289764404,
      "eval_runtime": 37.0376,
      "eval_samples_per_second": 27.0,
      "eval_steps_per_second": 1.134,
      "step": 680
    },
    {
      "epoch": 0.699537750385208,
      "grad_norm": 2.5892348289489746,
      "learning_rate": 4.288161400837458e-06,
      "loss": 1.5626,
      "step": 681
    },
    {
      "epoch": 0.7005649717514124,
      "grad_norm": 2.5460269451141357,
      "learning_rate": 4.286258089074991e-06,
      "loss": 1.4876,
      "step": 682
    },
    {
      "epoch": 0.7015921931176169,
      "grad_norm": 2.601632595062256,
      "learning_rate": 4.284354777312524e-06,
      "loss": 1.5547,
      "step": 683
    },
    {
      "epoch": 0.7026194144838213,
      "grad_norm": 2.6032373905181885,
      "learning_rate": 4.282451465550058e-06,
      "loss": 1.5133,
      "step": 684
    },
    {
      "epoch": 0.7036466358500256,
      "grad_norm": 2.6283886432647705,
      "learning_rate": 4.280548153787591e-06,
      "loss": 1.5375,
      "step": 685
    },
    {
      "epoch": 0.7046738572162301,
      "grad_norm": 2.5892159938812256,
      "learning_rate": 4.2786448420251245e-06,
      "loss": 1.5338,
      "step": 686
    },
    {
      "epoch": 0.7057010785824345,
      "grad_norm": 2.518629789352417,
      "learning_rate": 4.276741530262658e-06,
      "loss": 1.5003,
      "step": 687
    },
    {
      "epoch": 0.706728299948639,
      "grad_norm": 2.8705344200134277,
      "learning_rate": 4.27483821850019e-06,
      "loss": 1.5178,
      "step": 688
    },
    {
      "epoch": 0.7077555213148433,
      "grad_norm": 2.7153639793395996,
      "learning_rate": 4.272934906737724e-06,
      "loss": 1.5914,
      "step": 689
    },
    {
      "epoch": 0.7087827426810478,
      "grad_norm": 2.4585697650909424,
      "learning_rate": 4.271031594975257e-06,
      "loss": 1.5726,
      "step": 690
    },
    {
      "epoch": 0.7087827426810478,
      "eval_loss": 1.8637797832489014,
      "eval_runtime": 37.0465,
      "eval_samples_per_second": 26.993,
      "eval_steps_per_second": 1.134,
      "step": 690
    },
    {
      "epoch": 0.7098099640472522,
      "grad_norm": 2.6370503902435303,
      "learning_rate": 4.2691282832127905e-06,
      "loss": 1.5706,
      "step": 691
    },
    {
      "epoch": 0.7108371854134566,
      "grad_norm": 2.70202898979187,
      "learning_rate": 4.267224971450324e-06,
      "loss": 1.5298,
      "step": 692
    },
    {
      "epoch": 0.711864406779661,
      "grad_norm": 2.6149585247039795,
      "learning_rate": 4.265321659687857e-06,
      "loss": 1.477,
      "step": 693
    },
    {
      "epoch": 0.7128916281458655,
      "grad_norm": 2.9561493396759033,
      "learning_rate": 4.2634183479253906e-06,
      "loss": 1.5189,
      "step": 694
    },
    {
      "epoch": 0.7139188495120699,
      "grad_norm": 2.697680711746216,
      "learning_rate": 4.261515036162924e-06,
      "loss": 1.5001,
      "step": 695
    },
    {
      "epoch": 0.7149460708782742,
      "grad_norm": 2.55727219581604,
      "learning_rate": 4.259611724400457e-06,
      "loss": 1.533,
      "step": 696
    },
    {
      "epoch": 0.7159732922444787,
      "grad_norm": 2.854290723800659,
      "learning_rate": 4.257708412637991e-06,
      "loss": 1.5057,
      "step": 697
    },
    {
      "epoch": 0.7170005136106831,
      "grad_norm": 2.73824405670166,
      "learning_rate": 4.255805100875523e-06,
      "loss": 1.4567,
      "step": 698
    },
    {
      "epoch": 0.7180277349768875,
      "grad_norm": 2.730050802230835,
      "learning_rate": 4.2539017891130565e-06,
      "loss": 1.4873,
      "step": 699
    },
    {
      "epoch": 0.7190549563430919,
      "grad_norm": 2.966660737991333,
      "learning_rate": 4.25199847735059e-06,
      "loss": 1.574,
      "step": 700
    },
    {
      "epoch": 0.7190549563430919,
      "eval_loss": 1.8623589277267456,
      "eval_runtime": 37.0761,
      "eval_samples_per_second": 26.972,
      "eval_steps_per_second": 1.133,
      "step": 700
    },
    {
      "epoch": 0.7200821777092964,
      "grad_norm": 2.7526276111602783,
      "learning_rate": 4.250095165588123e-06,
      "loss": 1.4831,
      "step": 701
    },
    {
      "epoch": 0.7211093990755008,
      "grad_norm": 2.9722564220428467,
      "learning_rate": 4.2481918538256575e-06,
      "loss": 1.548,
      "step": 702
    },
    {
      "epoch": 0.7221366204417052,
      "grad_norm": 2.679373025894165,
      "learning_rate": 4.246288542063191e-06,
      "loss": 1.5145,
      "step": 703
    },
    {
      "epoch": 0.7231638418079096,
      "grad_norm": 2.733139753341675,
      "learning_rate": 4.244385230300724e-06,
      "loss": 1.5542,
      "step": 704
    },
    {
      "epoch": 0.724191063174114,
      "grad_norm": 2.6549174785614014,
      "learning_rate": 4.242481918538257e-06,
      "loss": 1.4958,
      "step": 705
    },
    {
      "epoch": 0.7252182845403184,
      "grad_norm": 2.562958002090454,
      "learning_rate": 4.24057860677579e-06,
      "loss": 1.524,
      "step": 706
    },
    {
      "epoch": 0.7262455059065228,
      "grad_norm": 2.641123056411743,
      "learning_rate": 4.2386752950133234e-06,
      "loss": 1.5723,
      "step": 707
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.8128373622894287,
      "learning_rate": 4.236771983250857e-06,
      "loss": 1.5151,
      "step": 708
    },
    {
      "epoch": 0.7282999486389317,
      "grad_norm": 2.676765203475952,
      "learning_rate": 4.23486867148839e-06,
      "loss": 1.5024,
      "step": 709
    },
    {
      "epoch": 0.7293271700051361,
      "grad_norm": 2.917178153991699,
      "learning_rate": 4.2329653597259235e-06,
      "loss": 1.5326,
      "step": 710
    },
    {
      "epoch": 0.7293271700051361,
      "eval_loss": 1.8676133155822754,
      "eval_runtime": 36.8304,
      "eval_samples_per_second": 27.152,
      "eval_steps_per_second": 1.14,
      "step": 710
    },
    {
      "epoch": 0.7303543913713405,
      "grad_norm": 2.8313796520233154,
      "learning_rate": 4.231062047963457e-06,
      "loss": 1.5623,
      "step": 711
    },
    {
      "epoch": 0.731381612737545,
      "grad_norm": 2.61542010307312,
      "learning_rate": 4.22915873620099e-06,
      "loss": 1.5725,
      "step": 712
    },
    {
      "epoch": 0.7324088341037494,
      "grad_norm": 2.5788609981536865,
      "learning_rate": 4.227255424438524e-06,
      "loss": 1.5381,
      "step": 713
    },
    {
      "epoch": 0.7334360554699538,
      "grad_norm": 2.720306158065796,
      "learning_rate": 4.225352112676057e-06,
      "loss": 1.572,
      "step": 714
    },
    {
      "epoch": 0.7344632768361582,
      "grad_norm": 2.780654191970825,
      "learning_rate": 4.2234488009135895e-06,
      "loss": 1.5084,
      "step": 715
    },
    {
      "epoch": 0.7354904982023626,
      "grad_norm": 2.4448273181915283,
      "learning_rate": 4.221545489151123e-06,
      "loss": 1.4614,
      "step": 716
    },
    {
      "epoch": 0.736517719568567,
      "grad_norm": 2.6579184532165527,
      "learning_rate": 4.219642177388656e-06,
      "loss": 1.5022,
      "step": 717
    },
    {
      "epoch": 0.7375449409347714,
      "grad_norm": 2.7092721462249756,
      "learning_rate": 4.2177388656261896e-06,
      "loss": 1.5307,
      "step": 718
    },
    {
      "epoch": 0.7385721623009759,
      "grad_norm": 2.65014910697937,
      "learning_rate": 4.215835553863723e-06,
      "loss": 1.4936,
      "step": 719
    },
    {
      "epoch": 0.7395993836671803,
      "grad_norm": 2.7403628826141357,
      "learning_rate": 4.213932242101256e-06,
      "loss": 1.4458,
      "step": 720
    },
    {
      "epoch": 0.7395993836671803,
      "eval_loss": 1.8778676986694336,
      "eval_runtime": 36.9378,
      "eval_samples_per_second": 27.073,
      "eval_steps_per_second": 1.137,
      "step": 720
    },
    {
      "epoch": 0.7406266050333847,
      "grad_norm": 2.3937759399414062,
      "learning_rate": 4.21202893033879e-06,
      "loss": 1.5086,
      "step": 721
    },
    {
      "epoch": 0.7416538263995891,
      "grad_norm": 2.5786867141723633,
      "learning_rate": 4.210125618576323e-06,
      "loss": 1.5321,
      "step": 722
    },
    {
      "epoch": 0.7426810477657936,
      "grad_norm": 2.607236862182617,
      "learning_rate": 4.208222306813856e-06,
      "loss": 1.5244,
      "step": 723
    },
    {
      "epoch": 0.743708269131998,
      "grad_norm": 2.5461602210998535,
      "learning_rate": 4.20631899505139e-06,
      "loss": 1.5841,
      "step": 724
    },
    {
      "epoch": 0.7447354904982023,
      "grad_norm": 2.5051612854003906,
      "learning_rate": 4.204415683288923e-06,
      "loss": 1.43,
      "step": 725
    },
    {
      "epoch": 0.7457627118644068,
      "grad_norm": 2.692816734313965,
      "learning_rate": 4.2025123715264565e-06,
      "loss": 1.5439,
      "step": 726
    },
    {
      "epoch": 0.7467899332306112,
      "grad_norm": 2.4827022552490234,
      "learning_rate": 4.20060905976399e-06,
      "loss": 1.4987,
      "step": 727
    },
    {
      "epoch": 0.7478171545968156,
      "grad_norm": 2.7261135578155518,
      "learning_rate": 4.198705748001523e-06,
      "loss": 1.4573,
      "step": 728
    },
    {
      "epoch": 0.74884437596302,
      "grad_norm": 2.5710837841033936,
      "learning_rate": 4.1968024362390566e-06,
      "loss": 1.564,
      "step": 729
    },
    {
      "epoch": 0.7498715973292245,
      "grad_norm": 2.61181640625,
      "learning_rate": 4.19489912447659e-06,
      "loss": 1.5211,
      "step": 730
    },
    {
      "epoch": 0.7498715973292245,
      "eval_loss": 1.8794434070587158,
      "eval_runtime": 36.9676,
      "eval_samples_per_second": 27.051,
      "eval_steps_per_second": 1.136,
      "step": 730
    },
    {
      "epoch": 0.7508988186954288,
      "grad_norm": 2.6946470737457275,
      "learning_rate": 4.192995812714123e-06,
      "loss": 1.5837,
      "step": 731
    },
    {
      "epoch": 0.7519260400616333,
      "grad_norm": 2.5337390899658203,
      "learning_rate": 4.191092500951656e-06,
      "loss": 1.5251,
      "step": 732
    },
    {
      "epoch": 0.7529532614278377,
      "grad_norm": 2.541532278060913,
      "learning_rate": 4.189189189189189e-06,
      "loss": 1.503,
      "step": 733
    },
    {
      "epoch": 0.7539804827940421,
      "grad_norm": 2.639124631881714,
      "learning_rate": 4.1872858774267225e-06,
      "loss": 1.4963,
      "step": 734
    },
    {
      "epoch": 0.7550077041602465,
      "grad_norm": 2.872861862182617,
      "learning_rate": 4.185382565664256e-06,
      "loss": 1.5632,
      "step": 735
    },
    {
      "epoch": 0.7560349255264509,
      "grad_norm": 2.588980197906494,
      "learning_rate": 4.183479253901789e-06,
      "loss": 1.4998,
      "step": 736
    },
    {
      "epoch": 0.7570621468926554,
      "grad_norm": 2.9199588298797607,
      "learning_rate": 4.181575942139323e-06,
      "loss": 1.466,
      "step": 737
    },
    {
      "epoch": 0.7580893682588598,
      "grad_norm": 2.5677788257598877,
      "learning_rate": 4.179672630376856e-06,
      "loss": 1.5247,
      "step": 738
    },
    {
      "epoch": 0.7591165896250642,
      "grad_norm": 2.481142520904541,
      "learning_rate": 4.177769318614389e-06,
      "loss": 1.4915,
      "step": 739
    },
    {
      "epoch": 0.7601438109912686,
      "grad_norm": 2.7072012424468994,
      "learning_rate": 4.175866006851923e-06,
      "loss": 1.5783,
      "step": 740
    },
    {
      "epoch": 0.7601438109912686,
      "eval_loss": 1.8817057609558105,
      "eval_runtime": 36.9635,
      "eval_samples_per_second": 27.054,
      "eval_steps_per_second": 1.136,
      "step": 740
    },
    {
      "epoch": 0.7611710323574731,
      "grad_norm": 2.5433032512664795,
      "learning_rate": 4.173962695089456e-06,
      "loss": 1.5457,
      "step": 741
    },
    {
      "epoch": 0.7621982537236774,
      "grad_norm": 2.702077627182007,
      "learning_rate": 4.1720593833269894e-06,
      "loss": 1.5285,
      "step": 742
    },
    {
      "epoch": 0.7632254750898819,
      "grad_norm": 2.6678309440612793,
      "learning_rate": 4.170156071564523e-06,
      "loss": 1.5381,
      "step": 743
    },
    {
      "epoch": 0.7642526964560863,
      "grad_norm": 2.6097185611724854,
      "learning_rate": 4.168252759802056e-06,
      "loss": 1.6076,
      "step": 744
    },
    {
      "epoch": 0.7652799178222907,
      "grad_norm": 2.5975704193115234,
      "learning_rate": 4.1663494480395895e-06,
      "loss": 1.515,
      "step": 745
    },
    {
      "epoch": 0.7663071391884951,
      "grad_norm": 2.6858558654785156,
      "learning_rate": 4.164446136277123e-06,
      "loss": 1.5499,
      "step": 746
    },
    {
      "epoch": 0.7673343605546995,
      "grad_norm": 2.565023899078369,
      "learning_rate": 4.162542824514656e-06,
      "loss": 1.5502,
      "step": 747
    },
    {
      "epoch": 0.768361581920904,
      "grad_norm": 2.61715030670166,
      "learning_rate": 4.16063951275219e-06,
      "loss": 1.5271,
      "step": 748
    },
    {
      "epoch": 0.7693888032871083,
      "grad_norm": 2.564382314682007,
      "learning_rate": 4.158736200989722e-06,
      "loss": 1.4756,
      "step": 749
    },
    {
      "epoch": 0.7704160246533128,
      "grad_norm": 2.64457368850708,
      "learning_rate": 4.1568328892272555e-06,
      "loss": 1.5246,
      "step": 750
    },
    {
      "epoch": 0.7704160246533128,
      "eval_loss": 1.8777766227722168,
      "eval_runtime": 36.9737,
      "eval_samples_per_second": 27.046,
      "eval_steps_per_second": 1.136,
      "step": 750
    },
    {
      "epoch": 0.7714432460195172,
      "grad_norm": 2.773602247238159,
      "learning_rate": 4.154929577464789e-06,
      "loss": 1.5337,
      "step": 751
    },
    {
      "epoch": 0.7724704673857217,
      "grad_norm": 2.5466997623443604,
      "learning_rate": 4.153026265702322e-06,
      "loss": 1.5269,
      "step": 752
    },
    {
      "epoch": 0.773497688751926,
      "grad_norm": 2.702888250350952,
      "learning_rate": 4.1511229539398556e-06,
      "loss": 1.4636,
      "step": 753
    },
    {
      "epoch": 0.7745249101181304,
      "grad_norm": 2.629251718521118,
      "learning_rate": 4.149219642177389e-06,
      "loss": 1.513,
      "step": 754
    },
    {
      "epoch": 0.7755521314843349,
      "grad_norm": 2.52544903755188,
      "learning_rate": 4.147316330414922e-06,
      "loss": 1.5225,
      "step": 755
    },
    {
      "epoch": 0.7765793528505393,
      "grad_norm": 2.4267778396606445,
      "learning_rate": 4.145413018652456e-06,
      "loss": 1.4377,
      "step": 756
    },
    {
      "epoch": 0.7776065742167437,
      "grad_norm": 2.6061577796936035,
      "learning_rate": 4.143509706889989e-06,
      "loss": 1.5402,
      "step": 757
    },
    {
      "epoch": 0.7786337955829481,
      "grad_norm": 2.6571085453033447,
      "learning_rate": 4.141606395127522e-06,
      "loss": 1.527,
      "step": 758
    },
    {
      "epoch": 0.7796610169491526,
      "grad_norm": 2.48740291595459,
      "learning_rate": 4.139703083365056e-06,
      "loss": 1.517,
      "step": 759
    },
    {
      "epoch": 0.7806882383153569,
      "grad_norm": 2.5044426918029785,
      "learning_rate": 4.137799771602588e-06,
      "loss": 1.5579,
      "step": 760
    },
    {
      "epoch": 0.7806882383153569,
      "eval_loss": 1.8778951168060303,
      "eval_runtime": 37.0045,
      "eval_samples_per_second": 27.024,
      "eval_steps_per_second": 1.135,
      "step": 760
    },
    {
      "epoch": 0.7817154596815614,
      "grad_norm": 2.4728074073791504,
      "learning_rate": 4.135896459840122e-06,
      "loss": 1.5321,
      "step": 761
    },
    {
      "epoch": 0.7827426810477658,
      "grad_norm": 2.6541659832000732,
      "learning_rate": 4.133993148077656e-06,
      "loss": 1.4866,
      "step": 762
    },
    {
      "epoch": 0.7837699024139703,
      "grad_norm": 2.526716947555542,
      "learning_rate": 4.132089836315189e-06,
      "loss": 1.5035,
      "step": 763
    },
    {
      "epoch": 0.7847971237801746,
      "grad_norm": 2.683306932449341,
      "learning_rate": 4.1301865245527226e-06,
      "loss": 1.5221,
      "step": 764
    },
    {
      "epoch": 0.785824345146379,
      "grad_norm": 2.4980056285858154,
      "learning_rate": 4.128283212790256e-06,
      "loss": 1.4643,
      "step": 765
    },
    {
      "epoch": 0.7868515665125835,
      "grad_norm": 2.3985488414764404,
      "learning_rate": 4.1263799010277884e-06,
      "loss": 1.4348,
      "step": 766
    },
    {
      "epoch": 0.7878787878787878,
      "grad_norm": 2.600689649581909,
      "learning_rate": 4.124476589265322e-06,
      "loss": 1.5647,
      "step": 767
    },
    {
      "epoch": 0.7889060092449923,
      "grad_norm": 2.5934395790100098,
      "learning_rate": 4.122573277502855e-06,
      "loss": 1.4887,
      "step": 768
    },
    {
      "epoch": 0.7899332306111967,
      "grad_norm": 2.5209388732910156,
      "learning_rate": 4.1206699657403885e-06,
      "loss": 1.4893,
      "step": 769
    },
    {
      "epoch": 0.7909604519774012,
      "grad_norm": 2.6656298637390137,
      "learning_rate": 4.118766653977922e-06,
      "loss": 1.5239,
      "step": 770
    },
    {
      "epoch": 0.7909604519774012,
      "eval_loss": 1.8737632036209106,
      "eval_runtime": 37.056,
      "eval_samples_per_second": 26.986,
      "eval_steps_per_second": 1.133,
      "step": 770
    },
    {
      "epoch": 0.7919876733436055,
      "grad_norm": 2.5147693157196045,
      "learning_rate": 4.116863342215455e-06,
      "loss": 1.5,
      "step": 771
    },
    {
      "epoch": 0.79301489470981,
      "grad_norm": 2.4653773307800293,
      "learning_rate": 4.114960030452989e-06,
      "loss": 1.4802,
      "step": 772
    },
    {
      "epoch": 0.7940421160760144,
      "grad_norm": 2.643160820007324,
      "learning_rate": 4.113056718690522e-06,
      "loss": 1.5086,
      "step": 773
    },
    {
      "epoch": 0.7950693374422187,
      "grad_norm": 2.4873125553131104,
      "learning_rate": 4.111153406928055e-06,
      "loss": 1.4558,
      "step": 774
    },
    {
      "epoch": 0.7960965588084232,
      "grad_norm": 2.541308879852295,
      "learning_rate": 4.109250095165589e-06,
      "loss": 1.5483,
      "step": 775
    },
    {
      "epoch": 0.7971237801746276,
      "grad_norm": 2.701874256134033,
      "learning_rate": 4.107346783403122e-06,
      "loss": 1.5551,
      "step": 776
    },
    {
      "epoch": 0.7981510015408321,
      "grad_norm": 2.457371234893799,
      "learning_rate": 4.105443471640655e-06,
      "loss": 1.4968,
      "step": 777
    },
    {
      "epoch": 0.7991782229070364,
      "grad_norm": 2.5294973850250244,
      "learning_rate": 4.103540159878188e-06,
      "loss": 1.4756,
      "step": 778
    },
    {
      "epoch": 0.8002054442732409,
      "grad_norm": 2.464932918548584,
      "learning_rate": 4.101636848115721e-06,
      "loss": 1.5215,
      "step": 779
    },
    {
      "epoch": 0.8012326656394453,
      "grad_norm": 2.5919454097747803,
      "learning_rate": 4.099733536353255e-06,
      "loss": 1.4807,
      "step": 780
    },
    {
      "epoch": 0.8012326656394453,
      "eval_loss": 1.875250220298767,
      "eval_runtime": 37.035,
      "eval_samples_per_second": 27.001,
      "eval_steps_per_second": 1.134,
      "step": 780
    },
    {
      "epoch": 0.8022598870056498,
      "grad_norm": 2.505749464035034,
      "learning_rate": 4.097830224590788e-06,
      "loss": 1.5224,
      "step": 781
    },
    {
      "epoch": 0.8032871083718541,
      "grad_norm": 2.747262954711914,
      "learning_rate": 4.095926912828322e-06,
      "loss": 1.6051,
      "step": 782
    },
    {
      "epoch": 0.8043143297380585,
      "grad_norm": 2.6147146224975586,
      "learning_rate": 4.094023601065855e-06,
      "loss": 1.565,
      "step": 783
    },
    {
      "epoch": 0.805341551104263,
      "grad_norm": 2.5381031036376953,
      "learning_rate": 4.092120289303388e-06,
      "loss": 1.511,
      "step": 784
    },
    {
      "epoch": 0.8063687724704673,
      "grad_norm": 2.6476190090179443,
      "learning_rate": 4.0902169775409215e-06,
      "loss": 1.5147,
      "step": 785
    },
    {
      "epoch": 0.8073959938366718,
      "grad_norm": 2.486919641494751,
      "learning_rate": 4.088313665778455e-06,
      "loss": 1.4697,
      "step": 786
    },
    {
      "epoch": 0.8084232152028762,
      "grad_norm": 2.7006654739379883,
      "learning_rate": 4.086410354015988e-06,
      "loss": 1.4965,
      "step": 787
    },
    {
      "epoch": 0.8094504365690807,
      "grad_norm": 2.7587921619415283,
      "learning_rate": 4.0845070422535216e-06,
      "loss": 1.4781,
      "step": 788
    },
    {
      "epoch": 0.810477657935285,
      "grad_norm": 2.64239501953125,
      "learning_rate": 4.082603730491055e-06,
      "loss": 1.4945,
      "step": 789
    },
    {
      "epoch": 0.8115048793014895,
      "grad_norm": 2.659919261932373,
      "learning_rate": 4.080700418728588e-06,
      "loss": 1.5378,
      "step": 790
    },
    {
      "epoch": 0.8115048793014895,
      "eval_loss": 1.881503939628601,
      "eval_runtime": 37.2277,
      "eval_samples_per_second": 26.862,
      "eval_steps_per_second": 1.128,
      "step": 790
    },
    {
      "epoch": 0.8125321006676939,
      "grad_norm": 2.820996046066284,
      "learning_rate": 4.078797106966122e-06,
      "loss": 1.5134,
      "step": 791
    },
    {
      "epoch": 0.8135593220338984,
      "grad_norm": 2.4667489528656006,
      "learning_rate": 4.076893795203655e-06,
      "loss": 1.4165,
      "step": 792
    },
    {
      "epoch": 0.8145865434001027,
      "grad_norm": 2.5528056621551514,
      "learning_rate": 4.074990483441188e-06,
      "loss": 1.481,
      "step": 793
    },
    {
      "epoch": 0.8156137647663071,
      "grad_norm": 2.632922887802124,
      "learning_rate": 4.073087171678721e-06,
      "loss": 1.5502,
      "step": 794
    },
    {
      "epoch": 0.8166409861325116,
      "grad_norm": 2.662536144256592,
      "learning_rate": 4.071183859916254e-06,
      "loss": 1.5302,
      "step": 795
    },
    {
      "epoch": 0.8176682074987159,
      "grad_norm": 2.662454605102539,
      "learning_rate": 4.069280548153788e-06,
      "loss": 1.5131,
      "step": 796
    },
    {
      "epoch": 0.8186954288649204,
      "grad_norm": 2.6387386322021484,
      "learning_rate": 4.067377236391321e-06,
      "loss": 1.4677,
      "step": 797
    },
    {
      "epoch": 0.8197226502311248,
      "grad_norm": 2.8879177570343018,
      "learning_rate": 4.065473924628854e-06,
      "loss": 1.5106,
      "step": 798
    },
    {
      "epoch": 0.8207498715973293,
      "grad_norm": 2.680610418319702,
      "learning_rate": 4.063570612866388e-06,
      "loss": 1.5389,
      "step": 799
    },
    {
      "epoch": 0.8217770929635336,
      "grad_norm": 2.659590244293213,
      "learning_rate": 4.061667301103921e-06,
      "loss": 1.5254,
      "step": 800
    },
    {
      "epoch": 0.8217770929635336,
      "eval_loss": 1.881851077079773,
      "eval_runtime": 37.3385,
      "eval_samples_per_second": 26.782,
      "eval_steps_per_second": 1.125,
      "step": 800
    },
    {
      "epoch": 0.8228043143297381,
      "grad_norm": 2.711195945739746,
      "learning_rate": 4.0597639893414544e-06,
      "loss": 1.5026,
      "step": 801
    },
    {
      "epoch": 0.8238315356959425,
      "grad_norm": 2.6387939453125,
      "learning_rate": 4.057860677578988e-06,
      "loss": 1.5101,
      "step": 802
    },
    {
      "epoch": 0.8248587570621468,
      "grad_norm": 2.507387638092041,
      "learning_rate": 4.055957365816521e-06,
      "loss": 1.5277,
      "step": 803
    },
    {
      "epoch": 0.8258859784283513,
      "grad_norm": 2.610370397567749,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 1.4695,
      "step": 804
    },
    {
      "epoch": 0.8269131997945557,
      "grad_norm": 2.8345787525177,
      "learning_rate": 4.052150742291588e-06,
      "loss": 1.467,
      "step": 805
    },
    {
      "epoch": 0.8279404211607602,
      "grad_norm": 2.5142805576324463,
      "learning_rate": 4.050247430529121e-06,
      "loss": 1.4637,
      "step": 806
    },
    {
      "epoch": 0.8289676425269645,
      "grad_norm": 2.7722535133361816,
      "learning_rate": 4.048344118766655e-06,
      "loss": 1.5427,
      "step": 807
    },
    {
      "epoch": 0.829994863893169,
      "grad_norm": 2.816789150238037,
      "learning_rate": 4.046440807004188e-06,
      "loss": 1.5422,
      "step": 808
    },
    {
      "epoch": 0.8310220852593734,
      "grad_norm": 2.6485376358032227,
      "learning_rate": 4.044537495241721e-06,
      "loss": 1.5078,
      "step": 809
    },
    {
      "epoch": 0.8320493066255779,
      "grad_norm": 2.4760921001434326,
      "learning_rate": 4.042634183479255e-06,
      "loss": 1.4827,
      "step": 810
    },
    {
      "epoch": 0.8320493066255779,
      "eval_loss": 1.8709155321121216,
      "eval_runtime": 36.9015,
      "eval_samples_per_second": 27.099,
      "eval_steps_per_second": 1.138,
      "step": 810
    },
    {
      "epoch": 0.8330765279917822,
      "grad_norm": 2.658062219619751,
      "learning_rate": 4.040730871716787e-06,
      "loss": 1.5434,
      "step": 811
    },
    {
      "epoch": 0.8341037493579867,
      "grad_norm": 2.5443618297576904,
      "learning_rate": 4.038827559954321e-06,
      "loss": 1.5165,
      "step": 812
    },
    {
      "epoch": 0.8351309707241911,
      "grad_norm": 2.698432207107544,
      "learning_rate": 4.036924248191854e-06,
      "loss": 1.5553,
      "step": 813
    },
    {
      "epoch": 0.8361581920903954,
      "grad_norm": 2.5598180294036865,
      "learning_rate": 4.035020936429387e-06,
      "loss": 1.5289,
      "step": 814
    },
    {
      "epoch": 0.8371854134565999,
      "grad_norm": 2.7002663612365723,
      "learning_rate": 4.033117624666921e-06,
      "loss": 1.509,
      "step": 815
    },
    {
      "epoch": 0.8382126348228043,
      "grad_norm": 2.7882776260375977,
      "learning_rate": 4.031214312904454e-06,
      "loss": 1.4787,
      "step": 816
    },
    {
      "epoch": 0.8392398561890088,
      "grad_norm": 2.5583086013793945,
      "learning_rate": 4.029311001141987e-06,
      "loss": 1.4642,
      "step": 817
    },
    {
      "epoch": 0.8402670775552131,
      "grad_norm": 2.584573268890381,
      "learning_rate": 4.027407689379521e-06,
      "loss": 1.4343,
      "step": 818
    },
    {
      "epoch": 0.8412942989214176,
      "grad_norm": 2.5062575340270996,
      "learning_rate": 4.025504377617054e-06,
      "loss": 1.4975,
      "step": 819
    },
    {
      "epoch": 0.842321520287622,
      "grad_norm": 2.3546457290649414,
      "learning_rate": 4.0236010658545875e-06,
      "loss": 1.4795,
      "step": 820
    },
    {
      "epoch": 0.842321520287622,
      "eval_loss": 1.8716342449188232,
      "eval_runtime": 37.0473,
      "eval_samples_per_second": 26.993,
      "eval_steps_per_second": 1.134,
      "step": 820
    },
    {
      "epoch": 0.8433487416538265,
      "grad_norm": 2.513228178024292,
      "learning_rate": 4.02169775409212e-06,
      "loss": 1.5191,
      "step": 821
    },
    {
      "epoch": 0.8443759630200308,
      "grad_norm": 2.6360220909118652,
      "learning_rate": 4.019794442329653e-06,
      "loss": 1.5739,
      "step": 822
    },
    {
      "epoch": 0.8454031843862352,
      "grad_norm": 2.5351850986480713,
      "learning_rate": 4.0178911305671876e-06,
      "loss": 1.4487,
      "step": 823
    },
    {
      "epoch": 0.8464304057524397,
      "grad_norm": 2.6693851947784424,
      "learning_rate": 4.015987818804721e-06,
      "loss": 1.5319,
      "step": 824
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 2.5210046768188477,
      "learning_rate": 4.014084507042254e-06,
      "loss": 1.4723,
      "step": 825
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 2.5799994468688965,
      "learning_rate": 4.012181195279788e-06,
      "loss": 1.4409,
      "step": 826
    },
    {
      "epoch": 0.8495120698510529,
      "grad_norm": 2.5206313133239746,
      "learning_rate": 4.010277883517321e-06,
      "loss": 1.5449,
      "step": 827
    },
    {
      "epoch": 0.8505392912172574,
      "grad_norm": 2.480584144592285,
      "learning_rate": 4.0083745717548535e-06,
      "loss": 1.5306,
      "step": 828
    },
    {
      "epoch": 0.8515665125834617,
      "grad_norm": 2.5229053497314453,
      "learning_rate": 4.006471259992387e-06,
      "loss": 1.5001,
      "step": 829
    },
    {
      "epoch": 0.8525937339496662,
      "grad_norm": 2.53357195854187,
      "learning_rate": 4.00456794822992e-06,
      "loss": 1.5188,
      "step": 830
    },
    {
      "epoch": 0.8525937339496662,
      "eval_loss": 1.8724256753921509,
      "eval_runtime": 37.1114,
      "eval_samples_per_second": 26.946,
      "eval_steps_per_second": 1.132,
      "step": 830
    },
    {
      "epoch": 0.8536209553158706,
      "grad_norm": 2.6796159744262695,
      "learning_rate": 4.002664636467454e-06,
      "loss": 1.5577,
      "step": 831
    },
    {
      "epoch": 0.8546481766820749,
      "grad_norm": 2.4717817306518555,
      "learning_rate": 4.000761324704987e-06,
      "loss": 1.5124,
      "step": 832
    },
    {
      "epoch": 0.8556753980482794,
      "grad_norm": 2.649057626724243,
      "learning_rate": 3.99885801294252e-06,
      "loss": 1.4516,
      "step": 833
    },
    {
      "epoch": 0.8567026194144838,
      "grad_norm": 2.592406749725342,
      "learning_rate": 3.996954701180054e-06,
      "loss": 1.5215,
      "step": 834
    },
    {
      "epoch": 0.8577298407806883,
      "grad_norm": 2.46956205368042,
      "learning_rate": 3.995051389417587e-06,
      "loss": 1.4324,
      "step": 835
    },
    {
      "epoch": 0.8587570621468926,
      "grad_norm": 2.8484537601470947,
      "learning_rate": 3.9931480776551204e-06,
      "loss": 1.4574,
      "step": 836
    },
    {
      "epoch": 0.8597842835130971,
      "grad_norm": 2.6475489139556885,
      "learning_rate": 3.991244765892654e-06,
      "loss": 1.5581,
      "step": 837
    },
    {
      "epoch": 0.8608115048793015,
      "grad_norm": 2.527153491973877,
      "learning_rate": 3.989341454130186e-06,
      "loss": 1.4713,
      "step": 838
    },
    {
      "epoch": 0.861838726245506,
      "grad_norm": 2.6767020225524902,
      "learning_rate": 3.98743814236772e-06,
      "loss": 1.4459,
      "step": 839
    },
    {
      "epoch": 0.8628659476117103,
      "grad_norm": 2.6406850814819336,
      "learning_rate": 3.985534830605253e-06,
      "loss": 1.4964,
      "step": 840
    },
    {
      "epoch": 0.8628659476117103,
      "eval_loss": 1.886762261390686,
      "eval_runtime": 37.1378,
      "eval_samples_per_second": 26.927,
      "eval_steps_per_second": 1.131,
      "step": 840
    },
    {
      "epoch": 0.8638931689779148,
      "grad_norm": 2.466372489929199,
      "learning_rate": 3.983631518842786e-06,
      "loss": 1.5136,
      "step": 841
    },
    {
      "epoch": 0.8649203903441192,
      "grad_norm": 2.830249071121216,
      "learning_rate": 3.98172820708032e-06,
      "loss": 1.5372,
      "step": 842
    },
    {
      "epoch": 0.8659476117103235,
      "grad_norm": 2.675889015197754,
      "learning_rate": 3.979824895317854e-06,
      "loss": 1.4881,
      "step": 843
    },
    {
      "epoch": 0.866974833076528,
      "grad_norm": 2.7418212890625,
      "learning_rate": 3.977921583555387e-06,
      "loss": 1.4541,
      "step": 844
    },
    {
      "epoch": 0.8680020544427324,
      "grad_norm": 2.572862386703491,
      "learning_rate": 3.97601827179292e-06,
      "loss": 1.5191,
      "step": 845
    },
    {
      "epoch": 0.8690292758089369,
      "grad_norm": 2.6524274349212646,
      "learning_rate": 3.974114960030453e-06,
      "loss": 1.4647,
      "step": 846
    },
    {
      "epoch": 0.8700564971751412,
      "grad_norm": 2.794415235519409,
      "learning_rate": 3.9722116482679866e-06,
      "loss": 1.504,
      "step": 847
    },
    {
      "epoch": 0.8710837185413457,
      "grad_norm": 2.5947723388671875,
      "learning_rate": 3.97030833650552e-06,
      "loss": 1.492,
      "step": 848
    },
    {
      "epoch": 0.8721109399075501,
      "grad_norm": 2.8539984226226807,
      "learning_rate": 3.968405024743053e-06,
      "loss": 1.5049,
      "step": 849
    },
    {
      "epoch": 0.8731381612737545,
      "grad_norm": 2.522786855697632,
      "learning_rate": 3.966501712980587e-06,
      "loss": 1.4223,
      "step": 850
    },
    {
      "epoch": 0.8731381612737545,
      "eval_loss": 1.8691684007644653,
      "eval_runtime": 37.1287,
      "eval_samples_per_second": 26.933,
      "eval_steps_per_second": 1.131,
      "step": 850
    },
    {
      "epoch": 0.8741653826399589,
      "grad_norm": 2.6483359336853027,
      "learning_rate": 3.96459840121812e-06,
      "loss": 1.4815,
      "step": 851
    },
    {
      "epoch": 0.8751926040061633,
      "grad_norm": 2.581429958343506,
      "learning_rate": 3.962695089455653e-06,
      "loss": 1.4837,
      "step": 852
    },
    {
      "epoch": 0.8762198253723678,
      "grad_norm": 2.5967583656311035,
      "learning_rate": 3.960791777693187e-06,
      "loss": 1.5289,
      "step": 853
    },
    {
      "epoch": 0.8772470467385721,
      "grad_norm": 2.7169225215911865,
      "learning_rate": 3.95888846593072e-06,
      "loss": 1.5173,
      "step": 854
    },
    {
      "epoch": 0.8782742681047766,
      "grad_norm": 2.644656181335449,
      "learning_rate": 3.956985154168253e-06,
      "loss": 1.5095,
      "step": 855
    },
    {
      "epoch": 0.879301489470981,
      "grad_norm": 2.6397511959075928,
      "learning_rate": 3.955081842405786e-06,
      "loss": 1.5422,
      "step": 856
    },
    {
      "epoch": 0.8803287108371854,
      "grad_norm": 2.691540479660034,
      "learning_rate": 3.953178530643319e-06,
      "loss": 1.5703,
      "step": 857
    },
    {
      "epoch": 0.8813559322033898,
      "grad_norm": 2.50526762008667,
      "learning_rate": 3.951275218880853e-06,
      "loss": 1.5885,
      "step": 858
    },
    {
      "epoch": 0.8823831535695943,
      "grad_norm": 2.3846609592437744,
      "learning_rate": 3.949371907118386e-06,
      "loss": 1.4248,
      "step": 859
    },
    {
      "epoch": 0.8834103749357987,
      "grad_norm": 2.7107760906219482,
      "learning_rate": 3.9474685953559194e-06,
      "loss": 1.5358,
      "step": 860
    },
    {
      "epoch": 0.8834103749357987,
      "eval_loss": 1.876391887664795,
      "eval_runtime": 37.1284,
      "eval_samples_per_second": 26.934,
      "eval_steps_per_second": 1.131,
      "step": 860
    },
    {
      "epoch": 0.884437596302003,
      "grad_norm": 2.6206252574920654,
      "learning_rate": 3.945565283593453e-06,
      "loss": 1.4907,
      "step": 861
    },
    {
      "epoch": 0.8854648176682075,
      "grad_norm": 2.6641664505004883,
      "learning_rate": 3.943661971830986e-06,
      "loss": 1.4888,
      "step": 862
    },
    {
      "epoch": 0.8864920390344119,
      "grad_norm": 2.7342002391815186,
      "learning_rate": 3.9417586600685195e-06,
      "loss": 1.5421,
      "step": 863
    },
    {
      "epoch": 0.8875192604006163,
      "grad_norm": 2.7120890617370605,
      "learning_rate": 3.939855348306053e-06,
      "loss": 1.5154,
      "step": 864
    },
    {
      "epoch": 0.8885464817668207,
      "grad_norm": 2.6350717544555664,
      "learning_rate": 3.937952036543586e-06,
      "loss": 1.5135,
      "step": 865
    },
    {
      "epoch": 0.8895737031330252,
      "grad_norm": 2.570857524871826,
      "learning_rate": 3.93604872478112e-06,
      "loss": 1.5072,
      "step": 866
    },
    {
      "epoch": 0.8906009244992296,
      "grad_norm": 2.855071544647217,
      "learning_rate": 3.934145413018653e-06,
      "loss": 1.5171,
      "step": 867
    },
    {
      "epoch": 0.891628145865434,
      "grad_norm": 2.5592195987701416,
      "learning_rate": 3.932242101256186e-06,
      "loss": 1.4695,
      "step": 868
    },
    {
      "epoch": 0.8926553672316384,
      "grad_norm": 2.6110830307006836,
      "learning_rate": 3.93033878949372e-06,
      "loss": 1.4802,
      "step": 869
    },
    {
      "epoch": 0.8936825885978429,
      "grad_norm": 2.7121548652648926,
      "learning_rate": 3.928435477731253e-06,
      "loss": 1.4422,
      "step": 870
    },
    {
      "epoch": 0.8936825885978429,
      "eval_loss": 1.8629941940307617,
      "eval_runtime": 37.1091,
      "eval_samples_per_second": 26.948,
      "eval_steps_per_second": 1.132,
      "step": 870
    },
    {
      "epoch": 0.8947098099640473,
      "grad_norm": 2.7763099670410156,
      "learning_rate": 3.9265321659687864e-06,
      "loss": 1.5275,
      "step": 871
    },
    {
      "epoch": 0.8957370313302516,
      "grad_norm": 2.596220016479492,
      "learning_rate": 3.924628854206319e-06,
      "loss": 1.5374,
      "step": 872
    },
    {
      "epoch": 0.8967642526964561,
      "grad_norm": 2.575990915298462,
      "learning_rate": 3.922725542443852e-06,
      "loss": 1.5071,
      "step": 873
    },
    {
      "epoch": 0.8977914740626605,
      "grad_norm": 2.7052223682403564,
      "learning_rate": 3.920822230681386e-06,
      "loss": 1.4656,
      "step": 874
    },
    {
      "epoch": 0.8988186954288649,
      "grad_norm": 2.550267219543457,
      "learning_rate": 3.918918918918919e-06,
      "loss": 1.4837,
      "step": 875
    },
    {
      "epoch": 0.8998459167950693,
      "grad_norm": 2.6948060989379883,
      "learning_rate": 3.917015607156452e-06,
      "loss": 1.5156,
      "step": 876
    },
    {
      "epoch": 0.9008731381612738,
      "grad_norm": 2.575596809387207,
      "learning_rate": 3.915112295393986e-06,
      "loss": 1.4881,
      "step": 877
    },
    {
      "epoch": 0.9019003595274782,
      "grad_norm": 2.9835500717163086,
      "learning_rate": 3.913208983631519e-06,
      "loss": 1.5729,
      "step": 878
    },
    {
      "epoch": 0.9029275808936826,
      "grad_norm": 2.5961520671844482,
      "learning_rate": 3.9113056718690525e-06,
      "loss": 1.5469,
      "step": 879
    },
    {
      "epoch": 0.903954802259887,
      "grad_norm": 2.518260955810547,
      "learning_rate": 3.909402360106586e-06,
      "loss": 1.4704,
      "step": 880
    },
    {
      "epoch": 0.903954802259887,
      "eval_loss": 1.8570330142974854,
      "eval_runtime": 37.1293,
      "eval_samples_per_second": 26.933,
      "eval_steps_per_second": 1.131,
      "step": 880
    },
    {
      "epoch": 0.9049820236260914,
      "grad_norm": 2.740879535675049,
      "learning_rate": 3.907499048344119e-06,
      "loss": 1.5092,
      "step": 881
    },
    {
      "epoch": 0.9060092449922958,
      "grad_norm": 2.5227973461151123,
      "learning_rate": 3.905595736581652e-06,
      "loss": 1.5252,
      "step": 882
    },
    {
      "epoch": 0.9070364663585002,
      "grad_norm": 2.483125925064087,
      "learning_rate": 3.903692424819186e-06,
      "loss": 1.5216,
      "step": 883
    },
    {
      "epoch": 0.9080636877247047,
      "grad_norm": 2.4979538917541504,
      "learning_rate": 3.901789113056719e-06,
      "loss": 1.4734,
      "step": 884
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.707669496536255,
      "learning_rate": 3.899885801294253e-06,
      "loss": 1.5347,
      "step": 885
    },
    {
      "epoch": 0.9101181304571135,
      "grad_norm": 2.6548409461975098,
      "learning_rate": 3.897982489531786e-06,
      "loss": 1.5079,
      "step": 886
    },
    {
      "epoch": 0.9111453518233179,
      "grad_norm": 2.6678848266601562,
      "learning_rate": 3.896079177769319e-06,
      "loss": 1.5659,
      "step": 887
    },
    {
      "epoch": 0.9121725731895224,
      "grad_norm": 2.570976734161377,
      "learning_rate": 3.894175866006853e-06,
      "loss": 1.5281,
      "step": 888
    },
    {
      "epoch": 0.9131997945557268,
      "grad_norm": 2.5939226150512695,
      "learning_rate": 3.892272554244385e-06,
      "loss": 1.56,
      "step": 889
    },
    {
      "epoch": 0.9142270159219312,
      "grad_norm": 2.4577109813690186,
      "learning_rate": 3.890369242481919e-06,
      "loss": 1.4702,
      "step": 890
    },
    {
      "epoch": 0.9142270159219312,
      "eval_loss": 1.8534926176071167,
      "eval_runtime": 37.3353,
      "eval_samples_per_second": 26.784,
      "eval_steps_per_second": 1.125,
      "step": 890
    },
    {
      "epoch": 0.9152542372881356,
      "grad_norm": 2.5012643337249756,
      "learning_rate": 3.888465930719452e-06,
      "loss": 1.5411,
      "step": 891
    },
    {
      "epoch": 0.91628145865434,
      "grad_norm": 2.470905065536499,
      "learning_rate": 3.886562618956985e-06,
      "loss": 1.4766,
      "step": 892
    },
    {
      "epoch": 0.9173086800205444,
      "grad_norm": 2.4610393047332764,
      "learning_rate": 3.884659307194519e-06,
      "loss": 1.4925,
      "step": 893
    },
    {
      "epoch": 0.9183359013867488,
      "grad_norm": 2.573151111602783,
      "learning_rate": 3.882755995432052e-06,
      "loss": 1.4389,
      "step": 894
    },
    {
      "epoch": 0.9193631227529533,
      "grad_norm": 2.628934144973755,
      "learning_rate": 3.8808526836695854e-06,
      "loss": 1.5099,
      "step": 895
    },
    {
      "epoch": 0.9203903441191577,
      "grad_norm": 2.639920949935913,
      "learning_rate": 3.878949371907119e-06,
      "loss": 1.4724,
      "step": 896
    },
    {
      "epoch": 0.9214175654853621,
      "grad_norm": 2.4775230884552,
      "learning_rate": 3.877046060144652e-06,
      "loss": 1.4657,
      "step": 897
    },
    {
      "epoch": 0.9224447868515665,
      "grad_norm": 2.5242490768432617,
      "learning_rate": 3.8751427483821855e-06,
      "loss": 1.459,
      "step": 898
    },
    {
      "epoch": 0.923472008217771,
      "grad_norm": 2.4444870948791504,
      "learning_rate": 3.873239436619718e-06,
      "loss": 1.494,
      "step": 899
    },
    {
      "epoch": 0.9244992295839753,
      "grad_norm": 2.4326930046081543,
      "learning_rate": 3.871336124857251e-06,
      "loss": 1.4449,
      "step": 900
    },
    {
      "epoch": 0.9244992295839753,
      "eval_loss": 1.8844366073608398,
      "eval_runtime": 37.432,
      "eval_samples_per_second": 26.715,
      "eval_steps_per_second": 1.122,
      "step": 900
    },
    {
      "epoch": 0.9255264509501797,
      "grad_norm": 2.621652603149414,
      "learning_rate": 3.869432813094785e-06,
      "loss": 1.5055,
      "step": 901
    },
    {
      "epoch": 0.9265536723163842,
      "grad_norm": 2.5909814834594727,
      "learning_rate": 3.867529501332318e-06,
      "loss": 1.5176,
      "step": 902
    },
    {
      "epoch": 0.9275808936825886,
      "grad_norm": 2.512760877609253,
      "learning_rate": 3.8656261895698515e-06,
      "loss": 1.5027,
      "step": 903
    },
    {
      "epoch": 0.928608115048793,
      "grad_norm": 2.6954784393310547,
      "learning_rate": 3.863722877807386e-06,
      "loss": 1.4912,
      "step": 904
    },
    {
      "epoch": 0.9296353364149974,
      "grad_norm": 2.481905221939087,
      "learning_rate": 3.861819566044919e-06,
      "loss": 1.529,
      "step": 905
    },
    {
      "epoch": 0.9306625577812019,
      "grad_norm": 2.612758159637451,
      "learning_rate": 3.859916254282452e-06,
      "loss": 1.5658,
      "step": 906
    },
    {
      "epoch": 0.9316897791474062,
      "grad_norm": 2.523171901702881,
      "learning_rate": 3.858012942519985e-06,
      "loss": 1.4804,
      "step": 907
    },
    {
      "epoch": 0.9327170005136107,
      "grad_norm": 2.5053422451019287,
      "learning_rate": 3.856109630757518e-06,
      "loss": 1.5007,
      "step": 908
    },
    {
      "epoch": 0.9337442218798151,
      "grad_norm": 2.6110904216766357,
      "learning_rate": 3.854206318995052e-06,
      "loss": 1.4758,
      "step": 909
    },
    {
      "epoch": 0.9347714432460195,
      "grad_norm": 2.7004992961883545,
      "learning_rate": 3.852303007232585e-06,
      "loss": 1.5255,
      "step": 910
    },
    {
      "epoch": 0.9347714432460195,
      "eval_loss": 1.892087697982788,
      "eval_runtime": 36.9356,
      "eval_samples_per_second": 27.074,
      "eval_steps_per_second": 1.137,
      "step": 910
    },
    {
      "epoch": 0.9357986646122239,
      "grad_norm": 2.617180109024048,
      "learning_rate": 3.850399695470118e-06,
      "loss": 1.4563,
      "step": 911
    },
    {
      "epoch": 0.9368258859784283,
      "grad_norm": 2.5294559001922607,
      "learning_rate": 3.848496383707652e-06,
      "loss": 1.4767,
      "step": 912
    },
    {
      "epoch": 0.9378531073446328,
      "grad_norm": 2.594238042831421,
      "learning_rate": 3.846593071945185e-06,
      "loss": 1.4617,
      "step": 913
    },
    {
      "epoch": 0.9388803287108372,
      "grad_norm": 2.6076996326446533,
      "learning_rate": 3.8446897601827185e-06,
      "loss": 1.446,
      "step": 914
    },
    {
      "epoch": 0.9399075500770416,
      "grad_norm": 2.6047134399414062,
      "learning_rate": 3.842786448420252e-06,
      "loss": 1.5698,
      "step": 915
    },
    {
      "epoch": 0.940934771443246,
      "grad_norm": 2.5670995712280273,
      "learning_rate": 3.840883136657785e-06,
      "loss": 1.5803,
      "step": 916
    },
    {
      "epoch": 0.9419619928094505,
      "grad_norm": 2.6153712272644043,
      "learning_rate": 3.838979824895318e-06,
      "loss": 1.5565,
      "step": 917
    },
    {
      "epoch": 0.9429892141756548,
      "grad_norm": 2.6707098484039307,
      "learning_rate": 3.837076513132851e-06,
      "loss": 1.4417,
      "step": 918
    },
    {
      "epoch": 0.9440164355418593,
      "grad_norm": 2.591336488723755,
      "learning_rate": 3.8351732013703844e-06,
      "loss": 1.5159,
      "step": 919
    },
    {
      "epoch": 0.9450436569080637,
      "grad_norm": 2.4689993858337402,
      "learning_rate": 3.833269889607918e-06,
      "loss": 1.4735,
      "step": 920
    },
    {
      "epoch": 0.9450436569080637,
      "eval_loss": 1.875109076499939,
      "eval_runtime": 37.0384,
      "eval_samples_per_second": 26.999,
      "eval_steps_per_second": 1.134,
      "step": 920
    },
    {
      "epoch": 0.9460708782742681,
      "grad_norm": 2.739758253097534,
      "learning_rate": 3.831366577845451e-06,
      "loss": 1.4566,
      "step": 921
    },
    {
      "epoch": 0.9470980996404725,
      "grad_norm": 2.65278697013855,
      "learning_rate": 3.8294632660829845e-06,
      "loss": 1.4545,
      "step": 922
    },
    {
      "epoch": 0.9481253210066769,
      "grad_norm": 2.678321123123169,
      "learning_rate": 3.827559954320518e-06,
      "loss": 1.4864,
      "step": 923
    },
    {
      "epoch": 0.9491525423728814,
      "grad_norm": 2.7516839504241943,
      "learning_rate": 3.825656642558051e-06,
      "loss": 1.5031,
      "step": 924
    },
    {
      "epoch": 0.9501797637390857,
      "grad_norm": 2.6125247478485107,
      "learning_rate": 3.823753330795585e-06,
      "loss": 1.521,
      "step": 925
    },
    {
      "epoch": 0.9512069851052902,
      "grad_norm": 2.5134871006011963,
      "learning_rate": 3.821850019033118e-06,
      "loss": 1.5019,
      "step": 926
    },
    {
      "epoch": 0.9522342064714946,
      "grad_norm": 2.5274763107299805,
      "learning_rate": 3.819946707270651e-06,
      "loss": 1.4358,
      "step": 927
    },
    {
      "epoch": 0.9532614278376991,
      "grad_norm": 2.480628728866577,
      "learning_rate": 3.818043395508185e-06,
      "loss": 1.483,
      "step": 928
    },
    {
      "epoch": 0.9542886492039034,
      "grad_norm": 2.6450936794281006,
      "learning_rate": 3.816140083745718e-06,
      "loss": 1.4892,
      "step": 929
    },
    {
      "epoch": 0.9553158705701078,
      "grad_norm": 2.6758296489715576,
      "learning_rate": 3.8142367719832514e-06,
      "loss": 1.5701,
      "step": 930
    },
    {
      "epoch": 0.9553158705701078,
      "eval_loss": 1.8651176691055298,
      "eval_runtime": 37.0553,
      "eval_samples_per_second": 26.987,
      "eval_steps_per_second": 1.133,
      "step": 930
    },
    {
      "epoch": 0.9563430919363123,
      "grad_norm": 2.6887295246124268,
      "learning_rate": 3.8123334602207844e-06,
      "loss": 1.4596,
      "step": 931
    },
    {
      "epoch": 0.9573703133025167,
      "grad_norm": 2.577507734298706,
      "learning_rate": 3.8104301484583177e-06,
      "loss": 1.4664,
      "step": 932
    },
    {
      "epoch": 0.9583975346687211,
      "grad_norm": 2.6431517601013184,
      "learning_rate": 3.808526836695851e-06,
      "loss": 1.4715,
      "step": 933
    },
    {
      "epoch": 0.9594247560349255,
      "grad_norm": 2.8451743125915527,
      "learning_rate": 3.8066235249333845e-06,
      "loss": 1.5096,
      "step": 934
    },
    {
      "epoch": 0.96045197740113,
      "grad_norm": 2.5525193214416504,
      "learning_rate": 3.804720213170918e-06,
      "loss": 1.4975,
      "step": 935
    },
    {
      "epoch": 0.9614791987673343,
      "grad_norm": 2.6353704929351807,
      "learning_rate": 3.8028169014084508e-06,
      "loss": 1.4979,
      "step": 936
    },
    {
      "epoch": 0.9625064201335388,
      "grad_norm": 2.923781156539917,
      "learning_rate": 3.800913589645984e-06,
      "loss": 1.5118,
      "step": 937
    },
    {
      "epoch": 0.9635336414997432,
      "grad_norm": 2.7363650798797607,
      "learning_rate": 3.7990102778835175e-06,
      "loss": 1.5016,
      "step": 938
    },
    {
      "epoch": 0.9645608628659477,
      "grad_norm": 2.7126598358154297,
      "learning_rate": 3.797106966121051e-06,
      "loss": 1.5022,
      "step": 939
    },
    {
      "epoch": 0.965588084232152,
      "grad_norm": 3.0102460384368896,
      "learning_rate": 3.7952036543585842e-06,
      "loss": 1.5638,
      "step": 940
    },
    {
      "epoch": 0.965588084232152,
      "eval_loss": 1.8475635051727295,
      "eval_runtime": 37.0871,
      "eval_samples_per_second": 26.964,
      "eval_steps_per_second": 1.132,
      "step": 940
    },
    {
      "epoch": 0.9666153055983564,
      "grad_norm": 2.5455925464630127,
      "learning_rate": 3.793300342596117e-06,
      "loss": 1.5218,
      "step": 941
    },
    {
      "epoch": 0.9676425269645609,
      "grad_norm": 2.5489683151245117,
      "learning_rate": 3.7913970308336505e-06,
      "loss": 1.5375,
      "step": 942
    },
    {
      "epoch": 0.9686697483307652,
      "grad_norm": 2.672626256942749,
      "learning_rate": 3.789493719071184e-06,
      "loss": 1.5217,
      "step": 943
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 2.498378038406372,
      "learning_rate": 3.7875904073087177e-06,
      "loss": 1.475,
      "step": 944
    },
    {
      "epoch": 0.9707241910631741,
      "grad_norm": 2.515190362930298,
      "learning_rate": 3.785687095546251e-06,
      "loss": 1.4341,
      "step": 945
    },
    {
      "epoch": 0.9717514124293786,
      "grad_norm": 2.7561733722686768,
      "learning_rate": 3.7837837837837844e-06,
      "loss": 1.4838,
      "step": 946
    },
    {
      "epoch": 0.9727786337955829,
      "grad_norm": 2.544776201248169,
      "learning_rate": 3.7818804720213178e-06,
      "loss": 1.4539,
      "step": 947
    },
    {
      "epoch": 0.9738058551617874,
      "grad_norm": 2.744417428970337,
      "learning_rate": 3.7799771602588507e-06,
      "loss": 1.4296,
      "step": 948
    },
    {
      "epoch": 0.9748330765279918,
      "grad_norm": 2.7759811878204346,
      "learning_rate": 3.778073848496384e-06,
      "loss": 1.4314,
      "step": 949
    },
    {
      "epoch": 0.9758602978941961,
      "grad_norm": 2.5204875469207764,
      "learning_rate": 3.7761705367339174e-06,
      "loss": 1.495,
      "step": 950
    },
    {
      "epoch": 0.9758602978941961,
      "eval_loss": 1.8491986989974976,
      "eval_runtime": 37.1266,
      "eval_samples_per_second": 26.935,
      "eval_steps_per_second": 1.131,
      "step": 950
    },
    {
      "epoch": 0.9768875192604006,
      "grad_norm": 2.5156631469726562,
      "learning_rate": 3.7742672249714508e-06,
      "loss": 1.5228,
      "step": 951
    },
    {
      "epoch": 0.977914740626605,
      "grad_norm": 2.469834089279175,
      "learning_rate": 3.772363913208984e-06,
      "loss": 1.5188,
      "step": 952
    },
    {
      "epoch": 0.9789419619928095,
      "grad_norm": 2.4933393001556396,
      "learning_rate": 3.770460601446517e-06,
      "loss": 1.4818,
      "step": 953
    },
    {
      "epoch": 0.9799691833590138,
      "grad_norm": 2.5382790565490723,
      "learning_rate": 3.7685572896840504e-06,
      "loss": 1.5107,
      "step": 954
    },
    {
      "epoch": 0.9809964047252183,
      "grad_norm": 2.6962392330169678,
      "learning_rate": 3.766653977921584e-06,
      "loss": 1.4534,
      "step": 955
    },
    {
      "epoch": 0.9820236260914227,
      "grad_norm": 2.5411603450775146,
      "learning_rate": 3.764750666159117e-06,
      "loss": 1.3942,
      "step": 956
    },
    {
      "epoch": 0.9830508474576272,
      "grad_norm": 2.7101845741271973,
      "learning_rate": 3.7628473543966505e-06,
      "loss": 1.5003,
      "step": 957
    },
    {
      "epoch": 0.9840780688238315,
      "grad_norm": 2.5780858993530273,
      "learning_rate": 3.7609440426341835e-06,
      "loss": 1.4508,
      "step": 958
    },
    {
      "epoch": 0.9851052901900359,
      "grad_norm": 2.814004421234131,
      "learning_rate": 3.759040730871717e-06,
      "loss": 1.5435,
      "step": 959
    },
    {
      "epoch": 0.9861325115562404,
      "grad_norm": 2.5701088905334473,
      "learning_rate": 3.75713741910925e-06,
      "loss": 1.4386,
      "step": 960
    },
    {
      "epoch": 0.9861325115562404,
      "eval_loss": 1.8556313514709473,
      "eval_runtime": 37.1375,
      "eval_samples_per_second": 26.927,
      "eval_steps_per_second": 1.131,
      "step": 960
    },
    {
      "epoch": 0.9871597329224447,
      "grad_norm": 2.407271146774292,
      "learning_rate": 3.7552341073467836e-06,
      "loss": 1.4338,
      "step": 961
    },
    {
      "epoch": 0.9881869542886492,
      "grad_norm": 2.433256149291992,
      "learning_rate": 3.753330795584317e-06,
      "loss": 1.47,
      "step": 962
    },
    {
      "epoch": 0.9892141756548536,
      "grad_norm": 2.679339647293091,
      "learning_rate": 3.75142748382185e-06,
      "loss": 1.5459,
      "step": 963
    },
    {
      "epoch": 0.9902413970210581,
      "grad_norm": 2.325840473175049,
      "learning_rate": 3.749524172059384e-06,
      "loss": 1.4385,
      "step": 964
    },
    {
      "epoch": 0.9912686183872624,
      "grad_norm": 2.7060739994049072,
      "learning_rate": 3.747620860296917e-06,
      "loss": 1.4938,
      "step": 965
    },
    {
      "epoch": 0.9922958397534669,
      "grad_norm": 2.589808464050293,
      "learning_rate": 3.7457175485344504e-06,
      "loss": 1.5258,
      "step": 966
    },
    {
      "epoch": 0.9933230611196713,
      "grad_norm": 2.4876766204833984,
      "learning_rate": 3.7438142367719837e-06,
      "loss": 1.5361,
      "step": 967
    },
    {
      "epoch": 0.9943502824858758,
      "grad_norm": 2.7370550632476807,
      "learning_rate": 3.741910925009517e-06,
      "loss": 1.4999,
      "step": 968
    },
    {
      "epoch": 0.9953775038520801,
      "grad_norm": 2.6552398204803467,
      "learning_rate": 3.7400076132470505e-06,
      "loss": 1.4612,
      "step": 969
    },
    {
      "epoch": 0.9964047252182845,
      "grad_norm": 2.5229146480560303,
      "learning_rate": 3.7381043014845834e-06,
      "loss": 1.4794,
      "step": 970
    },
    {
      "epoch": 0.9964047252182845,
      "eval_loss": 1.8614000082015991,
      "eval_runtime": 37.1311,
      "eval_samples_per_second": 26.932,
      "eval_steps_per_second": 1.131,
      "step": 970
    },
    {
      "epoch": 0.997431946584489,
      "grad_norm": 2.5604071617126465,
      "learning_rate": 3.7362009897221168e-06,
      "loss": 1.5078,
      "step": 971
    },
    {
      "epoch": 0.9984591679506933,
      "grad_norm": 2.614192485809326,
      "learning_rate": 3.73429767795965e-06,
      "loss": 1.4849,
      "step": 972
    },
    {
      "epoch": 0.9994863893168978,
      "grad_norm": 2.5245072841644287,
      "learning_rate": 3.7323943661971835e-06,
      "loss": 1.4993,
      "step": 973
    },
    {
      "epoch": 1.0005136106831023,
      "grad_norm": 2.5631163120269775,
      "learning_rate": 3.730491054434717e-06,
      "loss": 1.4136,
      "step": 974
    },
    {
      "epoch": 1.0015408320493067,
      "grad_norm": 2.625396728515625,
      "learning_rate": 3.72858774267225e-06,
      "loss": 1.3623,
      "step": 975
    },
    {
      "epoch": 1.002568053415511,
      "grad_norm": 2.7067251205444336,
      "learning_rate": 3.726684430909783e-06,
      "loss": 1.4527,
      "step": 976
    },
    {
      "epoch": 1.0035952747817154,
      "grad_norm": 2.5352942943573,
      "learning_rate": 3.7247811191473165e-06,
      "loss": 1.474,
      "step": 977
    },
    {
      "epoch": 1.00462249614792,
      "grad_norm": 2.818143367767334,
      "learning_rate": 3.72287780738485e-06,
      "loss": 1.4852,
      "step": 978
    },
    {
      "epoch": 1.0056497175141244,
      "grad_norm": 2.8269577026367188,
      "learning_rate": 3.7209744956223832e-06,
      "loss": 1.4667,
      "step": 979
    },
    {
      "epoch": 1.0066769388803287,
      "grad_norm": 2.837454319000244,
      "learning_rate": 3.719071183859916e-06,
      "loss": 1.4728,
      "step": 980
    },
    {
      "epoch": 1.0066769388803287,
      "eval_loss": 1.8701857328414917,
      "eval_runtime": 37.2859,
      "eval_samples_per_second": 26.82,
      "eval_steps_per_second": 1.126,
      "step": 980
    },
    {
      "epoch": 1.007704160246533,
      "grad_norm": 2.6601309776306152,
      "learning_rate": 3.7171678720974495e-06,
      "loss": 1.3917,
      "step": 981
    },
    {
      "epoch": 1.0087313816127375,
      "grad_norm": 2.761413097381592,
      "learning_rate": 3.715264560334983e-06,
      "loss": 1.5049,
      "step": 982
    },
    {
      "epoch": 1.009758602978942,
      "grad_norm": 2.681429147720337,
      "learning_rate": 3.7133612485725163e-06,
      "loss": 1.4579,
      "step": 983
    },
    {
      "epoch": 1.0107858243451464,
      "grad_norm": 2.693094253540039,
      "learning_rate": 3.71145793681005e-06,
      "loss": 1.4766,
      "step": 984
    },
    {
      "epoch": 1.0118130457113508,
      "grad_norm": 2.589165210723877,
      "learning_rate": 3.7095546250475834e-06,
      "loss": 1.4011,
      "step": 985
    },
    {
      "epoch": 1.0128402670775551,
      "grad_norm": 2.738140821456909,
      "learning_rate": 3.7076513132851168e-06,
      "loss": 1.5109,
      "step": 986
    },
    {
      "epoch": 1.0138674884437597,
      "grad_norm": 2.7863996028900146,
      "learning_rate": 3.7057480015226497e-06,
      "loss": 1.4344,
      "step": 987
    },
    {
      "epoch": 1.014894709809964,
      "grad_norm": 2.6011335849761963,
      "learning_rate": 3.703844689760183e-06,
      "loss": 1.3981,
      "step": 988
    },
    {
      "epoch": 1.0159219311761685,
      "grad_norm": 2.5715689659118652,
      "learning_rate": 3.7019413779977164e-06,
      "loss": 1.4876,
      "step": 989
    },
    {
      "epoch": 1.0169491525423728,
      "grad_norm": 2.6804113388061523,
      "learning_rate": 3.70003806623525e-06,
      "loss": 1.4576,
      "step": 990
    },
    {
      "epoch": 1.0169491525423728,
      "eval_loss": 1.8802876472473145,
      "eval_runtime": 37.3955,
      "eval_samples_per_second": 26.741,
      "eval_steps_per_second": 1.123,
      "step": 990
    },
    {
      "epoch": 1.0179763739085772,
      "grad_norm": 2.737604856491089,
      "learning_rate": 3.698134754472783e-06,
      "loss": 1.4399,
      "step": 991
    },
    {
      "epoch": 1.0190035952747818,
      "grad_norm": 2.658388137817383,
      "learning_rate": 3.696231442710316e-06,
      "loss": 1.5035,
      "step": 992
    },
    {
      "epoch": 1.0200308166409862,
      "grad_norm": 2.7263925075531006,
      "learning_rate": 3.6943281309478495e-06,
      "loss": 1.4469,
      "step": 993
    },
    {
      "epoch": 1.0210580380071905,
      "grad_norm": 2.7885773181915283,
      "learning_rate": 3.692424819185383e-06,
      "loss": 1.4641,
      "step": 994
    },
    {
      "epoch": 1.022085259373395,
      "grad_norm": 2.8929779529571533,
      "learning_rate": 3.690521507422916e-06,
      "loss": 1.4727,
      "step": 995
    },
    {
      "epoch": 1.0231124807395995,
      "grad_norm": 2.6605303287506104,
      "learning_rate": 3.6886181956604496e-06,
      "loss": 1.4153,
      "step": 996
    },
    {
      "epoch": 1.0241397021058039,
      "grad_norm": 2.8678219318389893,
      "learning_rate": 3.686714883897983e-06,
      "loss": 1.4199,
      "step": 997
    },
    {
      "epoch": 1.0251669234720082,
      "grad_norm": 2.884453058242798,
      "learning_rate": 3.684811572135516e-06,
      "loss": 1.4909,
      "step": 998
    },
    {
      "epoch": 1.0261941448382126,
      "grad_norm": 2.6681578159332275,
      "learning_rate": 3.6829082603730492e-06,
      "loss": 1.4903,
      "step": 999
    },
    {
      "epoch": 1.027221366204417,
      "grad_norm": 2.8013756275177,
      "learning_rate": 3.6810049486105826e-06,
      "loss": 1.449,
      "step": 1000
    },
    {
      "epoch": 1.027221366204417,
      "eval_loss": 1.8735578060150146,
      "eval_runtime": 37.4416,
      "eval_samples_per_second": 26.708,
      "eval_steps_per_second": 1.122,
      "step": 1000
    },
    {
      "epoch": 1.0282485875706215,
      "grad_norm": 2.787750244140625,
      "learning_rate": 3.679101636848116e-06,
      "loss": 1.4808,
      "step": 1001
    },
    {
      "epoch": 1.029275808936826,
      "grad_norm": 2.666874647140503,
      "learning_rate": 3.6771983250856493e-06,
      "loss": 1.4739,
      "step": 1002
    },
    {
      "epoch": 1.0303030303030303,
      "grad_norm": 2.720586061477661,
      "learning_rate": 3.6752950133231822e-06,
      "loss": 1.4439,
      "step": 1003
    },
    {
      "epoch": 1.0313302516692346,
      "grad_norm": 2.707771062850952,
      "learning_rate": 3.6733917015607156e-06,
      "loss": 1.4765,
      "step": 1004
    },
    {
      "epoch": 1.0323574730354392,
      "grad_norm": 2.658092737197876,
      "learning_rate": 3.6714883897982494e-06,
      "loss": 1.4822,
      "step": 1005
    },
    {
      "epoch": 1.0333846944016436,
      "grad_norm": 2.66788911819458,
      "learning_rate": 3.6695850780357828e-06,
      "loss": 1.4237,
      "step": 1006
    },
    {
      "epoch": 1.034411915767848,
      "grad_norm": 2.638956308364868,
      "learning_rate": 3.667681766273316e-06,
      "loss": 1.4168,
      "step": 1007
    },
    {
      "epoch": 1.0354391371340523,
      "grad_norm": 2.739384412765503,
      "learning_rate": 3.6657784545108495e-06,
      "loss": 1.4338,
      "step": 1008
    },
    {
      "epoch": 1.0364663585002567,
      "grad_norm": 2.599773406982422,
      "learning_rate": 3.6638751427483824e-06,
      "loss": 1.5188,
      "step": 1009
    },
    {
      "epoch": 1.0374935798664613,
      "grad_norm": 2.705876588821411,
      "learning_rate": 3.6619718309859158e-06,
      "loss": 1.468,
      "step": 1010
    },
    {
      "epoch": 1.0374935798664613,
      "eval_loss": 1.8751709461212158,
      "eval_runtime": 36.9681,
      "eval_samples_per_second": 27.05,
      "eval_steps_per_second": 1.136,
      "step": 1010
    },
    {
      "epoch": 1.0385208012326657,
      "grad_norm": 2.712103843688965,
      "learning_rate": 3.660068519223449e-06,
      "loss": 1.508,
      "step": 1011
    },
    {
      "epoch": 1.03954802259887,
      "grad_norm": 2.6621742248535156,
      "learning_rate": 3.6581652074609825e-06,
      "loss": 1.464,
      "step": 1012
    },
    {
      "epoch": 1.0405752439650744,
      "grad_norm": 2.6141257286071777,
      "learning_rate": 3.656261895698516e-06,
      "loss": 1.4429,
      "step": 1013
    },
    {
      "epoch": 1.041602465331279,
      "grad_norm": 2.5651750564575195,
      "learning_rate": 3.6543585839360492e-06,
      "loss": 1.4254,
      "step": 1014
    },
    {
      "epoch": 1.0426296866974833,
      "grad_norm": 2.598686933517456,
      "learning_rate": 3.652455272173582e-06,
      "loss": 1.436,
      "step": 1015
    },
    {
      "epoch": 1.0436569080636877,
      "grad_norm": 2.559375524520874,
      "learning_rate": 3.6505519604111155e-06,
      "loss": 1.4359,
      "step": 1016
    },
    {
      "epoch": 1.044684129429892,
      "grad_norm": 2.57920503616333,
      "learning_rate": 3.648648648648649e-06,
      "loss": 1.4524,
      "step": 1017
    },
    {
      "epoch": 1.0457113507960965,
      "grad_norm": 2.6629042625427246,
      "learning_rate": 3.6467453368861823e-06,
      "loss": 1.4901,
      "step": 1018
    },
    {
      "epoch": 1.046738572162301,
      "grad_norm": 2.6846625804901123,
      "learning_rate": 3.6448420251237156e-06,
      "loss": 1.4345,
      "step": 1019
    },
    {
      "epoch": 1.0477657935285054,
      "grad_norm": 2.7288694381713867,
      "learning_rate": 3.6429387133612486e-06,
      "loss": 1.4632,
      "step": 1020
    },
    {
      "epoch": 1.0477657935285054,
      "eval_loss": 1.878115177154541,
      "eval_runtime": 37.0627,
      "eval_samples_per_second": 26.981,
      "eval_steps_per_second": 1.133,
      "step": 1020
    },
    {
      "epoch": 1.0487930148947098,
      "grad_norm": 2.6689510345458984,
      "learning_rate": 3.641035401598782e-06,
      "loss": 1.4934,
      "step": 1021
    },
    {
      "epoch": 1.0498202362609141,
      "grad_norm": 2.6762588024139404,
      "learning_rate": 3.6391320898363153e-06,
      "loss": 1.4284,
      "step": 1022
    },
    {
      "epoch": 1.0508474576271187,
      "grad_norm": 2.7875330448150635,
      "learning_rate": 3.6372287780738487e-06,
      "loss": 1.4717,
      "step": 1023
    },
    {
      "epoch": 1.051874678993323,
      "grad_norm": 2.9239425659179688,
      "learning_rate": 3.635325466311382e-06,
      "loss": 1.4504,
      "step": 1024
    },
    {
      "epoch": 1.0529019003595275,
      "grad_norm": 2.700913429260254,
      "learning_rate": 3.633422154548916e-06,
      "loss": 1.4701,
      "step": 1025
    },
    {
      "epoch": 1.0539291217257318,
      "grad_norm": 2.693148612976074,
      "learning_rate": 3.6315188427864487e-06,
      "loss": 1.4889,
      "step": 1026
    },
    {
      "epoch": 1.0549563430919364,
      "grad_norm": 2.6215498447418213,
      "learning_rate": 3.629615531023982e-06,
      "loss": 1.4586,
      "step": 1027
    },
    {
      "epoch": 1.0559835644581408,
      "grad_norm": 2.704425573348999,
      "learning_rate": 3.6277122192615155e-06,
      "loss": 1.4182,
      "step": 1028
    },
    {
      "epoch": 1.0570107858243452,
      "grad_norm": 2.691352128982544,
      "learning_rate": 3.625808907499049e-06,
      "loss": 1.468,
      "step": 1029
    },
    {
      "epoch": 1.0580380071905495,
      "grad_norm": 2.769744634628296,
      "learning_rate": 3.623905595736582e-06,
      "loss": 1.4375,
      "step": 1030
    },
    {
      "epoch": 1.0580380071905495,
      "eval_loss": 1.8686742782592773,
      "eval_runtime": 37.1278,
      "eval_samples_per_second": 26.934,
      "eval_steps_per_second": 1.131,
      "step": 1030
    },
    {
      "epoch": 1.059065228556754,
      "grad_norm": 2.6012840270996094,
      "learning_rate": 3.6220022839741156e-06,
      "loss": 1.4334,
      "step": 1031
    },
    {
      "epoch": 1.0600924499229585,
      "grad_norm": 2.641416072845459,
      "learning_rate": 3.6200989722116485e-06,
      "loss": 1.4186,
      "step": 1032
    },
    {
      "epoch": 1.0611196712891628,
      "grad_norm": 2.821176052093506,
      "learning_rate": 3.618195660449182e-06,
      "loss": 1.4438,
      "step": 1033
    },
    {
      "epoch": 1.0621468926553672,
      "grad_norm": 2.719303607940674,
      "learning_rate": 3.6162923486867152e-06,
      "loss": 1.4519,
      "step": 1034
    },
    {
      "epoch": 1.0631741140215716,
      "grad_norm": 2.76882004737854,
      "learning_rate": 3.6143890369242486e-06,
      "loss": 1.3941,
      "step": 1035
    },
    {
      "epoch": 1.064201335387776,
      "grad_norm": 2.601137399673462,
      "learning_rate": 3.612485725161782e-06,
      "loss": 1.4167,
      "step": 1036
    },
    {
      "epoch": 1.0652285567539805,
      "grad_norm": 2.691140651702881,
      "learning_rate": 3.610582413399315e-06,
      "loss": 1.4361,
      "step": 1037
    },
    {
      "epoch": 1.066255778120185,
      "grad_norm": 2.5779640674591064,
      "learning_rate": 3.6086791016368482e-06,
      "loss": 1.4237,
      "step": 1038
    },
    {
      "epoch": 1.0672829994863893,
      "grad_norm": 2.6535801887512207,
      "learning_rate": 3.6067757898743816e-06,
      "loss": 1.436,
      "step": 1039
    },
    {
      "epoch": 1.0683102208525936,
      "grad_norm": 2.8365395069122314,
      "learning_rate": 3.604872478111915e-06,
      "loss": 1.5067,
      "step": 1040
    },
    {
      "epoch": 1.0683102208525936,
      "eval_loss": 1.8686083555221558,
      "eval_runtime": 37.1495,
      "eval_samples_per_second": 26.918,
      "eval_steps_per_second": 1.131,
      "step": 1040
    },
    {
      "epoch": 1.0693374422187982,
      "grad_norm": 2.704638957977295,
      "learning_rate": 3.6029691663494483e-06,
      "loss": 1.5105,
      "step": 1041
    },
    {
      "epoch": 1.0703646635850026,
      "grad_norm": 2.6670243740081787,
      "learning_rate": 3.6010658545869813e-06,
      "loss": 1.451,
      "step": 1042
    },
    {
      "epoch": 1.071391884951207,
      "grad_norm": 2.8808884620666504,
      "learning_rate": 3.5991625428245146e-06,
      "loss": 1.4795,
      "step": 1043
    },
    {
      "epoch": 1.0724191063174113,
      "grad_norm": 2.800075054168701,
      "learning_rate": 3.597259231062048e-06,
      "loss": 1.4758,
      "step": 1044
    },
    {
      "epoch": 1.073446327683616,
      "grad_norm": 2.4989686012268066,
      "learning_rate": 3.5953559192995818e-06,
      "loss": 1.42,
      "step": 1045
    },
    {
      "epoch": 1.0744735490498203,
      "grad_norm": 2.7234370708465576,
      "learning_rate": 3.593452607537115e-06,
      "loss": 1.4549,
      "step": 1046
    },
    {
      "epoch": 1.0755007704160247,
      "grad_norm": 2.7219443321228027,
      "learning_rate": 3.5915492957746485e-06,
      "loss": 1.5245,
      "step": 1047
    },
    {
      "epoch": 1.076527991782229,
      "grad_norm": 2.6963443756103516,
      "learning_rate": 3.589645984012182e-06,
      "loss": 1.4472,
      "step": 1048
    },
    {
      "epoch": 1.0775552131484334,
      "grad_norm": 2.7101387977600098,
      "learning_rate": 3.587742672249715e-06,
      "loss": 1.4659,
      "step": 1049
    },
    {
      "epoch": 1.078582434514638,
      "grad_norm": 2.7673099040985107,
      "learning_rate": 3.585839360487248e-06,
      "loss": 1.4284,
      "step": 1050
    },
    {
      "epoch": 1.078582434514638,
      "eval_loss": 1.879286289215088,
      "eval_runtime": 37.1517,
      "eval_samples_per_second": 26.917,
      "eval_steps_per_second": 1.13,
      "step": 1050
    },
    {
      "epoch": 1.0796096558808423,
      "grad_norm": 2.81614351272583,
      "learning_rate": 3.5839360487247815e-06,
      "loss": 1.4251,
      "step": 1051
    },
    {
      "epoch": 1.0806368772470467,
      "grad_norm": 2.770350933074951,
      "learning_rate": 3.582032736962315e-06,
      "loss": 1.4372,
      "step": 1052
    },
    {
      "epoch": 1.081664098613251,
      "grad_norm": 2.8099257946014404,
      "learning_rate": 3.5801294251998483e-06,
      "loss": 1.495,
      "step": 1053
    },
    {
      "epoch": 1.0826913199794557,
      "grad_norm": 2.582378387451172,
      "learning_rate": 3.578226113437381e-06,
      "loss": 1.4005,
      "step": 1054
    },
    {
      "epoch": 1.08371854134566,
      "grad_norm": 2.6215474605560303,
      "learning_rate": 3.5763228016749146e-06,
      "loss": 1.4428,
      "step": 1055
    },
    {
      "epoch": 1.0847457627118644,
      "grad_norm": 2.7204151153564453,
      "learning_rate": 3.574419489912448e-06,
      "loss": 1.5197,
      "step": 1056
    },
    {
      "epoch": 1.0857729840780688,
      "grad_norm": 2.6563117504119873,
      "learning_rate": 3.5725161781499813e-06,
      "loss": 1.457,
      "step": 1057
    },
    {
      "epoch": 1.0868002054442731,
      "grad_norm": 2.559816837310791,
      "learning_rate": 3.5706128663875147e-06,
      "loss": 1.4376,
      "step": 1058
    },
    {
      "epoch": 1.0878274268104777,
      "grad_norm": 2.6742372512817383,
      "learning_rate": 3.5687095546250476e-06,
      "loss": 1.4626,
      "step": 1059
    },
    {
      "epoch": 1.088854648176682,
      "grad_norm": 2.8052852153778076,
      "learning_rate": 3.566806242862581e-06,
      "loss": 1.4785,
      "step": 1060
    },
    {
      "epoch": 1.088854648176682,
      "eval_loss": 1.8775676488876343,
      "eval_runtime": 37.1523,
      "eval_samples_per_second": 26.916,
      "eval_steps_per_second": 1.13,
      "step": 1060
    },
    {
      "epoch": 1.0898818695428865,
      "grad_norm": 2.687411308288574,
      "learning_rate": 3.5649029311001143e-06,
      "loss": 1.4491,
      "step": 1061
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.662604808807373,
      "learning_rate": 3.5629996193376477e-06,
      "loss": 1.4274,
      "step": 1062
    },
    {
      "epoch": 1.0919363122752954,
      "grad_norm": 2.619046211242676,
      "learning_rate": 3.561096307575181e-06,
      "loss": 1.4449,
      "step": 1063
    },
    {
      "epoch": 1.0929635336414998,
      "grad_norm": 2.672553539276123,
      "learning_rate": 3.559192995812714e-06,
      "loss": 1.4719,
      "step": 1064
    },
    {
      "epoch": 1.0939907550077042,
      "grad_norm": 2.868760824203491,
      "learning_rate": 3.557289684050248e-06,
      "loss": 1.4955,
      "step": 1065
    },
    {
      "epoch": 1.0950179763739085,
      "grad_norm": 2.7635276317596436,
      "learning_rate": 3.555386372287781e-06,
      "loss": 1.4765,
      "step": 1066
    },
    {
      "epoch": 1.0960451977401129,
      "grad_norm": 2.883657455444336,
      "learning_rate": 3.5534830605253145e-06,
      "loss": 1.4923,
      "step": 1067
    },
    {
      "epoch": 1.0970724191063175,
      "grad_norm": 2.7595503330230713,
      "learning_rate": 3.551579748762848e-06,
      "loss": 1.4527,
      "step": 1068
    },
    {
      "epoch": 1.0980996404725218,
      "grad_norm": 2.8310179710388184,
      "learning_rate": 3.5496764370003812e-06,
      "loss": 1.4485,
      "step": 1069
    },
    {
      "epoch": 1.0991268618387262,
      "grad_norm": 2.699432134628296,
      "learning_rate": 3.5477731252379146e-06,
      "loss": 1.4322,
      "step": 1070
    },
    {
      "epoch": 1.0991268618387262,
      "eval_loss": 1.8706557750701904,
      "eval_runtime": 37.1509,
      "eval_samples_per_second": 26.917,
      "eval_steps_per_second": 1.131,
      "step": 1070
    },
    {
      "epoch": 1.1001540832049306,
      "grad_norm": 2.6249091625213623,
      "learning_rate": 3.5458698134754475e-06,
      "loss": 1.4031,
      "step": 1071
    },
    {
      "epoch": 1.1011813045711352,
      "grad_norm": 2.6882944107055664,
      "learning_rate": 3.543966501712981e-06,
      "loss": 1.4075,
      "step": 1072
    },
    {
      "epoch": 1.1022085259373395,
      "grad_norm": 2.6578996181488037,
      "learning_rate": 3.5420631899505142e-06,
      "loss": 1.4256,
      "step": 1073
    },
    {
      "epoch": 1.103235747303544,
      "grad_norm": 2.8124637603759766,
      "learning_rate": 3.5401598781880476e-06,
      "loss": 1.5263,
      "step": 1074
    },
    {
      "epoch": 1.1042629686697483,
      "grad_norm": 2.6022191047668457,
      "learning_rate": 3.538256566425581e-06,
      "loss": 1.3482,
      "step": 1075
    },
    {
      "epoch": 1.1052901900359529,
      "grad_norm": 2.753009080886841,
      "learning_rate": 3.536353254663114e-06,
      "loss": 1.483,
      "step": 1076
    },
    {
      "epoch": 1.1063174114021572,
      "grad_norm": 2.575666904449463,
      "learning_rate": 3.5344499429006473e-06,
      "loss": 1.4325,
      "step": 1077
    },
    {
      "epoch": 1.1073446327683616,
      "grad_norm": 2.594136953353882,
      "learning_rate": 3.5325466311381806e-06,
      "loss": 1.4187,
      "step": 1078
    },
    {
      "epoch": 1.108371854134566,
      "grad_norm": 2.807074785232544,
      "learning_rate": 3.530643319375714e-06,
      "loss": 1.4763,
      "step": 1079
    },
    {
      "epoch": 1.1093990755007703,
      "grad_norm": 2.5865330696105957,
      "learning_rate": 3.5287400076132474e-06,
      "loss": 1.3896,
      "step": 1080
    },
    {
      "epoch": 1.1093990755007703,
      "eval_loss": 1.8865792751312256,
      "eval_runtime": 37.1514,
      "eval_samples_per_second": 26.917,
      "eval_steps_per_second": 1.131,
      "step": 1080
    },
    {
      "epoch": 1.110426296866975,
      "grad_norm": 2.715419054031372,
      "learning_rate": 3.5268366958507803e-06,
      "loss": 1.4259,
      "step": 1081
    },
    {
      "epoch": 1.1114535182331793,
      "grad_norm": 2.6837334632873535,
      "learning_rate": 3.5249333840883137e-06,
      "loss": 1.4581,
      "step": 1082
    },
    {
      "epoch": 1.1124807395993837,
      "grad_norm": 2.722128391265869,
      "learning_rate": 3.523030072325847e-06,
      "loss": 1.4486,
      "step": 1083
    },
    {
      "epoch": 1.113507960965588,
      "grad_norm": 2.7390329837799072,
      "learning_rate": 3.5211267605633804e-06,
      "loss": 1.4225,
      "step": 1084
    },
    {
      "epoch": 1.1145351823317924,
      "grad_norm": 2.740271806716919,
      "learning_rate": 3.519223448800914e-06,
      "loss": 1.4274,
      "step": 1085
    },
    {
      "epoch": 1.115562403697997,
      "grad_norm": 2.8321313858032227,
      "learning_rate": 3.5173201370384475e-06,
      "loss": 1.4337,
      "step": 1086
    },
    {
      "epoch": 1.1165896250642013,
      "grad_norm": 2.740826368331909,
      "learning_rate": 3.515416825275981e-06,
      "loss": 1.4331,
      "step": 1087
    },
    {
      "epoch": 1.1176168464304057,
      "grad_norm": 2.7382850646972656,
      "learning_rate": 3.513513513513514e-06,
      "loss": 1.4495,
      "step": 1088
    },
    {
      "epoch": 1.11864406779661,
      "grad_norm": 2.7677972316741943,
      "learning_rate": 3.511610201751047e-06,
      "loss": 1.4628,
      "step": 1089
    },
    {
      "epoch": 1.1196712891628147,
      "grad_norm": 2.8850181102752686,
      "learning_rate": 3.5097068899885806e-06,
      "loss": 1.3739,
      "step": 1090
    },
    {
      "epoch": 1.1196712891628147,
      "eval_loss": 1.8794182538986206,
      "eval_runtime": 37.4388,
      "eval_samples_per_second": 26.71,
      "eval_steps_per_second": 1.122,
      "step": 1090
    },
    {
      "epoch": 1.120698510529019,
      "grad_norm": 2.73730731010437,
      "learning_rate": 3.507803578226114e-06,
      "loss": 1.4444,
      "step": 1091
    },
    {
      "epoch": 1.1217257318952234,
      "grad_norm": 3.027599811553955,
      "learning_rate": 3.5059002664636473e-06,
      "loss": 1.5123,
      "step": 1092
    },
    {
      "epoch": 1.1227529532614278,
      "grad_norm": 2.613658905029297,
      "learning_rate": 3.5039969547011802e-06,
      "loss": 1.3798,
      "step": 1093
    },
    {
      "epoch": 1.1237801746276324,
      "grad_norm": 2.6396987438201904,
      "learning_rate": 3.5020936429387136e-06,
      "loss": 1.4357,
      "step": 1094
    },
    {
      "epoch": 1.1248073959938367,
      "grad_norm": 2.625540018081665,
      "learning_rate": 3.500190331176247e-06,
      "loss": 1.4354,
      "step": 1095
    },
    {
      "epoch": 1.125834617360041,
      "grad_norm": 2.8413279056549072,
      "learning_rate": 3.4982870194137803e-06,
      "loss": 1.4846,
      "step": 1096
    },
    {
      "epoch": 1.1268618387262455,
      "grad_norm": 2.7898473739624023,
      "learning_rate": 3.4963837076513137e-06,
      "loss": 1.421,
      "step": 1097
    },
    {
      "epoch": 1.1278890600924498,
      "grad_norm": 2.6385879516601562,
      "learning_rate": 3.4944803958888466e-06,
      "loss": 1.4241,
      "step": 1098
    },
    {
      "epoch": 1.1289162814586544,
      "grad_norm": 2.677098512649536,
      "learning_rate": 3.49257708412638e-06,
      "loss": 1.3671,
      "step": 1099
    },
    {
      "epoch": 1.1299435028248588,
      "grad_norm": 2.8582913875579834,
      "learning_rate": 3.4906737723639133e-06,
      "loss": 1.4886,
      "step": 1100
    },
    {
      "epoch": 1.1299435028248588,
      "eval_loss": 1.8696506023406982,
      "eval_runtime": 37.462,
      "eval_samples_per_second": 26.694,
      "eval_steps_per_second": 1.121,
      "step": 1100
    },
    {
      "epoch": 1.1309707241910631,
      "grad_norm": 2.7157933712005615,
      "learning_rate": 3.4887704606014467e-06,
      "loss": 1.4964,
      "step": 1101
    },
    {
      "epoch": 1.1319979455572675,
      "grad_norm": 2.6703319549560547,
      "learning_rate": 3.48686714883898e-06,
      "loss": 1.4509,
      "step": 1102
    },
    {
      "epoch": 1.133025166923472,
      "grad_norm": 2.6307899951934814,
      "learning_rate": 3.484963837076513e-06,
      "loss": 1.485,
      "step": 1103
    },
    {
      "epoch": 1.1340523882896765,
      "grad_norm": 2.8333141803741455,
      "learning_rate": 3.4830605253140464e-06,
      "loss": 1.432,
      "step": 1104
    },
    {
      "epoch": 1.1350796096558808,
      "grad_norm": 2.545804023742676,
      "learning_rate": 3.4811572135515797e-06,
      "loss": 1.3799,
      "step": 1105
    },
    {
      "epoch": 1.1361068310220852,
      "grad_norm": 2.7070441246032715,
      "learning_rate": 3.4792539017891135e-06,
      "loss": 1.4193,
      "step": 1106
    },
    {
      "epoch": 1.1371340523882898,
      "grad_norm": 2.697140693664551,
      "learning_rate": 3.477350590026647e-06,
      "loss": 1.398,
      "step": 1107
    },
    {
      "epoch": 1.1381612737544942,
      "grad_norm": 2.784795045852661,
      "learning_rate": 3.4754472782641802e-06,
      "loss": 1.4708,
      "step": 1108
    },
    {
      "epoch": 1.1391884951206985,
      "grad_norm": 2.8518831729888916,
      "learning_rate": 3.4735439665017136e-06,
      "loss": 1.4707,
      "step": 1109
    },
    {
      "epoch": 1.140215716486903,
      "grad_norm": 3.0481607913970947,
      "learning_rate": 3.4716406547392465e-06,
      "loss": 1.4385,
      "step": 1110
    },
    {
      "epoch": 1.140215716486903,
      "eval_loss": 1.8367345333099365,
      "eval_runtime": 36.9539,
      "eval_samples_per_second": 27.061,
      "eval_steps_per_second": 1.137,
      "step": 1110
    },
    {
      "epoch": 1.1412429378531073,
      "grad_norm": 2.803053140640259,
      "learning_rate": 3.46973734297678e-06,
      "loss": 1.4361,
      "step": 1111
    },
    {
      "epoch": 1.1422701592193119,
      "grad_norm": 2.762638568878174,
      "learning_rate": 3.4678340312143133e-06,
      "loss": 1.4522,
      "step": 1112
    },
    {
      "epoch": 1.1432973805855162,
      "grad_norm": 2.9458558559417725,
      "learning_rate": 3.4659307194518466e-06,
      "loss": 1.4914,
      "step": 1113
    },
    {
      "epoch": 1.1443246019517206,
      "grad_norm": 2.774721622467041,
      "learning_rate": 3.46402740768938e-06,
      "loss": 1.4723,
      "step": 1114
    },
    {
      "epoch": 1.145351823317925,
      "grad_norm": 2.6556880474090576,
      "learning_rate": 3.462124095926913e-06,
      "loss": 1.4073,
      "step": 1115
    },
    {
      "epoch": 1.1463790446841293,
      "grad_norm": 2.7482454776763916,
      "learning_rate": 3.4602207841644463e-06,
      "loss": 1.4022,
      "step": 1116
    },
    {
      "epoch": 1.147406266050334,
      "grad_norm": 2.627567768096924,
      "learning_rate": 3.4583174724019797e-06,
      "loss": 1.4428,
      "step": 1117
    },
    {
      "epoch": 1.1484334874165383,
      "grad_norm": 2.750666856765747,
      "learning_rate": 3.456414160639513e-06,
      "loss": 1.4972,
      "step": 1118
    },
    {
      "epoch": 1.1494607087827426,
      "grad_norm": 2.842184543609619,
      "learning_rate": 3.4545108488770464e-06,
      "loss": 1.4753,
      "step": 1119
    },
    {
      "epoch": 1.150487930148947,
      "grad_norm": 2.829437732696533,
      "learning_rate": 3.4526075371145793e-06,
      "loss": 1.4525,
      "step": 1120
    },
    {
      "epoch": 1.150487930148947,
      "eval_loss": 1.8498426675796509,
      "eval_runtime": 37.0234,
      "eval_samples_per_second": 27.01,
      "eval_steps_per_second": 1.134,
      "step": 1120
    },
    {
      "epoch": 1.1515151515151516,
      "grad_norm": 2.7674920558929443,
      "learning_rate": 3.4507042253521127e-06,
      "loss": 1.4516,
      "step": 1121
    },
    {
      "epoch": 1.152542372881356,
      "grad_norm": 2.726806879043579,
      "learning_rate": 3.448800913589646e-06,
      "loss": 1.5113,
      "step": 1122
    },
    {
      "epoch": 1.1535695942475603,
      "grad_norm": 2.7698283195495605,
      "learning_rate": 3.4468976018271794e-06,
      "loss": 1.4295,
      "step": 1123
    },
    {
      "epoch": 1.1545968156137647,
      "grad_norm": 2.6224706172943115,
      "learning_rate": 3.4449942900647128e-06,
      "loss": 1.5179,
      "step": 1124
    },
    {
      "epoch": 1.1556240369799693,
      "grad_norm": 2.7266266345977783,
      "learning_rate": 3.443090978302246e-06,
      "loss": 1.4784,
      "step": 1125
    },
    {
      "epoch": 1.1566512583461737,
      "grad_norm": 2.631751298904419,
      "learning_rate": 3.44118766653978e-06,
      "loss": 1.4659,
      "step": 1126
    },
    {
      "epoch": 1.157678479712378,
      "grad_norm": 2.7077064514160156,
      "learning_rate": 3.439284354777313e-06,
      "loss": 1.4654,
      "step": 1127
    },
    {
      "epoch": 1.1587057010785824,
      "grad_norm": 2.735102415084839,
      "learning_rate": 3.4373810430148462e-06,
      "loss": 1.4604,
      "step": 1128
    },
    {
      "epoch": 1.1597329224447868,
      "grad_norm": 2.9314727783203125,
      "learning_rate": 3.4354777312523796e-06,
      "loss": 1.3964,
      "step": 1129
    },
    {
      "epoch": 1.1607601438109914,
      "grad_norm": 2.658930778503418,
      "learning_rate": 3.433574419489913e-06,
      "loss": 1.4331,
      "step": 1130
    },
    {
      "epoch": 1.1607601438109914,
      "eval_loss": 1.8545702695846558,
      "eval_runtime": 37.0493,
      "eval_samples_per_second": 26.991,
      "eval_steps_per_second": 1.134,
      "step": 1130
    },
    {
      "epoch": 1.1617873651771957,
      "grad_norm": 2.8146262168884277,
      "learning_rate": 3.4316711077274463e-06,
      "loss": 1.4711,
      "step": 1131
    },
    {
      "epoch": 1.1628145865434,
      "grad_norm": 2.6918272972106934,
      "learning_rate": 3.4297677959649792e-06,
      "loss": 1.4327,
      "step": 1132
    },
    {
      "epoch": 1.1638418079096045,
      "grad_norm": 2.680551528930664,
      "learning_rate": 3.4278644842025126e-06,
      "loss": 1.4886,
      "step": 1133
    },
    {
      "epoch": 1.1648690292758088,
      "grad_norm": 2.617286443710327,
      "learning_rate": 3.425961172440046e-06,
      "loss": 1.4167,
      "step": 1134
    },
    {
      "epoch": 1.1658962506420134,
      "grad_norm": 2.6772518157958984,
      "learning_rate": 3.4240578606775793e-06,
      "loss": 1.5274,
      "step": 1135
    },
    {
      "epoch": 1.1669234720082178,
      "grad_norm": 2.640723705291748,
      "learning_rate": 3.4221545489151127e-06,
      "loss": 1.4398,
      "step": 1136
    },
    {
      "epoch": 1.1679506933744221,
      "grad_norm": 2.7104315757751465,
      "learning_rate": 3.4202512371526456e-06,
      "loss": 1.4367,
      "step": 1137
    },
    {
      "epoch": 1.1689779147406265,
      "grad_norm": 2.831645965576172,
      "learning_rate": 3.418347925390179e-06,
      "loss": 1.5329,
      "step": 1138
    },
    {
      "epoch": 1.170005136106831,
      "grad_norm": 2.7174274921417236,
      "learning_rate": 3.4164446136277124e-06,
      "loss": 1.481,
      "step": 1139
    },
    {
      "epoch": 1.1710323574730355,
      "grad_norm": 2.6496286392211914,
      "learning_rate": 3.4145413018652457e-06,
      "loss": 1.4489,
      "step": 1140
    },
    {
      "epoch": 1.1710323574730355,
      "eval_loss": 1.8524384498596191,
      "eval_runtime": 37.1347,
      "eval_samples_per_second": 26.929,
      "eval_steps_per_second": 1.131,
      "step": 1140
    },
    {
      "epoch": 1.1720595788392398,
      "grad_norm": 2.610058069229126,
      "learning_rate": 3.412637990102779e-06,
      "loss": 1.4755,
      "step": 1141
    },
    {
      "epoch": 1.1730868002054442,
      "grad_norm": 2.642070770263672,
      "learning_rate": 3.4107346783403125e-06,
      "loss": 1.4981,
      "step": 1142
    },
    {
      "epoch": 1.1741140215716488,
      "grad_norm": 2.57493257522583,
      "learning_rate": 3.4088313665778454e-06,
      "loss": 1.4039,
      "step": 1143
    },
    {
      "epoch": 1.1751412429378532,
      "grad_norm": 2.5919628143310547,
      "learning_rate": 3.4069280548153788e-06,
      "loss": 1.4391,
      "step": 1144
    },
    {
      "epoch": 1.1761684643040575,
      "grad_norm": 2.698862314224243,
      "learning_rate": 3.405024743052912e-06,
      "loss": 1.4198,
      "step": 1145
    },
    {
      "epoch": 1.177195685670262,
      "grad_norm": 2.6387715339660645,
      "learning_rate": 3.403121431290446e-06,
      "loss": 1.4373,
      "step": 1146
    },
    {
      "epoch": 1.1782229070364663,
      "grad_norm": 2.764643430709839,
      "learning_rate": 3.4012181195279793e-06,
      "loss": 1.4348,
      "step": 1147
    },
    {
      "epoch": 1.1792501284026708,
      "grad_norm": 2.721038579940796,
      "learning_rate": 3.3993148077655126e-06,
      "loss": 1.4232,
      "step": 1148
    },
    {
      "epoch": 1.1802773497688752,
      "grad_norm": 2.8942360877990723,
      "learning_rate": 3.3974114960030456e-06,
      "loss": 1.4405,
      "step": 1149
    },
    {
      "epoch": 1.1813045711350796,
      "grad_norm": 2.687194585800171,
      "learning_rate": 3.395508184240579e-06,
      "loss": 1.3854,
      "step": 1150
    },
    {
      "epoch": 1.1813045711350796,
      "eval_loss": 1.8701989650726318,
      "eval_runtime": 37.132,
      "eval_samples_per_second": 26.931,
      "eval_steps_per_second": 1.131,
      "step": 1150
    },
    {
      "epoch": 1.182331792501284,
      "grad_norm": 2.7137982845306396,
      "learning_rate": 3.3936048724781123e-06,
      "loss": 1.4174,
      "step": 1151
    },
    {
      "epoch": 1.1833590138674885,
      "grad_norm": 2.8916432857513428,
      "learning_rate": 3.3917015607156457e-06,
      "loss": 1.5043,
      "step": 1152
    },
    {
      "epoch": 1.184386235233693,
      "grad_norm": 2.6625723838806152,
      "learning_rate": 3.389798248953179e-06,
      "loss": 1.3832,
      "step": 1153
    },
    {
      "epoch": 1.1854134565998973,
      "grad_norm": 2.8155555725097656,
      "learning_rate": 3.387894937190712e-06,
      "loss": 1.4098,
      "step": 1154
    },
    {
      "epoch": 1.1864406779661016,
      "grad_norm": 2.634751796722412,
      "learning_rate": 3.3859916254282453e-06,
      "loss": 1.4522,
      "step": 1155
    },
    {
      "epoch": 1.1874678993323062,
      "grad_norm": 2.7085063457489014,
      "learning_rate": 3.3840883136657787e-06,
      "loss": 1.4539,
      "step": 1156
    },
    {
      "epoch": 1.1884951206985106,
      "grad_norm": 2.65588116645813,
      "learning_rate": 3.382185001903312e-06,
      "loss": 1.5084,
      "step": 1157
    },
    {
      "epoch": 1.189522342064715,
      "grad_norm": 2.582489252090454,
      "learning_rate": 3.3802816901408454e-06,
      "loss": 1.4218,
      "step": 1158
    },
    {
      "epoch": 1.1905495634309193,
      "grad_norm": 2.6364378929138184,
      "learning_rate": 3.3783783783783788e-06,
      "loss": 1.4522,
      "step": 1159
    },
    {
      "epoch": 1.1915767847971237,
      "grad_norm": 2.6329915523529053,
      "learning_rate": 3.3764750666159117e-06,
      "loss": 1.442,
      "step": 1160
    },
    {
      "epoch": 1.1915767847971237,
      "eval_loss": 1.8593298196792603,
      "eval_runtime": 37.1143,
      "eval_samples_per_second": 26.944,
      "eval_steps_per_second": 1.132,
      "step": 1160
    },
    {
      "epoch": 1.1926040061633283,
      "grad_norm": 2.610050916671753,
      "learning_rate": 3.374571754853445e-06,
      "loss": 1.4397,
      "step": 1161
    },
    {
      "epoch": 1.1936312275295327,
      "grad_norm": 2.5503945350646973,
      "learning_rate": 3.3726684430909784e-06,
      "loss": 1.4211,
      "step": 1162
    },
    {
      "epoch": 1.194658448895737,
      "grad_norm": 2.6648080348968506,
      "learning_rate": 3.370765131328512e-06,
      "loss": 1.3938,
      "step": 1163
    },
    {
      "epoch": 1.1956856702619414,
      "grad_norm": 2.6705355644226074,
      "learning_rate": 3.368861819566045e-06,
      "loss": 1.4107,
      "step": 1164
    },
    {
      "epoch": 1.1967128916281458,
      "grad_norm": 2.7582640647888184,
      "learning_rate": 3.366958507803578e-06,
      "loss": 1.4327,
      "step": 1165
    },
    {
      "epoch": 1.1977401129943503,
      "grad_norm": 2.7768008708953857,
      "learning_rate": 3.365055196041112e-06,
      "loss": 1.454,
      "step": 1166
    },
    {
      "epoch": 1.1987673343605547,
      "grad_norm": 2.628997325897217,
      "learning_rate": 3.3631518842786452e-06,
      "loss": 1.3988,
      "step": 1167
    },
    {
      "epoch": 1.199794555726759,
      "grad_norm": 2.637505531311035,
      "learning_rate": 3.3612485725161786e-06,
      "loss": 1.4766,
      "step": 1168
    },
    {
      "epoch": 1.2008217770929634,
      "grad_norm": 2.6935489177703857,
      "learning_rate": 3.359345260753712e-06,
      "loss": 1.4063,
      "step": 1169
    },
    {
      "epoch": 1.201848998459168,
      "grad_norm": 2.7827870845794678,
      "learning_rate": 3.3574419489912453e-06,
      "loss": 1.4882,
      "step": 1170
    },
    {
      "epoch": 1.201848998459168,
      "eval_loss": 1.8575458526611328,
      "eval_runtime": 37.1207,
      "eval_samples_per_second": 26.939,
      "eval_steps_per_second": 1.131,
      "step": 1170
    },
    {
      "epoch": 1.2028762198253724,
      "grad_norm": 2.789053201675415,
      "learning_rate": 3.3555386372287783e-06,
      "loss": 1.4699,
      "step": 1171
    },
    {
      "epoch": 1.2039034411915768,
      "grad_norm": 3.053184747695923,
      "learning_rate": 3.3536353254663116e-06,
      "loss": 1.4993,
      "step": 1172
    },
    {
      "epoch": 1.2049306625577811,
      "grad_norm": 2.6963112354278564,
      "learning_rate": 3.351732013703845e-06,
      "loss": 1.3985,
      "step": 1173
    },
    {
      "epoch": 1.2059578839239857,
      "grad_norm": 2.689101219177246,
      "learning_rate": 3.3498287019413784e-06,
      "loss": 1.4464,
      "step": 1174
    },
    {
      "epoch": 1.20698510529019,
      "grad_norm": 2.7748899459838867,
      "learning_rate": 3.3479253901789117e-06,
      "loss": 1.4554,
      "step": 1175
    },
    {
      "epoch": 1.2080123266563945,
      "grad_norm": 2.8673830032348633,
      "learning_rate": 3.346022078416445e-06,
      "loss": 1.4796,
      "step": 1176
    },
    {
      "epoch": 1.2090395480225988,
      "grad_norm": 2.7719533443450928,
      "learning_rate": 3.344118766653978e-06,
      "loss": 1.3716,
      "step": 1177
    },
    {
      "epoch": 1.2100667693888032,
      "grad_norm": 2.71771240234375,
      "learning_rate": 3.3422154548915114e-06,
      "loss": 1.4554,
      "step": 1178
    },
    {
      "epoch": 1.2110939907550078,
      "grad_norm": 2.7095048427581787,
      "learning_rate": 3.3403121431290448e-06,
      "loss": 1.4233,
      "step": 1179
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 2.631638526916504,
      "learning_rate": 3.338408831366578e-06,
      "loss": 1.4581,
      "step": 1180
    },
    {
      "epoch": 1.2121212121212122,
      "eval_loss": 1.855039119720459,
      "eval_runtime": 37.1083,
      "eval_samples_per_second": 26.948,
      "eval_steps_per_second": 1.132,
      "step": 1180
    },
    {
      "epoch": 1.2131484334874165,
      "grad_norm": 2.657257318496704,
      "learning_rate": 3.3365055196041115e-06,
      "loss": 1.3864,
      "step": 1181
    },
    {
      "epoch": 1.214175654853621,
      "grad_norm": 2.894169569015503,
      "learning_rate": 3.3346022078416444e-06,
      "loss": 1.4673,
      "step": 1182
    },
    {
      "epoch": 1.2152028762198253,
      "grad_norm": 2.852644920349121,
      "learning_rate": 3.3326988960791778e-06,
      "loss": 1.5026,
      "step": 1183
    },
    {
      "epoch": 1.2162300975860298,
      "grad_norm": 2.720878839492798,
      "learning_rate": 3.330795584316711e-06,
      "loss": 1.4281,
      "step": 1184
    },
    {
      "epoch": 1.2172573189522342,
      "grad_norm": 2.6779472827911377,
      "learning_rate": 3.3288922725542445e-06,
      "loss": 1.3559,
      "step": 1185
    },
    {
      "epoch": 1.2182845403184386,
      "grad_norm": 2.836782217025757,
      "learning_rate": 3.326988960791778e-06,
      "loss": 1.4185,
      "step": 1186
    },
    {
      "epoch": 1.219311761684643,
      "grad_norm": 2.9181265830993652,
      "learning_rate": 3.3250856490293117e-06,
      "loss": 1.4379,
      "step": 1187
    },
    {
      "epoch": 1.2203389830508475,
      "grad_norm": 2.7315778732299805,
      "learning_rate": 3.323182337266845e-06,
      "loss": 1.4824,
      "step": 1188
    },
    {
      "epoch": 1.221366204417052,
      "grad_norm": 2.8483448028564453,
      "learning_rate": 3.321279025504378e-06,
      "loss": 1.4766,
      "step": 1189
    },
    {
      "epoch": 1.2223934257832563,
      "grad_norm": 2.736323118209839,
      "learning_rate": 3.3193757137419113e-06,
      "loss": 1.4647,
      "step": 1190
    },
    {
      "epoch": 1.2223934257832563,
      "eval_loss": 1.8581864833831787,
      "eval_runtime": 37.3401,
      "eval_samples_per_second": 26.781,
      "eval_steps_per_second": 1.125,
      "step": 1190
    },
    {
      "epoch": 1.2234206471494606,
      "grad_norm": 2.8863027095794678,
      "learning_rate": 3.3174724019794447e-06,
      "loss": 1.4533,
      "step": 1191
    },
    {
      "epoch": 1.2244478685156652,
      "grad_norm": 2.951228380203247,
      "learning_rate": 3.315569090216978e-06,
      "loss": 1.495,
      "step": 1192
    },
    {
      "epoch": 1.2254750898818696,
      "grad_norm": 2.7486696243286133,
      "learning_rate": 3.3136657784545114e-06,
      "loss": 1.4082,
      "step": 1193
    },
    {
      "epoch": 1.226502311248074,
      "grad_norm": 2.9105849266052246,
      "learning_rate": 3.3117624666920443e-06,
      "loss": 1.4988,
      "step": 1194
    },
    {
      "epoch": 1.2275295326142783,
      "grad_norm": 2.8165998458862305,
      "learning_rate": 3.3098591549295777e-06,
      "loss": 1.4162,
      "step": 1195
    },
    {
      "epoch": 1.2285567539804827,
      "grad_norm": 2.8999876976013184,
      "learning_rate": 3.307955843167111e-06,
      "loss": 1.3909,
      "step": 1196
    },
    {
      "epoch": 1.2295839753466873,
      "grad_norm": 2.8325014114379883,
      "learning_rate": 3.3060525314046444e-06,
      "loss": 1.4625,
      "step": 1197
    },
    {
      "epoch": 1.2306111967128917,
      "grad_norm": 2.6553618907928467,
      "learning_rate": 3.304149219642178e-06,
      "loss": 1.4453,
      "step": 1198
    },
    {
      "epoch": 1.231638418079096,
      "grad_norm": 2.584757089614868,
      "learning_rate": 3.3022459078797107e-06,
      "loss": 1.4412,
      "step": 1199
    },
    {
      "epoch": 1.2326656394453004,
      "grad_norm": 2.68215274810791,
      "learning_rate": 3.300342596117244e-06,
      "loss": 1.4511,
      "step": 1200
    },
    {
      "epoch": 1.2326656394453004,
      "eval_loss": 1.849502682685852,
      "eval_runtime": 37.394,
      "eval_samples_per_second": 26.742,
      "eval_steps_per_second": 1.123,
      "step": 1200
    },
    {
      "epoch": 1.2336928608115048,
      "grad_norm": 2.735494375228882,
      "learning_rate": 3.2984392843547775e-06,
      "loss": 1.4433,
      "step": 1201
    },
    {
      "epoch": 1.2347200821777093,
      "grad_norm": 2.6691815853118896,
      "learning_rate": 3.296535972592311e-06,
      "loss": 1.4033,
      "step": 1202
    },
    {
      "epoch": 1.2357473035439137,
      "grad_norm": 2.6378355026245117,
      "learning_rate": 3.294632660829844e-06,
      "loss": 1.378,
      "step": 1203
    },
    {
      "epoch": 1.236774524910118,
      "grad_norm": 2.8933019638061523,
      "learning_rate": 3.292729349067377e-06,
      "loss": 1.4227,
      "step": 1204
    },
    {
      "epoch": 1.2378017462763227,
      "grad_norm": 2.8933019638061523,
      "learning_rate": 3.292729349067377e-06,
      "loss": 1.4204,
      "step": 1205
    },
    {
      "epoch": 1.238828967642527,
      "grad_norm": 2.786475896835327,
      "learning_rate": 3.2908260373049105e-06,
      "loss": 1.4165,
      "step": 1206
    },
    {
      "epoch": 1.2398561890087314,
      "grad_norm": 2.936825752258301,
      "learning_rate": 3.288922725542444e-06,
      "loss": 1.4712,
      "step": 1207
    },
    {
      "epoch": 1.2408834103749358,
      "grad_norm": 2.7255940437316895,
      "learning_rate": 3.2870194137799776e-06,
      "loss": 1.4177,
      "step": 1208
    },
    {
      "epoch": 1.2419106317411401,
      "grad_norm": 2.6526482105255127,
      "learning_rate": 3.285116102017511e-06,
      "loss": 1.3975,
      "step": 1209
    },
    {
      "epoch": 1.2429378531073447,
      "grad_norm": 2.7539143562316895,
      "learning_rate": 3.2832127902550444e-06,
      "loss": 1.4717,
      "step": 1210
    },
    {
      "epoch": 1.2429378531073447,
      "eval_loss": 1.8445062637329102,
      "eval_runtime": 36.9889,
      "eval_samples_per_second": 27.035,
      "eval_steps_per_second": 1.135,
      "step": 1210
    },
    {
      "epoch": 1.243965074473549,
      "grad_norm": 2.907252073287964,
      "learning_rate": 3.2813094784925777e-06,
      "loss": 1.4484,
      "step": 1211
    },
    {
      "epoch": 1.2449922958397535,
      "grad_norm": 2.7312281131744385,
      "learning_rate": 3.2794061667301107e-06,
      "loss": 1.3863,
      "step": 1212
    },
    {
      "epoch": 1.2460195172059578,
      "grad_norm": 2.739184617996216,
      "learning_rate": 3.277502854967644e-06,
      "loss": 1.4302,
      "step": 1213
    },
    {
      "epoch": 1.2470467385721622,
      "grad_norm": 2.798630475997925,
      "learning_rate": 3.2755995432051774e-06,
      "loss": 1.4134,
      "step": 1214
    },
    {
      "epoch": 1.2480739599383668,
      "grad_norm": 2.835686683654785,
      "learning_rate": 3.2736962314427107e-06,
      "loss": 1.4258,
      "step": 1215
    },
    {
      "epoch": 1.2491011813045712,
      "grad_norm": 2.7404251098632812,
      "learning_rate": 3.271792919680244e-06,
      "loss": 1.4347,
      "step": 1216
    },
    {
      "epoch": 1.2501284026707755,
      "grad_norm": 2.8362114429473877,
      "learning_rate": 3.269889607917777e-06,
      "loss": 1.4909,
      "step": 1217
    },
    {
      "epoch": 1.25115562403698,
      "grad_norm": 2.842283248901367,
      "learning_rate": 3.2679862961553104e-06,
      "loss": 1.4065,
      "step": 1218
    },
    {
      "epoch": 1.2521828454031843,
      "grad_norm": 2.717536449432373,
      "learning_rate": 3.2660829843928438e-06,
      "loss": 1.4446,
      "step": 1219
    },
    {
      "epoch": 1.2532100667693888,
      "grad_norm": 2.687009572982788,
      "learning_rate": 3.264179672630377e-06,
      "loss": 1.4826,
      "step": 1220
    },
    {
      "epoch": 1.2532100667693888,
      "eval_loss": 1.8576964139938354,
      "eval_runtime": 37.0759,
      "eval_samples_per_second": 26.972,
      "eval_steps_per_second": 1.133,
      "step": 1220
    },
    {
      "epoch": 1.2542372881355932,
      "grad_norm": 2.7234137058258057,
      "learning_rate": 3.2622763608679105e-06,
      "loss": 1.4503,
      "step": 1221
    },
    {
      "epoch": 1.2552645095017976,
      "grad_norm": 2.733876943588257,
      "learning_rate": 3.2603730491054434e-06,
      "loss": 1.4076,
      "step": 1222
    },
    {
      "epoch": 1.2562917308680022,
      "grad_norm": 2.6480257511138916,
      "learning_rate": 3.258469737342977e-06,
      "loss": 1.422,
      "step": 1223
    },
    {
      "epoch": 1.2573189522342065,
      "grad_norm": 2.7006945610046387,
      "learning_rate": 3.25656642558051e-06,
      "loss": 1.3975,
      "step": 1224
    },
    {
      "epoch": 1.258346173600411,
      "grad_norm": 2.6176297664642334,
      "learning_rate": 3.2546631138180435e-06,
      "loss": 1.4138,
      "step": 1225
    },
    {
      "epoch": 1.2593733949666153,
      "grad_norm": 2.717742443084717,
      "learning_rate": 3.252759802055577e-06,
      "loss": 1.4092,
      "step": 1226
    },
    {
      "epoch": 1.2604006163328196,
      "grad_norm": 2.693643093109131,
      "learning_rate": 3.25085649029311e-06,
      "loss": 1.4221,
      "step": 1227
    },
    {
      "epoch": 1.2614278376990242,
      "grad_norm": 2.709026575088501,
      "learning_rate": 3.248953178530644e-06,
      "loss": 1.4474,
      "step": 1228
    },
    {
      "epoch": 1.2624550590652286,
      "grad_norm": 2.628399610519409,
      "learning_rate": 3.247049866768177e-06,
      "loss": 1.4075,
      "step": 1229
    },
    {
      "epoch": 1.263482280431433,
      "grad_norm": 2.681164503097534,
      "learning_rate": 3.2451465550057103e-06,
      "loss": 1.3683,
      "step": 1230
    },
    {
      "epoch": 1.263482280431433,
      "eval_loss": 1.8543845415115356,
      "eval_runtime": 37.1602,
      "eval_samples_per_second": 26.91,
      "eval_steps_per_second": 1.13,
      "step": 1230
    },
    {
      "epoch": 1.2645095017976373,
      "grad_norm": 2.6538004875183105,
      "learning_rate": 3.2432432432432437e-06,
      "loss": 1.4214,
      "step": 1231
    },
    {
      "epoch": 1.2655367231638417,
      "grad_norm": 2.714805841445923,
      "learning_rate": 3.241339931480777e-06,
      "loss": 1.4617,
      "step": 1232
    },
    {
      "epoch": 1.2665639445300463,
      "grad_norm": 2.743016481399536,
      "learning_rate": 3.2394366197183104e-06,
      "loss": 1.4556,
      "step": 1233
    },
    {
      "epoch": 1.2675911658962506,
      "grad_norm": 2.618269681930542,
      "learning_rate": 3.2375333079558434e-06,
      "loss": 1.419,
      "step": 1234
    },
    {
      "epoch": 1.268618387262455,
      "grad_norm": 2.672124147415161,
      "learning_rate": 3.2356299961933767e-06,
      "loss": 1.4477,
      "step": 1235
    },
    {
      "epoch": 1.2696456086286596,
      "grad_norm": 2.831037998199463,
      "learning_rate": 3.23372668443091e-06,
      "loss": 1.4552,
      "step": 1236
    },
    {
      "epoch": 1.270672829994864,
      "grad_norm": 2.713193893432617,
      "learning_rate": 3.2318233726684435e-06,
      "loss": 1.4653,
      "step": 1237
    },
    {
      "epoch": 1.2717000513610683,
      "grad_norm": 2.645097017288208,
      "learning_rate": 3.229920060905977e-06,
      "loss": 1.4087,
      "step": 1238
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 2.741934061050415,
      "learning_rate": 3.2280167491435098e-06,
      "loss": 1.4155,
      "step": 1239
    },
    {
      "epoch": 1.273754494093477,
      "grad_norm": 2.5912556648254395,
      "learning_rate": 3.226113437381043e-06,
      "loss": 1.4214,
      "step": 1240
    },
    {
      "epoch": 1.273754494093477,
      "eval_loss": 1.8524503707885742,
      "eval_runtime": 37.1424,
      "eval_samples_per_second": 26.923,
      "eval_steps_per_second": 1.131,
      "step": 1240
    },
    {
      "epoch": 1.2747817154596817,
      "grad_norm": 2.5774216651916504,
      "learning_rate": 3.2242101256185765e-06,
      "loss": 1.4309,
      "step": 1241
    },
    {
      "epoch": 1.275808936825886,
      "grad_norm": 2.7143211364746094,
      "learning_rate": 3.22230681385611e-06,
      "loss": 1.4928,
      "step": 1242
    },
    {
      "epoch": 1.2768361581920904,
      "grad_norm": 2.8101789951324463,
      "learning_rate": 3.220403502093643e-06,
      "loss": 1.4325,
      "step": 1243
    },
    {
      "epoch": 1.2778633795582948,
      "grad_norm": 2.704641580581665,
      "learning_rate": 3.218500190331176e-06,
      "loss": 1.4729,
      "step": 1244
    },
    {
      "epoch": 1.2788906009244991,
      "grad_norm": 2.649195671081543,
      "learning_rate": 3.2165968785687095e-06,
      "loss": 1.4186,
      "step": 1245
    },
    {
      "epoch": 1.2799178222907037,
      "grad_norm": 2.7863168716430664,
      "learning_rate": 3.214693566806243e-06,
      "loss": 1.4474,
      "step": 1246
    },
    {
      "epoch": 1.280945043656908,
      "grad_norm": 2.693589687347412,
      "learning_rate": 3.2127902550437762e-06,
      "loss": 1.4293,
      "step": 1247
    },
    {
      "epoch": 1.2819722650231125,
      "grad_norm": 2.7287509441375732,
      "learning_rate": 3.21088694328131e-06,
      "loss": 1.4305,
      "step": 1248
    },
    {
      "epoch": 1.2829994863893168,
      "grad_norm": 2.5521676540374756,
      "learning_rate": 3.2089836315188434e-06,
      "loss": 1.4203,
      "step": 1249
    },
    {
      "epoch": 1.2840267077555212,
      "grad_norm": 2.745425224304199,
      "learning_rate": 3.2070803197563767e-06,
      "loss": 1.4234,
      "step": 1250
    },
    {
      "epoch": 1.2840267077555212,
      "eval_loss": 1.8586679697036743,
      "eval_runtime": 37.1173,
      "eval_samples_per_second": 26.942,
      "eval_steps_per_second": 1.132,
      "step": 1250
    },
    {
      "epoch": 1.2850539291217258,
      "grad_norm": 2.687504768371582,
      "learning_rate": 3.2051770079939097e-06,
      "loss": 1.443,
      "step": 1251
    },
    {
      "epoch": 1.2860811504879301,
      "grad_norm": 2.7841391563415527,
      "learning_rate": 3.203273696231443e-06,
      "loss": 1.4722,
      "step": 1252
    },
    {
      "epoch": 1.2871083718541345,
      "grad_norm": 2.7715537548065186,
      "learning_rate": 3.2013703844689764e-06,
      "loss": 1.4063,
      "step": 1253
    },
    {
      "epoch": 1.288135593220339,
      "grad_norm": 2.8183677196502686,
      "learning_rate": 3.1994670727065098e-06,
      "loss": 1.4534,
      "step": 1254
    },
    {
      "epoch": 1.2891628145865435,
      "grad_norm": 2.795215606689453,
      "learning_rate": 3.197563760944043e-06,
      "loss": 1.4779,
      "step": 1255
    },
    {
      "epoch": 1.2901900359527478,
      "grad_norm": 2.7606441974639893,
      "learning_rate": 3.195660449181576e-06,
      "loss": 1.4668,
      "step": 1256
    },
    {
      "epoch": 1.2912172573189522,
      "grad_norm": 2.629718065261841,
      "learning_rate": 3.1937571374191094e-06,
      "loss": 1.4114,
      "step": 1257
    },
    {
      "epoch": 1.2922444786851566,
      "grad_norm": 2.776845932006836,
      "learning_rate": 3.191853825656643e-06,
      "loss": 1.4348,
      "step": 1258
    },
    {
      "epoch": 1.2932717000513612,
      "grad_norm": 2.672982931137085,
      "learning_rate": 3.189950513894176e-06,
      "loss": 1.4274,
      "step": 1259
    },
    {
      "epoch": 1.2942989214175655,
      "grad_norm": 2.5834884643554688,
      "learning_rate": 3.1880472021317095e-06,
      "loss": 1.4282,
      "step": 1260
    },
    {
      "epoch": 1.2942989214175655,
      "eval_loss": 1.8541877269744873,
      "eval_runtime": 37.1372,
      "eval_samples_per_second": 26.927,
      "eval_steps_per_second": 1.131,
      "step": 1260
    },
    {
      "epoch": 1.29532614278377,
      "grad_norm": 2.752054452896118,
      "learning_rate": 3.1861438903692425e-06,
      "loss": 1.3797,
      "step": 1261
    },
    {
      "epoch": 1.2963533641499743,
      "grad_norm": 2.723689556121826,
      "learning_rate": 3.184240578606776e-06,
      "loss": 1.4137,
      "step": 1262
    },
    {
      "epoch": 1.2973805855161786,
      "grad_norm": 2.806439161300659,
      "learning_rate": 3.182337266844309e-06,
      "loss": 1.4396,
      "step": 1263
    },
    {
      "epoch": 1.2984078068823832,
      "grad_norm": 2.8140552043914795,
      "learning_rate": 3.1804339550818426e-06,
      "loss": 1.4677,
      "step": 1264
    },
    {
      "epoch": 1.2994350282485876,
      "grad_norm": 2.78153920173645,
      "learning_rate": 3.178530643319376e-06,
      "loss": 1.433,
      "step": 1265
    },
    {
      "epoch": 1.300462249614792,
      "grad_norm": 2.6926140785217285,
      "learning_rate": 3.176627331556909e-06,
      "loss": 1.3901,
      "step": 1266
    },
    {
      "epoch": 1.3014894709809965,
      "grad_norm": 2.636629581451416,
      "learning_rate": 3.1747240197944422e-06,
      "loss": 1.4331,
      "step": 1267
    },
    {
      "epoch": 1.3025166923472007,
      "grad_norm": 2.625173330307007,
      "learning_rate": 3.172820708031976e-06,
      "loss": 1.3918,
      "step": 1268
    },
    {
      "epoch": 1.3035439137134053,
      "grad_norm": 2.745666980743408,
      "learning_rate": 3.1709173962695094e-06,
      "loss": 1.4669,
      "step": 1269
    },
    {
      "epoch": 1.3045711350796096,
      "grad_norm": 2.8781471252441406,
      "learning_rate": 3.1690140845070427e-06,
      "loss": 1.4586,
      "step": 1270
    },
    {
      "epoch": 1.3045711350796096,
      "eval_loss": 1.8759416341781616,
      "eval_runtime": 37.1372,
      "eval_samples_per_second": 26.927,
      "eval_steps_per_second": 1.131,
      "step": 1270
    },
    {
      "epoch": 1.305598356445814,
      "grad_norm": 2.832078456878662,
      "learning_rate": 3.167110772744576e-06,
      "loss": 1.4385,
      "step": 1271
    },
    {
      "epoch": 1.3066255778120186,
      "grad_norm": 2.7153384685516357,
      "learning_rate": 3.1652074609821095e-06,
      "loss": 1.427,
      "step": 1272
    },
    {
      "epoch": 1.307652799178223,
      "grad_norm": 2.6645848751068115,
      "learning_rate": 3.1633041492196424e-06,
      "loss": 1.4147,
      "step": 1273
    },
    {
      "epoch": 1.3086800205444273,
      "grad_norm": 2.7231268882751465,
      "learning_rate": 3.1614008374571758e-06,
      "loss": 1.436,
      "step": 1274
    },
    {
      "epoch": 1.3097072419106317,
      "grad_norm": 2.8131442070007324,
      "learning_rate": 3.159497525694709e-06,
      "loss": 1.4357,
      "step": 1275
    },
    {
      "epoch": 1.310734463276836,
      "grad_norm": 2.6366660594940186,
      "learning_rate": 3.1575942139322425e-06,
      "loss": 1.4089,
      "step": 1276
    },
    {
      "epoch": 1.3117616846430407,
      "grad_norm": 2.7849857807159424,
      "learning_rate": 3.155690902169776e-06,
      "loss": 1.4486,
      "step": 1277
    },
    {
      "epoch": 1.312788906009245,
      "grad_norm": 2.771298408508301,
      "learning_rate": 3.1537875904073088e-06,
      "loss": 1.4085,
      "step": 1278
    },
    {
      "epoch": 1.3138161273754494,
      "grad_norm": 2.7618181705474854,
      "learning_rate": 3.151884278644842e-06,
      "loss": 1.4272,
      "step": 1279
    },
    {
      "epoch": 1.3148433487416538,
      "grad_norm": 2.6741080284118652,
      "learning_rate": 3.1499809668823755e-06,
      "loss": 1.4135,
      "step": 1280
    },
    {
      "epoch": 1.3148433487416538,
      "eval_loss": 1.8567897081375122,
      "eval_runtime": 37.1429,
      "eval_samples_per_second": 26.923,
      "eval_steps_per_second": 1.131,
      "step": 1280
    },
    {
      "epoch": 1.3158705701078581,
      "grad_norm": 2.6671698093414307,
      "learning_rate": 3.148077655119909e-06,
      "loss": 1.4598,
      "step": 1281
    },
    {
      "epoch": 1.3168977914740627,
      "grad_norm": 2.6641244888305664,
      "learning_rate": 3.1461743433574422e-06,
      "loss": 1.3998,
      "step": 1282
    },
    {
      "epoch": 1.317925012840267,
      "grad_norm": 2.747417449951172,
      "learning_rate": 3.144271031594975e-06,
      "loss": 1.4602,
      "step": 1283
    },
    {
      "epoch": 1.3189522342064715,
      "grad_norm": 2.708984136581421,
      "learning_rate": 3.1423677198325085e-06,
      "loss": 1.4562,
      "step": 1284
    },
    {
      "epoch": 1.319979455572676,
      "grad_norm": 2.5973010063171387,
      "learning_rate": 3.140464408070042e-06,
      "loss": 1.4121,
      "step": 1285
    },
    {
      "epoch": 1.3210066769388804,
      "grad_norm": 2.6334621906280518,
      "learning_rate": 3.1385610963075753e-06,
      "loss": 1.4049,
      "step": 1286
    },
    {
      "epoch": 1.3220338983050848,
      "grad_norm": 2.772655963897705,
      "learning_rate": 3.1366577845451086e-06,
      "loss": 1.3953,
      "step": 1287
    },
    {
      "epoch": 1.3230611196712891,
      "grad_norm": 2.5006916522979736,
      "learning_rate": 3.134754472782642e-06,
      "loss": 1.3842,
      "step": 1288
    },
    {
      "epoch": 1.3240883410374935,
      "grad_norm": 2.6052703857421875,
      "learning_rate": 3.1328511610201758e-06,
      "loss": 1.3801,
      "step": 1289
    },
    {
      "epoch": 1.325115562403698,
      "grad_norm": 2.703693151473999,
      "learning_rate": 3.1309478492577087e-06,
      "loss": 1.3672,
      "step": 1290
    },
    {
      "epoch": 1.325115562403698,
      "eval_loss": 1.8470649719238281,
      "eval_runtime": 37.2714,
      "eval_samples_per_second": 26.83,
      "eval_steps_per_second": 1.127,
      "step": 1290
    },
    {
      "epoch": 1.3261427837699025,
      "grad_norm": 2.6864330768585205,
      "learning_rate": 3.129044537495242e-06,
      "loss": 1.4252,
      "step": 1291
    },
    {
      "epoch": 1.3271700051361068,
      "grad_norm": 2.7768399715423584,
      "learning_rate": 3.1271412257327754e-06,
      "loss": 1.4483,
      "step": 1292
    },
    {
      "epoch": 1.3281972265023112,
      "grad_norm": 2.6658599376678467,
      "learning_rate": 3.125237913970309e-06,
      "loss": 1.4059,
      "step": 1293
    },
    {
      "epoch": 1.3292244478685156,
      "grad_norm": 2.754718065261841,
      "learning_rate": 3.123334602207842e-06,
      "loss": 1.3787,
      "step": 1294
    },
    {
      "epoch": 1.3302516692347202,
      "grad_norm": 2.7994842529296875,
      "learning_rate": 3.121431290445375e-06,
      "loss": 1.4623,
      "step": 1295
    },
    {
      "epoch": 1.3312788906009245,
      "grad_norm": 2.805941343307495,
      "learning_rate": 3.1195279786829085e-06,
      "loss": 1.5186,
      "step": 1296
    },
    {
      "epoch": 1.332306111967129,
      "grad_norm": 2.6829605102539062,
      "learning_rate": 3.117624666920442e-06,
      "loss": 1.4052,
      "step": 1297
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 2.7221593856811523,
      "learning_rate": 3.115721355157975e-06,
      "loss": 1.4031,
      "step": 1298
    },
    {
      "epoch": 1.3343605546995376,
      "grad_norm": 2.579638719558716,
      "learning_rate": 3.1138180433955085e-06,
      "loss": 1.3627,
      "step": 1299
    },
    {
      "epoch": 1.3353877760657422,
      "grad_norm": 2.861189842224121,
      "learning_rate": 3.1119147316330415e-06,
      "loss": 1.4405,
      "step": 1300
    },
    {
      "epoch": 1.3353877760657422,
      "eval_loss": 1.8476718664169312,
      "eval_runtime": 37.4298,
      "eval_samples_per_second": 26.717,
      "eval_steps_per_second": 1.122,
      "step": 1300
    },
    {
      "epoch": 1.3364149974319466,
      "grad_norm": 2.8968658447265625,
      "learning_rate": 3.110011419870575e-06,
      "loss": 1.4774,
      "step": 1301
    },
    {
      "epoch": 1.337442218798151,
      "grad_norm": 2.8708558082580566,
      "learning_rate": 3.1081081081081082e-06,
      "loss": 1.4085,
      "step": 1302
    },
    {
      "epoch": 1.3384694401643555,
      "grad_norm": 2.802922487258911,
      "learning_rate": 3.1062047963456416e-06,
      "loss": 1.4486,
      "step": 1303
    },
    {
      "epoch": 1.33949666153056,
      "grad_norm": 2.9320907592773438,
      "learning_rate": 3.104301484583175e-06,
      "loss": 1.4932,
      "step": 1304
    },
    {
      "epoch": 1.3405238828967643,
      "grad_norm": 2.7708749771118164,
      "learning_rate": 3.1023981728207083e-06,
      "loss": 1.4601,
      "step": 1305
    },
    {
      "epoch": 1.3415511042629686,
      "grad_norm": 2.7464756965637207,
      "learning_rate": 3.1004948610582412e-06,
      "loss": 1.4245,
      "step": 1306
    },
    {
      "epoch": 1.342578325629173,
      "grad_norm": 2.767235517501831,
      "learning_rate": 3.0985915492957746e-06,
      "loss": 1.4199,
      "step": 1307
    },
    {
      "epoch": 1.3436055469953776,
      "grad_norm": 2.6810076236724854,
      "learning_rate": 3.096688237533308e-06,
      "loss": 1.3851,
      "step": 1308
    },
    {
      "epoch": 1.344632768361582,
      "grad_norm": 2.6390531063079834,
      "learning_rate": 3.0947849257708418e-06,
      "loss": 1.4688,
      "step": 1309
    },
    {
      "epoch": 1.3456599897277863,
      "grad_norm": 2.729029893875122,
      "learning_rate": 3.092881614008375e-06,
      "loss": 1.4248,
      "step": 1310
    },
    {
      "epoch": 1.3456599897277863,
      "eval_loss": 1.854402780532837,
      "eval_runtime": 36.9357,
      "eval_samples_per_second": 27.074,
      "eval_steps_per_second": 1.137,
      "step": 1310
    },
    {
      "epoch": 1.3466872110939907,
      "grad_norm": 2.8098297119140625,
      "learning_rate": 3.0909783022459085e-06,
      "loss": 1.4761,
      "step": 1311
    },
    {
      "epoch": 1.347714432460195,
      "grad_norm": 2.8076772689819336,
      "learning_rate": 3.0890749904834414e-06,
      "loss": 1.4119,
      "step": 1312
    },
    {
      "epoch": 1.3487416538263997,
      "grad_norm": 2.8214125633239746,
      "learning_rate": 3.0871716787209748e-06,
      "loss": 1.4693,
      "step": 1313
    },
    {
      "epoch": 1.349768875192604,
      "grad_norm": 2.830083131790161,
      "learning_rate": 3.085268366958508e-06,
      "loss": 1.438,
      "step": 1314
    },
    {
      "epoch": 1.3507960965588084,
      "grad_norm": 2.7671611309051514,
      "learning_rate": 3.0833650551960415e-06,
      "loss": 1.4417,
      "step": 1315
    },
    {
      "epoch": 1.3518233179250128,
      "grad_norm": 2.6464383602142334,
      "learning_rate": 3.081461743433575e-06,
      "loss": 1.4096,
      "step": 1316
    },
    {
      "epoch": 1.3528505392912171,
      "grad_norm": 2.651646137237549,
      "learning_rate": 3.079558431671108e-06,
      "loss": 1.4264,
      "step": 1317
    },
    {
      "epoch": 1.3538777606574217,
      "grad_norm": 2.6222681999206543,
      "learning_rate": 3.077655119908641e-06,
      "loss": 1.3836,
      "step": 1318
    },
    {
      "epoch": 1.354904982023626,
      "grad_norm": 2.5674147605895996,
      "learning_rate": 3.0757518081461745e-06,
      "loss": 1.4206,
      "step": 1319
    },
    {
      "epoch": 1.3559322033898304,
      "grad_norm": 2.7767961025238037,
      "learning_rate": 3.073848496383708e-06,
      "loss": 1.4348,
      "step": 1320
    },
    {
      "epoch": 1.3559322033898304,
      "eval_loss": 1.8607728481292725,
      "eval_runtime": 36.99,
      "eval_samples_per_second": 27.034,
      "eval_steps_per_second": 1.135,
      "step": 1320
    },
    {
      "epoch": 1.356959424756035,
      "grad_norm": 2.7811837196350098,
      "learning_rate": 3.0719451846212413e-06,
      "loss": 1.4082,
      "step": 1321
    },
    {
      "epoch": 1.3579866461222394,
      "grad_norm": 2.5633227825164795,
      "learning_rate": 3.0700418728587746e-06,
      "loss": 1.4352,
      "step": 1322
    },
    {
      "epoch": 1.3590138674884438,
      "grad_norm": 2.7205374240875244,
      "learning_rate": 3.0681385610963076e-06,
      "loss": 1.405,
      "step": 1323
    },
    {
      "epoch": 1.3600410888546481,
      "grad_norm": 2.82409930229187,
      "learning_rate": 3.066235249333841e-06,
      "loss": 1.448,
      "step": 1324
    },
    {
      "epoch": 1.3610683102208525,
      "grad_norm": 2.7932820320129395,
      "learning_rate": 3.0643319375713743e-06,
      "loss": 1.4785,
      "step": 1325
    },
    {
      "epoch": 1.362095531587057,
      "grad_norm": 2.68669056892395,
      "learning_rate": 3.0624286258089076e-06,
      "loss": 1.4232,
      "step": 1326
    },
    {
      "epoch": 1.3631227529532615,
      "grad_norm": 2.6492018699645996,
      "learning_rate": 3.060525314046441e-06,
      "loss": 1.4052,
      "step": 1327
    },
    {
      "epoch": 1.3641499743194658,
      "grad_norm": 2.7677114009857178,
      "learning_rate": 3.058622002283974e-06,
      "loss": 1.4426,
      "step": 1328
    },
    {
      "epoch": 1.3651771956856702,
      "grad_norm": 2.7124569416046143,
      "learning_rate": 3.0567186905215077e-06,
      "loss": 1.4631,
      "step": 1329
    },
    {
      "epoch": 1.3662044170518746,
      "grad_norm": 2.7084693908691406,
      "learning_rate": 3.054815378759041e-06,
      "loss": 1.4656,
      "step": 1330
    },
    {
      "epoch": 1.3662044170518746,
      "eval_loss": 1.8457012176513672,
      "eval_runtime": 37.0141,
      "eval_samples_per_second": 27.017,
      "eval_steps_per_second": 1.135,
      "step": 1330
    },
    {
      "epoch": 1.3672316384180792,
      "grad_norm": 2.7612688541412354,
      "learning_rate": 3.0529120669965745e-06,
      "loss": 1.455,
      "step": 1331
    },
    {
      "epoch": 1.3682588597842835,
      "grad_norm": 2.5172924995422363,
      "learning_rate": 3.051008755234108e-06,
      "loss": 1.3915,
      "step": 1332
    },
    {
      "epoch": 1.3692860811504879,
      "grad_norm": 2.7225120067596436,
      "learning_rate": 3.049105443471641e-06,
      "loss": 1.4025,
      "step": 1333
    },
    {
      "epoch": 1.3703133025166925,
      "grad_norm": 2.835470676422119,
      "learning_rate": 3.0472021317091745e-06,
      "loss": 1.5423,
      "step": 1334
    },
    {
      "epoch": 1.3713405238828968,
      "grad_norm": 2.6247284412384033,
      "learning_rate": 3.0452988199467075e-06,
      "loss": 1.4562,
      "step": 1335
    },
    {
      "epoch": 1.3723677452491012,
      "grad_norm": 2.6346960067749023,
      "learning_rate": 3.043395508184241e-06,
      "loss": 1.4566,
      "step": 1336
    },
    {
      "epoch": 1.3733949666153056,
      "grad_norm": 2.6278724670410156,
      "learning_rate": 3.041492196421774e-06,
      "loss": 1.403,
      "step": 1337
    },
    {
      "epoch": 1.37442218798151,
      "grad_norm": 2.8066110610961914,
      "learning_rate": 3.0395888846593076e-06,
      "loss": 1.4433,
      "step": 1338
    },
    {
      "epoch": 1.3754494093477145,
      "grad_norm": 2.6565256118774414,
      "learning_rate": 3.037685572896841e-06,
      "loss": 1.4177,
      "step": 1339
    },
    {
      "epoch": 1.376476630713919,
      "grad_norm": 2.6529884338378906,
      "learning_rate": 3.035782261134374e-06,
      "loss": 1.4378,
      "step": 1340
    },
    {
      "epoch": 1.376476630713919,
      "eval_loss": 1.8618167638778687,
      "eval_runtime": 37.025,
      "eval_samples_per_second": 27.009,
      "eval_steps_per_second": 1.134,
      "step": 1340
    },
    {
      "epoch": 1.3775038520801233,
      "grad_norm": 2.820838689804077,
      "learning_rate": 3.0338789493719072e-06,
      "loss": 1.4471,
      "step": 1341
    },
    {
      "epoch": 1.3785310734463276,
      "grad_norm": 2.6786739826202393,
      "learning_rate": 3.0319756376094406e-06,
      "loss": 1.4393,
      "step": 1342
    },
    {
      "epoch": 1.379558294812532,
      "grad_norm": 2.7962610721588135,
      "learning_rate": 3.030072325846974e-06,
      "loss": 1.4407,
      "step": 1343
    },
    {
      "epoch": 1.3805855161787366,
      "grad_norm": 2.739959716796875,
      "learning_rate": 3.0281690140845073e-06,
      "loss": 1.4059,
      "step": 1344
    },
    {
      "epoch": 1.381612737544941,
      "grad_norm": 2.7812039852142334,
      "learning_rate": 3.0262657023220403e-06,
      "loss": 1.4449,
      "step": 1345
    },
    {
      "epoch": 1.3826399589111453,
      "grad_norm": 2.749671459197998,
      "learning_rate": 3.0243623905595736e-06,
      "loss": 1.4459,
      "step": 1346
    },
    {
      "epoch": 1.3836671802773497,
      "grad_norm": 2.7494630813598633,
      "learning_rate": 3.022459078797107e-06,
      "loss": 1.4204,
      "step": 1347
    },
    {
      "epoch": 1.384694401643554,
      "grad_norm": 2.7188141345977783,
      "learning_rate": 3.0205557670346404e-06,
      "loss": 1.4249,
      "step": 1348
    },
    {
      "epoch": 1.3857216230097587,
      "grad_norm": 2.712442636489868,
      "learning_rate": 3.018652455272174e-06,
      "loss": 1.3975,
      "step": 1349
    },
    {
      "epoch": 1.386748844375963,
      "grad_norm": 2.9888060092926025,
      "learning_rate": 3.0167491435097075e-06,
      "loss": 1.5258,
      "step": 1350
    },
    {
      "epoch": 1.386748844375963,
      "eval_loss": 1.8591758012771606,
      "eval_runtime": 37.0078,
      "eval_samples_per_second": 27.021,
      "eval_steps_per_second": 1.135,
      "step": 1350
    },
    {
      "epoch": 1.3877760657421674,
      "grad_norm": 2.789698600769043,
      "learning_rate": 3.014845831747241e-06,
      "loss": 1.4305,
      "step": 1351
    },
    {
      "epoch": 1.388803287108372,
      "grad_norm": 2.747840166091919,
      "learning_rate": 3.012942519984774e-06,
      "loss": 1.4031,
      "step": 1352
    },
    {
      "epoch": 1.3898305084745763,
      "grad_norm": 2.7078518867492676,
      "learning_rate": 3.011039208222307e-06,
      "loss": 1.4139,
      "step": 1353
    },
    {
      "epoch": 1.3908577298407807,
      "grad_norm": 2.615677833557129,
      "learning_rate": 3.0091358964598405e-06,
      "loss": 1.3728,
      "step": 1354
    },
    {
      "epoch": 1.391884951206985,
      "grad_norm": 2.7485580444335938,
      "learning_rate": 3.007232584697374e-06,
      "loss": 1.4272,
      "step": 1355
    },
    {
      "epoch": 1.3929121725731894,
      "grad_norm": 2.7168128490448,
      "learning_rate": 3.0053292729349073e-06,
      "loss": 1.4353,
      "step": 1356
    },
    {
      "epoch": 1.393939393939394,
      "grad_norm": 2.7295777797698975,
      "learning_rate": 3.00342596117244e-06,
      "loss": 1.4324,
      "step": 1357
    },
    {
      "epoch": 1.3949666153055984,
      "grad_norm": 2.8791208267211914,
      "learning_rate": 3.0015226494099736e-06,
      "loss": 1.4303,
      "step": 1358
    },
    {
      "epoch": 1.3959938366718028,
      "grad_norm": 2.647028684616089,
      "learning_rate": 2.999619337647507e-06,
      "loss": 1.402,
      "step": 1359
    },
    {
      "epoch": 1.3970210580380071,
      "grad_norm": 2.869901180267334,
      "learning_rate": 2.9977160258850403e-06,
      "loss": 1.404,
      "step": 1360
    },
    {
      "epoch": 1.3970210580380071,
      "eval_loss": 1.8588491678237915,
      "eval_runtime": 37.0159,
      "eval_samples_per_second": 27.015,
      "eval_steps_per_second": 1.135,
      "step": 1360
    },
    {
      "epoch": 1.3980482794042115,
      "grad_norm": 2.7664403915405273,
      "learning_rate": 2.9958127141225736e-06,
      "loss": 1.3895,
      "step": 1361
    },
    {
      "epoch": 1.399075500770416,
      "grad_norm": 2.8101043701171875,
      "learning_rate": 2.9939094023601066e-06,
      "loss": 1.4629,
      "step": 1362
    },
    {
      "epoch": 1.4001027221366205,
      "grad_norm": 2.7114098072052,
      "learning_rate": 2.99200609059764e-06,
      "loss": 1.4365,
      "step": 1363
    },
    {
      "epoch": 1.4011299435028248,
      "grad_norm": 2.6739253997802734,
      "learning_rate": 2.9901027788351733e-06,
      "loss": 1.382,
      "step": 1364
    },
    {
      "epoch": 1.4021571648690292,
      "grad_norm": 2.8102970123291016,
      "learning_rate": 2.9881994670727067e-06,
      "loss": 1.392,
      "step": 1365
    },
    {
      "epoch": 1.4031843862352336,
      "grad_norm": 2.645071506500244,
      "learning_rate": 2.98629615531024e-06,
      "loss": 1.4157,
      "step": 1366
    },
    {
      "epoch": 1.4042116076014381,
      "grad_norm": 2.6110310554504395,
      "learning_rate": 2.984392843547773e-06,
      "loss": 1.3996,
      "step": 1367
    },
    {
      "epoch": 1.4052388289676425,
      "grad_norm": 2.784090042114258,
      "learning_rate": 2.9824895317853063e-06,
      "loss": 1.4178,
      "step": 1368
    },
    {
      "epoch": 1.4062660503338469,
      "grad_norm": 2.7293829917907715,
      "learning_rate": 2.98058622002284e-06,
      "loss": 1.4193,
      "step": 1369
    },
    {
      "epoch": 1.4072932717000515,
      "grad_norm": 2.6924493312835693,
      "learning_rate": 2.9786829082603735e-06,
      "loss": 1.476,
      "step": 1370
    },
    {
      "epoch": 1.4072932717000515,
      "eval_loss": 1.8642805814743042,
      "eval_runtime": 37.0452,
      "eval_samples_per_second": 26.994,
      "eval_steps_per_second": 1.134,
      "step": 1370
    },
    {
      "epoch": 1.4083204930662558,
      "grad_norm": 2.7714414596557617,
      "learning_rate": 2.976779596497907e-06,
      "loss": 1.4574,
      "step": 1371
    },
    {
      "epoch": 1.4093477144324602,
      "grad_norm": 2.7013490200042725,
      "learning_rate": 2.97487628473544e-06,
      "loss": 1.4718,
      "step": 1372
    },
    {
      "epoch": 1.4103749357986646,
      "grad_norm": 2.9120547771453857,
      "learning_rate": 2.9729729729729736e-06,
      "loss": 1.4298,
      "step": 1373
    },
    {
      "epoch": 1.411402157164869,
      "grad_norm": 2.9127988815307617,
      "learning_rate": 2.9710696612105065e-06,
      "loss": 1.4703,
      "step": 1374
    },
    {
      "epoch": 1.4124293785310735,
      "grad_norm": 2.696772575378418,
      "learning_rate": 2.96916634944804e-06,
      "loss": 1.412,
      "step": 1375
    },
    {
      "epoch": 1.413456599897278,
      "grad_norm": 2.8330435752868652,
      "learning_rate": 2.9672630376855732e-06,
      "loss": 1.4731,
      "step": 1376
    },
    {
      "epoch": 1.4144838212634823,
      "grad_norm": 2.7354564666748047,
      "learning_rate": 2.9653597259231066e-06,
      "loss": 1.4342,
      "step": 1377
    },
    {
      "epoch": 1.4155110426296866,
      "grad_norm": 2.7386012077331543,
      "learning_rate": 2.96345641416064e-06,
      "loss": 1.3935,
      "step": 1378
    },
    {
      "epoch": 1.416538263995891,
      "grad_norm": 2.5881052017211914,
      "learning_rate": 2.961553102398173e-06,
      "loss": 1.3282,
      "step": 1379
    },
    {
      "epoch": 1.4175654853620956,
      "grad_norm": 2.805558443069458,
      "learning_rate": 2.9596497906357063e-06,
      "loss": 1.4691,
      "step": 1380
    },
    {
      "epoch": 1.4175654853620956,
      "eval_loss": 1.84714937210083,
      "eval_runtime": 37.0349,
      "eval_samples_per_second": 27.002,
      "eval_steps_per_second": 1.134,
      "step": 1380
    },
    {
      "epoch": 1.4185927067283,
      "grad_norm": 2.8639137744903564,
      "learning_rate": 2.9577464788732396e-06,
      "loss": 1.4745,
      "step": 1381
    },
    {
      "epoch": 1.4196199280945043,
      "grad_norm": 2.6688785552978516,
      "learning_rate": 2.955843167110773e-06,
      "loss": 1.4016,
      "step": 1382
    },
    {
      "epoch": 1.420647149460709,
      "grad_norm": 2.6851978302001953,
      "learning_rate": 2.9539398553483064e-06,
      "loss": 1.4205,
      "step": 1383
    },
    {
      "epoch": 1.4216743708269133,
      "grad_norm": 2.7381017208099365,
      "learning_rate": 2.9520365435858393e-06,
      "loss": 1.4315,
      "step": 1384
    },
    {
      "epoch": 1.4227015921931176,
      "grad_norm": 2.9249424934387207,
      "learning_rate": 2.9501332318233727e-06,
      "loss": 1.49,
      "step": 1385
    },
    {
      "epoch": 1.423728813559322,
      "grad_norm": 2.765490770339966,
      "learning_rate": 2.948229920060906e-06,
      "loss": 1.4359,
      "step": 1386
    },
    {
      "epoch": 1.4247560349255264,
      "grad_norm": 2.578561544418335,
      "learning_rate": 2.9463266082984394e-06,
      "loss": 1.4032,
      "step": 1387
    },
    {
      "epoch": 1.425783256291731,
      "grad_norm": 2.6673238277435303,
      "learning_rate": 2.9444232965359727e-06,
      "loss": 1.4144,
      "step": 1388
    },
    {
      "epoch": 1.4268104776579353,
      "grad_norm": 2.737516164779663,
      "learning_rate": 2.9425199847735057e-06,
      "loss": 1.4375,
      "step": 1389
    },
    {
      "epoch": 1.4278376990241397,
      "grad_norm": 2.656177282333374,
      "learning_rate": 2.94061667301104e-06,
      "loss": 1.4007,
      "step": 1390
    },
    {
      "epoch": 1.4278376990241397,
      "eval_loss": 1.8427128791809082,
      "eval_runtime": 37.1322,
      "eval_samples_per_second": 26.931,
      "eval_steps_per_second": 1.131,
      "step": 1390
    },
    {
      "epoch": 1.428864920390344,
      "grad_norm": 2.6011996269226074,
      "learning_rate": 2.938713361248573e-06,
      "loss": 1.339,
      "step": 1391
    },
    {
      "epoch": 1.4298921417565484,
      "grad_norm": 2.6374619007110596,
      "learning_rate": 2.936810049486106e-06,
      "loss": 1.4092,
      "step": 1392
    },
    {
      "epoch": 1.430919363122753,
      "grad_norm": 2.6879520416259766,
      "learning_rate": 2.9349067377236396e-06,
      "loss": 1.4235,
      "step": 1393
    },
    {
      "epoch": 1.4319465844889574,
      "grad_norm": 2.8400580883026123,
      "learning_rate": 2.933003425961173e-06,
      "loss": 1.381,
      "step": 1394
    },
    {
      "epoch": 1.4329738058551618,
      "grad_norm": 2.742767810821533,
      "learning_rate": 2.9311001141987063e-06,
      "loss": 1.4055,
      "step": 1395
    },
    {
      "epoch": 1.4340010272213661,
      "grad_norm": 2.734239101409912,
      "learning_rate": 2.9291968024362392e-06,
      "loss": 1.3997,
      "step": 1396
    },
    {
      "epoch": 1.4350282485875705,
      "grad_norm": 2.6337532997131348,
      "learning_rate": 2.9272934906737726e-06,
      "loss": 1.3825,
      "step": 1397
    },
    {
      "epoch": 1.436055469953775,
      "grad_norm": 2.668790578842163,
      "learning_rate": 2.925390178911306e-06,
      "loss": 1.3918,
      "step": 1398
    },
    {
      "epoch": 1.4370826913199795,
      "grad_norm": 2.8307507038116455,
      "learning_rate": 2.9234868671488393e-06,
      "loss": 1.4183,
      "step": 1399
    },
    {
      "epoch": 1.4381099126861838,
      "grad_norm": 2.6661787033081055,
      "learning_rate": 2.9215835553863727e-06,
      "loss": 1.4225,
      "step": 1400
    },
    {
      "epoch": 1.4381099126861838,
      "eval_loss": 1.8518847227096558,
      "eval_runtime": 37.3672,
      "eval_samples_per_second": 26.761,
      "eval_steps_per_second": 1.124,
      "step": 1400
    },
    {
      "epoch": 1.4391371340523884,
      "grad_norm": 2.698025703430176,
      "learning_rate": 2.9196802436239056e-06,
      "loss": 1.4115,
      "step": 1401
    },
    {
      "epoch": 1.4401643554185928,
      "grad_norm": 2.6792502403259277,
      "learning_rate": 2.917776931861439e-06,
      "loss": 1.3869,
      "step": 1402
    },
    {
      "epoch": 1.4411915767847971,
      "grad_norm": 2.693514347076416,
      "learning_rate": 2.9158736200989723e-06,
      "loss": 1.3875,
      "step": 1403
    },
    {
      "epoch": 1.4422187981510015,
      "grad_norm": 2.6383583545684814,
      "learning_rate": 2.9139703083365057e-06,
      "loss": 1.3942,
      "step": 1404
    },
    {
      "epoch": 1.4432460195172059,
      "grad_norm": 2.6992499828338623,
      "learning_rate": 2.912066996574039e-06,
      "loss": 1.4243,
      "step": 1405
    },
    {
      "epoch": 1.4442732408834105,
      "grad_norm": 2.7901880741119385,
      "learning_rate": 2.910163684811572e-06,
      "loss": 1.3559,
      "step": 1406
    },
    {
      "epoch": 1.4453004622496148,
      "grad_norm": 2.9373207092285156,
      "learning_rate": 2.9082603730491054e-06,
      "loss": 1.4734,
      "step": 1407
    },
    {
      "epoch": 1.4463276836158192,
      "grad_norm": 2.8224740028381348,
      "learning_rate": 2.9063570612866387e-06,
      "loss": 1.4474,
      "step": 1408
    },
    {
      "epoch": 1.4473549049820236,
      "grad_norm": 2.9534835815429688,
      "learning_rate": 2.904453749524172e-06,
      "loss": 1.3691,
      "step": 1409
    },
    {
      "epoch": 1.448382126348228,
      "grad_norm": 2.8153076171875,
      "learning_rate": 2.902550437761706e-06,
      "loss": 1.3954,
      "step": 1410
    },
    {
      "epoch": 1.448382126348228,
      "eval_loss": 1.842891812324524,
      "eval_runtime": 36.9359,
      "eval_samples_per_second": 27.074,
      "eval_steps_per_second": 1.137,
      "step": 1410
    },
    {
      "epoch": 1.4494093477144325,
      "grad_norm": 2.8490853309631348,
      "learning_rate": 2.9006471259992392e-06,
      "loss": 1.4114,
      "step": 1411
    },
    {
      "epoch": 1.450436569080637,
      "grad_norm": 2.7475955486297607,
      "learning_rate": 2.8987438142367726e-06,
      "loss": 1.4068,
      "step": 1412
    },
    {
      "epoch": 1.4514637904468413,
      "grad_norm": 2.7209911346435547,
      "learning_rate": 2.8968405024743055e-06,
      "loss": 1.4658,
      "step": 1413
    },
    {
      "epoch": 1.4524910118130456,
      "grad_norm": 2.578934907913208,
      "learning_rate": 2.894937190711839e-06,
      "loss": 1.3686,
      "step": 1414
    },
    {
      "epoch": 1.45351823317925,
      "grad_norm": 2.722381114959717,
      "learning_rate": 2.8930338789493723e-06,
      "loss": 1.4337,
      "step": 1415
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 2.702178478240967,
      "learning_rate": 2.8911305671869056e-06,
      "loss": 1.4163,
      "step": 1416
    },
    {
      "epoch": 1.455572675911659,
      "grad_norm": 2.6742405891418457,
      "learning_rate": 2.889227255424439e-06,
      "loss": 1.3585,
      "step": 1417
    },
    {
      "epoch": 1.4565998972778633,
      "grad_norm": 2.6732680797576904,
      "learning_rate": 2.887323943661972e-06,
      "loss": 1.406,
      "step": 1418
    },
    {
      "epoch": 1.457627118644068,
      "grad_norm": 2.698864459991455,
      "learning_rate": 2.8854206318995053e-06,
      "loss": 1.3872,
      "step": 1419
    },
    {
      "epoch": 1.4586543400102723,
      "grad_norm": 2.7534351348876953,
      "learning_rate": 2.8835173201370386e-06,
      "loss": 1.4055,
      "step": 1420
    },
    {
      "epoch": 1.4586543400102723,
      "eval_loss": 1.8564038276672363,
      "eval_runtime": 37.0708,
      "eval_samples_per_second": 26.975,
      "eval_steps_per_second": 1.133,
      "step": 1420
    },
    {
      "epoch": 1.4596815613764766,
      "grad_norm": 2.9538612365722656,
      "learning_rate": 2.881614008374572e-06,
      "loss": 1.4596,
      "step": 1421
    },
    {
      "epoch": 1.460708782742681,
      "grad_norm": 2.829641103744507,
      "learning_rate": 2.8797106966121054e-06,
      "loss": 1.4265,
      "step": 1422
    },
    {
      "epoch": 1.4617360041088854,
      "grad_norm": 2.6814422607421875,
      "learning_rate": 2.8778073848496383e-06,
      "loss": 1.4112,
      "step": 1423
    },
    {
      "epoch": 1.46276322547509,
      "grad_norm": 2.7880561351776123,
      "learning_rate": 2.8759040730871717e-06,
      "loss": 1.4435,
      "step": 1424
    },
    {
      "epoch": 1.4637904468412943,
      "grad_norm": 2.676196336746216,
      "learning_rate": 2.874000761324705e-06,
      "loss": 1.4104,
      "step": 1425
    },
    {
      "epoch": 1.4648176682074987,
      "grad_norm": 2.7417874336242676,
      "learning_rate": 2.8720974495622384e-06,
      "loss": 1.4335,
      "step": 1426
    },
    {
      "epoch": 1.465844889573703,
      "grad_norm": 2.7870116233825684,
      "learning_rate": 2.8701941377997718e-06,
      "loss": 1.4226,
      "step": 1427
    },
    {
      "epoch": 1.4668721109399074,
      "grad_norm": 2.765194892883301,
      "learning_rate": 2.8682908260373047e-06,
      "loss": 1.4582,
      "step": 1428
    },
    {
      "epoch": 1.467899332306112,
      "grad_norm": 2.608175277709961,
      "learning_rate": 2.866387514274838e-06,
      "loss": 1.4285,
      "step": 1429
    },
    {
      "epoch": 1.4689265536723164,
      "grad_norm": 2.7573461532592773,
      "learning_rate": 2.864484202512372e-06,
      "loss": 1.4395,
      "step": 1430
    },
    {
      "epoch": 1.4689265536723164,
      "eval_loss": 1.8600319623947144,
      "eval_runtime": 37.1192,
      "eval_samples_per_second": 26.94,
      "eval_steps_per_second": 1.131,
      "step": 1430
    },
    {
      "epoch": 1.4699537750385208,
      "grad_norm": 2.635175943374634,
      "learning_rate": 2.8625808907499052e-06,
      "loss": 1.402,
      "step": 1431
    },
    {
      "epoch": 1.4709809964047253,
      "grad_norm": 2.7318811416625977,
      "learning_rate": 2.8606775789874386e-06,
      "loss": 1.4352,
      "step": 1432
    },
    {
      "epoch": 1.4720082177709297,
      "grad_norm": 2.783669948577881,
      "learning_rate": 2.858774267224972e-06,
      "loss": 1.4052,
      "step": 1433
    },
    {
      "epoch": 1.473035439137134,
      "grad_norm": 2.7327511310577393,
      "learning_rate": 2.8568709554625053e-06,
      "loss": 1.4209,
      "step": 1434
    },
    {
      "epoch": 1.4740626605033385,
      "grad_norm": 2.76542329788208,
      "learning_rate": 2.8549676437000382e-06,
      "loss": 1.4753,
      "step": 1435
    },
    {
      "epoch": 1.4750898818695428,
      "grad_norm": 2.8031232357025146,
      "learning_rate": 2.8530643319375716e-06,
      "loss": 1.4399,
      "step": 1436
    },
    {
      "epoch": 1.4761171032357474,
      "grad_norm": 2.737053155899048,
      "learning_rate": 2.851161020175105e-06,
      "loss": 1.4085,
      "step": 1437
    },
    {
      "epoch": 1.4771443246019518,
      "grad_norm": 2.609987735748291,
      "learning_rate": 2.8492577084126383e-06,
      "loss": 1.3809,
      "step": 1438
    },
    {
      "epoch": 1.4781715459681561,
      "grad_norm": 2.7482123374938965,
      "learning_rate": 2.8473543966501717e-06,
      "loss": 1.4234,
      "step": 1439
    },
    {
      "epoch": 1.4791987673343605,
      "grad_norm": 2.80330753326416,
      "learning_rate": 2.8454510848877046e-06,
      "loss": 1.4665,
      "step": 1440
    },
    {
      "epoch": 1.4791987673343605,
      "eval_loss": 1.8637592792510986,
      "eval_runtime": 37.1066,
      "eval_samples_per_second": 26.949,
      "eval_steps_per_second": 1.132,
      "step": 1440
    },
    {
      "epoch": 1.4802259887005649,
      "grad_norm": 2.698328733444214,
      "learning_rate": 2.843547773125238e-06,
      "loss": 1.4224,
      "step": 1441
    },
    {
      "epoch": 1.4812532100667695,
      "grad_norm": 2.663471221923828,
      "learning_rate": 2.8416444613627714e-06,
      "loss": 1.4205,
      "step": 1442
    },
    {
      "epoch": 1.4822804314329738,
      "grad_norm": 2.7955918312072754,
      "learning_rate": 2.8397411496003047e-06,
      "loss": 1.3694,
      "step": 1443
    },
    {
      "epoch": 1.4833076527991782,
      "grad_norm": 2.7320404052734375,
      "learning_rate": 2.837837837837838e-06,
      "loss": 1.4208,
      "step": 1444
    },
    {
      "epoch": 1.4843348741653826,
      "grad_norm": 2.7786989212036133,
      "learning_rate": 2.835934526075371e-06,
      "loss": 1.4392,
      "step": 1445
    },
    {
      "epoch": 1.485362095531587,
      "grad_norm": 2.6282734870910645,
      "learning_rate": 2.8340312143129044e-06,
      "loss": 1.3866,
      "step": 1446
    },
    {
      "epoch": 1.4863893168977915,
      "grad_norm": 2.615072011947632,
      "learning_rate": 2.8321279025504377e-06,
      "loss": 1.3695,
      "step": 1447
    },
    {
      "epoch": 1.487416538263996,
      "grad_norm": 2.884423017501831,
      "learning_rate": 2.830224590787971e-06,
      "loss": 1.4867,
      "step": 1448
    },
    {
      "epoch": 1.4884437596302003,
      "grad_norm": 2.7745864391326904,
      "learning_rate": 2.8283212790255045e-06,
      "loss": 1.42,
      "step": 1449
    },
    {
      "epoch": 1.4894709809964048,
      "grad_norm": 2.718130111694336,
      "learning_rate": 2.8264179672630383e-06,
      "loss": 1.4665,
      "step": 1450
    },
    {
      "epoch": 1.4894709809964048,
      "eval_loss": 1.8556917905807495,
      "eval_runtime": 37.123,
      "eval_samples_per_second": 26.938,
      "eval_steps_per_second": 1.131,
      "step": 1450
    },
    {
      "epoch": 1.4904982023626092,
      "grad_norm": 2.7502355575561523,
      "learning_rate": 2.8245146555005716e-06,
      "loss": 1.3946,
      "step": 1451
    },
    {
      "epoch": 1.4915254237288136,
      "grad_norm": 2.635500192642212,
      "learning_rate": 2.8226113437381046e-06,
      "loss": 1.3612,
      "step": 1452
    },
    {
      "epoch": 1.492552645095018,
      "grad_norm": 2.617408275604248,
      "learning_rate": 2.820708031975638e-06,
      "loss": 1.4146,
      "step": 1453
    },
    {
      "epoch": 1.4935798664612223,
      "grad_norm": 2.7706692218780518,
      "learning_rate": 2.8188047202131713e-06,
      "loss": 1.4474,
      "step": 1454
    },
    {
      "epoch": 1.494607087827427,
      "grad_norm": 2.7609055042266846,
      "learning_rate": 2.8169014084507046e-06,
      "loss": 1.4618,
      "step": 1455
    },
    {
      "epoch": 1.4956343091936313,
      "grad_norm": 2.757258892059326,
      "learning_rate": 2.814998096688238e-06,
      "loss": 1.4131,
      "step": 1456
    },
    {
      "epoch": 1.4966615305598356,
      "grad_norm": 2.5547049045562744,
      "learning_rate": 2.813094784925771e-06,
      "loss": 1.3692,
      "step": 1457
    },
    {
      "epoch": 1.49768875192604,
      "grad_norm": 2.817617177963257,
      "learning_rate": 2.8111914731633043e-06,
      "loss": 1.4664,
      "step": 1458
    },
    {
      "epoch": 1.4987159732922444,
      "grad_norm": 2.6999337673187256,
      "learning_rate": 2.8092881614008377e-06,
      "loss": 1.4126,
      "step": 1459
    },
    {
      "epoch": 1.499743194658449,
      "grad_norm": 2.571727991104126,
      "learning_rate": 2.807384849638371e-06,
      "loss": 1.3828,
      "step": 1460
    },
    {
      "epoch": 1.499743194658449,
      "eval_loss": 1.8413318395614624,
      "eval_runtime": 37.1803,
      "eval_samples_per_second": 26.896,
      "eval_steps_per_second": 1.13,
      "step": 1460
    },
    {
      "epoch": 1.5007704160246533,
      "grad_norm": 2.615567684173584,
      "learning_rate": 2.8054815378759044e-06,
      "loss": 1.4336,
      "step": 1461
    },
    {
      "epoch": 1.5017976373908577,
      "grad_norm": 2.696854591369629,
      "learning_rate": 2.8035782261134378e-06,
      "loss": 1.4015,
      "step": 1462
    },
    {
      "epoch": 1.5028248587570623,
      "grad_norm": 2.7395496368408203,
      "learning_rate": 2.8016749143509707e-06,
      "loss": 1.4009,
      "step": 1463
    },
    {
      "epoch": 1.5038520801232664,
      "grad_norm": 2.8281755447387695,
      "learning_rate": 2.799771602588504e-06,
      "loss": 1.4069,
      "step": 1464
    },
    {
      "epoch": 1.504879301489471,
      "grad_norm": 2.641793727874756,
      "learning_rate": 2.7978682908260374e-06,
      "loss": 1.4341,
      "step": 1465
    },
    {
      "epoch": 1.5059065228556754,
      "grad_norm": 2.8969004154205322,
      "learning_rate": 2.795964979063571e-06,
      "loss": 1.4493,
      "step": 1466
    },
    {
      "epoch": 1.5069337442218798,
      "grad_norm": 2.758255958557129,
      "learning_rate": 2.794061667301104e-06,
      "loss": 1.423,
      "step": 1467
    },
    {
      "epoch": 1.5079609655880843,
      "grad_norm": 2.7737441062927246,
      "learning_rate": 2.792158355538637e-06,
      "loss": 1.43,
      "step": 1468
    },
    {
      "epoch": 1.5089881869542885,
      "grad_norm": 2.798095226287842,
      "learning_rate": 2.7902550437761705e-06,
      "loss": 1.4276,
      "step": 1469
    },
    {
      "epoch": 1.510015408320493,
      "grad_norm": 2.760437488555908,
      "learning_rate": 2.7883517320137042e-06,
      "loss": 1.3927,
      "step": 1470
    },
    {
      "epoch": 1.510015408320493,
      "eval_loss": 1.8517959117889404,
      "eval_runtime": 37.163,
      "eval_samples_per_second": 26.908,
      "eval_steps_per_second": 1.13,
      "step": 1470
    },
    {
      "epoch": 1.5110426296866974,
      "grad_norm": 2.689023733139038,
      "learning_rate": 2.7864484202512376e-06,
      "loss": 1.387,
      "step": 1471
    },
    {
      "epoch": 1.5120698510529018,
      "grad_norm": 2.876424551010132,
      "learning_rate": 2.784545108488771e-06,
      "loss": 1.4183,
      "step": 1472
    },
    {
      "epoch": 1.5130970724191064,
      "grad_norm": 2.789368152618408,
      "learning_rate": 2.7826417967263043e-06,
      "loss": 1.4239,
      "step": 1473
    },
    {
      "epoch": 1.5141242937853108,
      "grad_norm": 2.945103883743286,
      "learning_rate": 2.7807384849638373e-06,
      "loss": 1.458,
      "step": 1474
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 2.694589614868164,
      "learning_rate": 2.7788351732013706e-06,
      "loss": 1.4145,
      "step": 1475
    },
    {
      "epoch": 1.5161787365177197,
      "grad_norm": 2.7276177406311035,
      "learning_rate": 2.776931861438904e-06,
      "loss": 1.4063,
      "step": 1476
    },
    {
      "epoch": 1.5172059578839239,
      "grad_norm": 2.694197177886963,
      "learning_rate": 2.7750285496764374e-06,
      "loss": 1.4125,
      "step": 1477
    },
    {
      "epoch": 1.5182331792501285,
      "grad_norm": 2.832883834838867,
      "learning_rate": 2.7731252379139707e-06,
      "loss": 1.4176,
      "step": 1478
    },
    {
      "epoch": 1.5192604006163328,
      "grad_norm": 2.67136549949646,
      "learning_rate": 2.771221926151504e-06,
      "loss": 1.3633,
      "step": 1479
    },
    {
      "epoch": 1.5202876219825372,
      "grad_norm": 2.9618499279022217,
      "learning_rate": 2.769318614389037e-06,
      "loss": 1.4283,
      "step": 1480
    },
    {
      "epoch": 1.5202876219825372,
      "eval_loss": 1.8486552238464355,
      "eval_runtime": 37.1808,
      "eval_samples_per_second": 26.896,
      "eval_steps_per_second": 1.13,
      "step": 1480
    },
    {
      "epoch": 1.5213148433487418,
      "grad_norm": 2.91403865814209,
      "learning_rate": 2.7674153026265704e-06,
      "loss": 1.4254,
      "step": 1481
    },
    {
      "epoch": 1.522342064714946,
      "grad_norm": 2.9121804237365723,
      "learning_rate": 2.7655119908641037e-06,
      "loss": 1.4251,
      "step": 1482
    },
    {
      "epoch": 1.5233692860811505,
      "grad_norm": 2.7484965324401855,
      "learning_rate": 2.763608679101637e-06,
      "loss": 1.4141,
      "step": 1483
    },
    {
      "epoch": 1.5243965074473549,
      "grad_norm": 2.913203716278076,
      "learning_rate": 2.7617053673391705e-06,
      "loss": 1.4516,
      "step": 1484
    },
    {
      "epoch": 1.5254237288135593,
      "grad_norm": 2.7844395637512207,
      "learning_rate": 2.7598020555767034e-06,
      "loss": 1.45,
      "step": 1485
    },
    {
      "epoch": 1.5264509501797638,
      "grad_norm": 2.8250396251678467,
      "learning_rate": 2.7578987438142368e-06,
      "loss": 1.3746,
      "step": 1486
    },
    {
      "epoch": 1.5274781715459682,
      "grad_norm": 2.749629497528076,
      "learning_rate": 2.75599543205177e-06,
      "loss": 1.433,
      "step": 1487
    },
    {
      "epoch": 1.5285053929121726,
      "grad_norm": 2.7559521198272705,
      "learning_rate": 2.7540921202893035e-06,
      "loss": 1.452,
      "step": 1488
    },
    {
      "epoch": 1.529532614278377,
      "grad_norm": 2.825148344039917,
      "learning_rate": 2.752188808526837e-06,
      "loss": 1.427,
      "step": 1489
    },
    {
      "epoch": 1.5305598356445813,
      "grad_norm": 2.645327091217041,
      "learning_rate": 2.75028549676437e-06,
      "loss": 1.4089,
      "step": 1490
    },
    {
      "epoch": 1.5305598356445813,
      "eval_loss": 1.8497828245162964,
      "eval_runtime": 37.3756,
      "eval_samples_per_second": 26.755,
      "eval_steps_per_second": 1.124,
      "step": 1490
    },
    {
      "epoch": 1.531587057010786,
      "grad_norm": 2.753156900405884,
      "learning_rate": 2.7483821850019036e-06,
      "loss": 1.4163,
      "step": 1491
    },
    {
      "epoch": 1.5326142783769903,
      "grad_norm": 2.8842933177948,
      "learning_rate": 2.746478873239437e-06,
      "loss": 1.4041,
      "step": 1492
    },
    {
      "epoch": 1.5336414997431946,
      "grad_norm": 2.7529265880584717,
      "learning_rate": 2.7445755614769703e-06,
      "loss": 1.41,
      "step": 1493
    },
    {
      "epoch": 1.5346687211093992,
      "grad_norm": 2.6907176971435547,
      "learning_rate": 2.7426722497145037e-06,
      "loss": 1.3841,
      "step": 1494
    },
    {
      "epoch": 1.5356959424756034,
      "grad_norm": 2.793073892593384,
      "learning_rate": 2.740768937952037e-06,
      "loss": 1.3885,
      "step": 1495
    },
    {
      "epoch": 1.536723163841808,
      "grad_norm": 2.7895116806030273,
      "learning_rate": 2.7388656261895704e-06,
      "loss": 1.4065,
      "step": 1496
    },
    {
      "epoch": 1.5377503852080123,
      "grad_norm": 2.903615713119507,
      "learning_rate": 2.7369623144271033e-06,
      "loss": 1.4812,
      "step": 1497
    },
    {
      "epoch": 1.5387776065742167,
      "grad_norm": 2.7081339359283447,
      "learning_rate": 2.7350590026646367e-06,
      "loss": 1.403,
      "step": 1498
    },
    {
      "epoch": 1.5398048279404213,
      "grad_norm": 2.75414776802063,
      "learning_rate": 2.73315569090217e-06,
      "loss": 1.3629,
      "step": 1499
    },
    {
      "epoch": 1.5408320493066254,
      "grad_norm": 2.795982599258423,
      "learning_rate": 2.7312523791397034e-06,
      "loss": 1.4551,
      "step": 1500
    },
    {
      "epoch": 1.5408320493066254,
      "eval_loss": 1.8510199785232544,
      "eval_runtime": 37.3806,
      "eval_samples_per_second": 26.752,
      "eval_steps_per_second": 1.124,
      "step": 1500
    },
    {
      "epoch": 1.54185927067283,
      "grad_norm": 2.894488573074341,
      "learning_rate": 2.7293490673772368e-06,
      "loss": 1.4794,
      "step": 1501
    },
    {
      "epoch": 1.5428864920390344,
      "grad_norm": 2.7982053756713867,
      "learning_rate": 2.7274457556147697e-06,
      "loss": 1.4365,
      "step": 1502
    },
    {
      "epoch": 1.5439137134052388,
      "grad_norm": 2.913367509841919,
      "learning_rate": 2.725542443852303e-06,
      "loss": 1.4132,
      "step": 1503
    },
    {
      "epoch": 1.5449409347714433,
      "grad_norm": 2.6658358573913574,
      "learning_rate": 2.7236391320898365e-06,
      "loss": 1.4288,
      "step": 1504
    },
    {
      "epoch": 1.5459681561376477,
      "grad_norm": 2.710700035095215,
      "learning_rate": 2.72173582032737e-06,
      "loss": 1.3936,
      "step": 1505
    },
    {
      "epoch": 1.546995377503852,
      "grad_norm": 2.7881481647491455,
      "learning_rate": 2.719832508564903e-06,
      "loss": 1.423,
      "step": 1506
    },
    {
      "epoch": 1.5480225988700564,
      "grad_norm": 2.7621042728424072,
      "learning_rate": 2.717929196802436e-06,
      "loss": 1.4218,
      "step": 1507
    },
    {
      "epoch": 1.5490498202362608,
      "grad_norm": 2.7504117488861084,
      "learning_rate": 2.7160258850399695e-06,
      "loss": 1.3918,
      "step": 1508
    },
    {
      "epoch": 1.5500770416024654,
      "grad_norm": 2.719702959060669,
      "learning_rate": 2.714122573277503e-06,
      "loss": 1.4154,
      "step": 1509
    },
    {
      "epoch": 1.5511042629686698,
      "grad_norm": 2.6698062419891357,
      "learning_rate": 2.712219261515036e-06,
      "loss": 1.4084,
      "step": 1510
    },
    {
      "epoch": 1.5511042629686698,
      "eval_loss": 1.8468451499938965,
      "eval_runtime": 36.9179,
      "eval_samples_per_second": 27.087,
      "eval_steps_per_second": 1.138,
      "step": 1510
    },
    {
      "epoch": 1.5521314843348741,
      "grad_norm": 2.816978693008423,
      "learning_rate": 2.71031594975257e-06,
      "loss": 1.4195,
      "step": 1511
    },
    {
      "epoch": 1.5531587057010787,
      "grad_norm": 2.8047056198120117,
      "learning_rate": 2.7084126379901034e-06,
      "loss": 1.4395,
      "step": 1512
    },
    {
      "epoch": 1.5541859270672829,
      "grad_norm": 2.8461363315582275,
      "learning_rate": 2.7065093262276367e-06,
      "loss": 1.4472,
      "step": 1513
    },
    {
      "epoch": 1.5552131484334875,
      "grad_norm": 2.712890863418579,
      "learning_rate": 2.7046060144651697e-06,
      "loss": 1.4538,
      "step": 1514
    },
    {
      "epoch": 1.5562403697996918,
      "grad_norm": 2.844165802001953,
      "learning_rate": 2.702702702702703e-06,
      "loss": 1.4719,
      "step": 1515
    },
    {
      "epoch": 1.5572675911658962,
      "grad_norm": 2.629138231277466,
      "learning_rate": 2.7007993909402364e-06,
      "loss": 1.3441,
      "step": 1516
    },
    {
      "epoch": 1.5582948125321008,
      "grad_norm": 2.7072019577026367,
      "learning_rate": 2.6988960791777697e-06,
      "loss": 1.4692,
      "step": 1517
    },
    {
      "epoch": 1.559322033898305,
      "grad_norm": 2.739567279815674,
      "learning_rate": 2.696992767415303e-06,
      "loss": 1.4215,
      "step": 1518
    },
    {
      "epoch": 1.5603492552645095,
      "grad_norm": 2.7846791744232178,
      "learning_rate": 2.695089455652836e-06,
      "loss": 1.4151,
      "step": 1519
    },
    {
      "epoch": 1.5613764766307139,
      "grad_norm": 2.7417876720428467,
      "learning_rate": 2.6931861438903694e-06,
      "loss": 1.3797,
      "step": 1520
    },
    {
      "epoch": 1.5613764766307139,
      "eval_loss": 1.847011923789978,
      "eval_runtime": 37.0386,
      "eval_samples_per_second": 26.999,
      "eval_steps_per_second": 1.134,
      "step": 1520
    },
    {
      "epoch": 1.5624036979969183,
      "grad_norm": 2.743271827697754,
      "learning_rate": 2.6912828321279028e-06,
      "loss": 1.4372,
      "step": 1521
    },
    {
      "epoch": 1.5634309193631228,
      "grad_norm": 2.8119285106658936,
      "learning_rate": 2.689379520365436e-06,
      "loss": 1.4075,
      "step": 1522
    },
    {
      "epoch": 1.5644581407293272,
      "grad_norm": 2.7443952560424805,
      "learning_rate": 2.6874762086029695e-06,
      "loss": 1.4442,
      "step": 1523
    },
    {
      "epoch": 1.5654853620955316,
      "grad_norm": 2.7869346141815186,
      "learning_rate": 2.6855728968405024e-06,
      "loss": 1.3948,
      "step": 1524
    },
    {
      "epoch": 1.5665125834617362,
      "grad_norm": 2.750185251235962,
      "learning_rate": 2.683669585078036e-06,
      "loss": 1.4313,
      "step": 1525
    },
    {
      "epoch": 1.5675398048279403,
      "grad_norm": 2.787389039993286,
      "learning_rate": 2.681766273315569e-06,
      "loss": 1.3748,
      "step": 1526
    },
    {
      "epoch": 1.568567026194145,
      "grad_norm": 2.6675543785095215,
      "learning_rate": 2.6798629615531025e-06,
      "loss": 1.4089,
      "step": 1527
    },
    {
      "epoch": 1.5695942475603493,
      "grad_norm": 2.6468663215637207,
      "learning_rate": 2.677959649790636e-06,
      "loss": 1.4148,
      "step": 1528
    },
    {
      "epoch": 1.5706214689265536,
      "grad_norm": 2.7241196632385254,
      "learning_rate": 2.676056338028169e-06,
      "loss": 1.4498,
      "step": 1529
    },
    {
      "epoch": 1.5716486902927582,
      "grad_norm": 2.705486536026001,
      "learning_rate": 2.674153026265702e-06,
      "loss": 1.4324,
      "step": 1530
    },
    {
      "epoch": 1.5716486902927582,
      "eval_loss": 1.8467438220977783,
      "eval_runtime": 37.0576,
      "eval_samples_per_second": 26.985,
      "eval_steps_per_second": 1.133,
      "step": 1530
    },
    {
      "epoch": 1.5726759116589624,
      "grad_norm": 2.805429458618164,
      "learning_rate": 2.672249714503236e-06,
      "loss": 1.4475,
      "step": 1531
    },
    {
      "epoch": 1.573703133025167,
      "grad_norm": 2.7074034214019775,
      "learning_rate": 2.6703464027407693e-06,
      "loss": 1.3997,
      "step": 1532
    },
    {
      "epoch": 1.5747303543913713,
      "grad_norm": 2.737571954727173,
      "learning_rate": 2.6684430909783027e-06,
      "loss": 1.4238,
      "step": 1533
    },
    {
      "epoch": 1.5757575757575757,
      "grad_norm": 2.703036069869995,
      "learning_rate": 2.666539779215836e-06,
      "loss": 1.4107,
      "step": 1534
    },
    {
      "epoch": 1.5767847971237803,
      "grad_norm": 2.7987451553344727,
      "learning_rate": 2.6646364674533694e-06,
      "loss": 1.3927,
      "step": 1535
    },
    {
      "epoch": 1.5778120184899846,
      "grad_norm": 2.622497081756592,
      "learning_rate": 2.6627331556909024e-06,
      "loss": 1.3215,
      "step": 1536
    },
    {
      "epoch": 1.578839239856189,
      "grad_norm": 2.5890607833862305,
      "learning_rate": 2.6608298439284357e-06,
      "loss": 1.3994,
      "step": 1537
    },
    {
      "epoch": 1.5798664612223934,
      "grad_norm": 2.6438405513763428,
      "learning_rate": 2.658926532165969e-06,
      "loss": 1.4042,
      "step": 1538
    },
    {
      "epoch": 1.5808936825885977,
      "grad_norm": 2.690793752670288,
      "learning_rate": 2.6570232204035024e-06,
      "loss": 1.3673,
      "step": 1539
    },
    {
      "epoch": 1.5819209039548023,
      "grad_norm": 2.7485456466674805,
      "learning_rate": 2.655119908641036e-06,
      "loss": 1.3979,
      "step": 1540
    },
    {
      "epoch": 1.5819209039548023,
      "eval_loss": 1.852581262588501,
      "eval_runtime": 37.0688,
      "eval_samples_per_second": 26.977,
      "eval_steps_per_second": 1.133,
      "step": 1540
    },
    {
      "epoch": 1.5829481253210067,
      "grad_norm": 2.8531415462493896,
      "learning_rate": 2.6532165968785687e-06,
      "loss": 1.4094,
      "step": 1541
    },
    {
      "epoch": 1.583975346687211,
      "grad_norm": 2.638291358947754,
      "learning_rate": 2.651313285116102e-06,
      "loss": 1.3708,
      "step": 1542
    },
    {
      "epoch": 1.5850025680534157,
      "grad_norm": 2.717550754547119,
      "learning_rate": 2.6494099733536355e-06,
      "loss": 1.3821,
      "step": 1543
    },
    {
      "epoch": 1.5860297894196198,
      "grad_norm": 2.91959810256958,
      "learning_rate": 2.647506661591169e-06,
      "loss": 1.5532,
      "step": 1544
    },
    {
      "epoch": 1.5870570107858244,
      "grad_norm": 2.7281882762908936,
      "learning_rate": 2.645603349828702e-06,
      "loss": 1.4361,
      "step": 1545
    },
    {
      "epoch": 1.5880842321520288,
      "grad_norm": 2.732283592224121,
      "learning_rate": 2.643700038066235e-06,
      "loss": 1.3669,
      "step": 1546
    },
    {
      "epoch": 1.5891114535182331,
      "grad_norm": 2.731055498123169,
      "learning_rate": 2.6417967263037685e-06,
      "loss": 1.4223,
      "step": 1547
    },
    {
      "epoch": 1.5901386748844377,
      "grad_norm": 2.831334352493286,
      "learning_rate": 2.639893414541302e-06,
      "loss": 1.4454,
      "step": 1548
    },
    {
      "epoch": 1.5911658962506419,
      "grad_norm": 2.920548677444458,
      "learning_rate": 2.6379901027788352e-06,
      "loss": 1.4686,
      "step": 1549
    },
    {
      "epoch": 1.5921931176168465,
      "grad_norm": 2.6671719551086426,
      "learning_rate": 2.6360867910163686e-06,
      "loss": 1.398,
      "step": 1550
    },
    {
      "epoch": 1.5921931176168465,
      "eval_loss": 1.857794165611267,
      "eval_runtime": 37.0817,
      "eval_samples_per_second": 26.967,
      "eval_steps_per_second": 1.133,
      "step": 1550
    },
    {
      "epoch": 1.5932203389830508,
      "grad_norm": 2.685175657272339,
      "learning_rate": 2.6341834792539024e-06,
      "loss": 1.4176,
      "step": 1551
    },
    {
      "epoch": 1.5942475603492552,
      "grad_norm": 2.7412822246551514,
      "learning_rate": 2.6322801674914357e-06,
      "loss": 1.4195,
      "step": 1552
    },
    {
      "epoch": 1.5952747817154598,
      "grad_norm": 2.7461724281311035,
      "learning_rate": 2.6303768557289687e-06,
      "loss": 1.4081,
      "step": 1553
    },
    {
      "epoch": 1.5963020030816641,
      "grad_norm": 2.6494853496551514,
      "learning_rate": 2.628473543966502e-06,
      "loss": 1.3759,
      "step": 1554
    },
    {
      "epoch": 1.5973292244478685,
      "grad_norm": 2.736752986907959,
      "learning_rate": 2.6265702322040354e-06,
      "loss": 1.4415,
      "step": 1555
    },
    {
      "epoch": 1.5983564458140729,
      "grad_norm": 2.733452796936035,
      "learning_rate": 2.6246669204415688e-06,
      "loss": 1.3867,
      "step": 1556
    },
    {
      "epoch": 1.5993836671802772,
      "grad_norm": 2.692986249923706,
      "learning_rate": 2.622763608679102e-06,
      "loss": 1.3811,
      "step": 1557
    },
    {
      "epoch": 1.6004108885464818,
      "grad_norm": 2.580052614212036,
      "learning_rate": 2.620860296916635e-06,
      "loss": 1.399,
      "step": 1558
    },
    {
      "epoch": 1.6014381099126862,
      "grad_norm": 2.651503086090088,
      "learning_rate": 2.6189569851541684e-06,
      "loss": 1.34,
      "step": 1559
    },
    {
      "epoch": 1.6024653312788906,
      "grad_norm": 2.8157193660736084,
      "learning_rate": 2.617053673391702e-06,
      "loss": 1.4437,
      "step": 1560
    },
    {
      "epoch": 1.6024653312788906,
      "eval_loss": 1.8588496446609497,
      "eval_runtime": 37.0786,
      "eval_samples_per_second": 26.97,
      "eval_steps_per_second": 1.133,
      "step": 1560
    },
    {
      "epoch": 1.6034925526450952,
      "grad_norm": 2.893589973449707,
      "learning_rate": 2.615150361629235e-06,
      "loss": 1.4775,
      "step": 1561
    },
    {
      "epoch": 1.6045197740112993,
      "grad_norm": 2.719688653945923,
      "learning_rate": 2.6132470498667685e-06,
      "loss": 1.432,
      "step": 1562
    },
    {
      "epoch": 1.605546995377504,
      "grad_norm": 2.5626275539398193,
      "learning_rate": 2.6113437381043015e-06,
      "loss": 1.4118,
      "step": 1563
    },
    {
      "epoch": 1.6065742167437083,
      "grad_norm": 2.6879632472991943,
      "learning_rate": 2.609440426341835e-06,
      "loss": 1.398,
      "step": 1564
    },
    {
      "epoch": 1.6076014381099126,
      "grad_norm": 2.890155792236328,
      "learning_rate": 2.607537114579368e-06,
      "loss": 1.4336,
      "step": 1565
    },
    {
      "epoch": 1.6086286594761172,
      "grad_norm": 2.8333349227905273,
      "learning_rate": 2.6056338028169015e-06,
      "loss": 1.4148,
      "step": 1566
    },
    {
      "epoch": 1.6096558808423214,
      "grad_norm": 2.729215383529663,
      "learning_rate": 2.603730491054435e-06,
      "loss": 1.3831,
      "step": 1567
    },
    {
      "epoch": 1.610683102208526,
      "grad_norm": 2.800276279449463,
      "learning_rate": 2.601827179291968e-06,
      "loss": 1.4315,
      "step": 1568
    },
    {
      "epoch": 1.6117103235747303,
      "grad_norm": 2.7900125980377197,
      "learning_rate": 2.599923867529501e-06,
      "loss": 1.4257,
      "step": 1569
    },
    {
      "epoch": 1.6127375449409347,
      "grad_norm": 2.7753336429595947,
      "learning_rate": 2.5980205557670346e-06,
      "loss": 1.4089,
      "step": 1570
    },
    {
      "epoch": 1.6127375449409347,
      "eval_loss": 1.8533762693405151,
      "eval_runtime": 37.0837,
      "eval_samples_per_second": 26.966,
      "eval_steps_per_second": 1.133,
      "step": 1570
    },
    {
      "epoch": 1.6137647663071393,
      "grad_norm": 2.9164156913757324,
      "learning_rate": 2.5961172440045684e-06,
      "loss": 1.4467,
      "step": 1571
    },
    {
      "epoch": 1.6147919876733436,
      "grad_norm": 2.747676134109497,
      "learning_rate": 2.5942139322421017e-06,
      "loss": 1.4521,
      "step": 1572
    },
    {
      "epoch": 1.615819209039548,
      "grad_norm": 2.6693432331085205,
      "learning_rate": 2.592310620479635e-06,
      "loss": 1.3982,
      "step": 1573
    },
    {
      "epoch": 1.6168464304057526,
      "grad_norm": 2.5721216201782227,
      "learning_rate": 2.5904073087171684e-06,
      "loss": 1.3657,
      "step": 1574
    },
    {
      "epoch": 1.6178736517719567,
      "grad_norm": 2.706143617630005,
      "learning_rate": 2.5885039969547014e-06,
      "loss": 1.4315,
      "step": 1575
    },
    {
      "epoch": 1.6189008731381613,
      "grad_norm": 2.59006404876709,
      "learning_rate": 2.5866006851922347e-06,
      "loss": 1.372,
      "step": 1576
    },
    {
      "epoch": 1.6199280945043657,
      "grad_norm": 2.7117512226104736,
      "learning_rate": 2.584697373429768e-06,
      "loss": 1.3834,
      "step": 1577
    },
    {
      "epoch": 1.62095531587057,
      "grad_norm": 2.718029737472534,
      "learning_rate": 2.5827940616673015e-06,
      "loss": 1.3978,
      "step": 1578
    },
    {
      "epoch": 1.6219825372367747,
      "grad_norm": 2.7660539150238037,
      "learning_rate": 2.580890749904835e-06,
      "loss": 1.4505,
      "step": 1579
    },
    {
      "epoch": 1.6230097586029788,
      "grad_norm": 2.7671236991882324,
      "learning_rate": 2.5789874381423678e-06,
      "loss": 1.3951,
      "step": 1580
    },
    {
      "epoch": 1.6230097586029788,
      "eval_loss": 1.8412485122680664,
      "eval_runtime": 37.1162,
      "eval_samples_per_second": 26.942,
      "eval_steps_per_second": 1.132,
      "step": 1580
    },
    {
      "epoch": 1.6240369799691834,
      "grad_norm": 2.6504807472229004,
      "learning_rate": 2.577084126379901e-06,
      "loss": 1.3957,
      "step": 1581
    },
    {
      "epoch": 1.6250642013353878,
      "grad_norm": 2.688166618347168,
      "learning_rate": 2.5751808146174345e-06,
      "loss": 1.3686,
      "step": 1582
    },
    {
      "epoch": 1.6260914227015921,
      "grad_norm": 2.7034637928009033,
      "learning_rate": 2.573277502854968e-06,
      "loss": 1.4201,
      "step": 1583
    },
    {
      "epoch": 1.6271186440677967,
      "grad_norm": 2.6097683906555176,
      "learning_rate": 2.5713741910925012e-06,
      "loss": 1.3881,
      "step": 1584
    },
    {
      "epoch": 1.6281458654340009,
      "grad_norm": 2.6197433471679688,
      "learning_rate": 2.569470879330034e-06,
      "loss": 1.3962,
      "step": 1585
    },
    {
      "epoch": 1.6291730868002054,
      "grad_norm": 3.0436367988586426,
      "learning_rate": 2.5675675675675675e-06,
      "loss": 1.4714,
      "step": 1586
    },
    {
      "epoch": 1.6302003081664098,
      "grad_norm": 2.708744764328003,
      "learning_rate": 2.565664255805101e-06,
      "loss": 1.4007,
      "step": 1587
    },
    {
      "epoch": 1.6312275295326142,
      "grad_norm": 2.708279848098755,
      "learning_rate": 2.5637609440426343e-06,
      "loss": 1.4284,
      "step": 1588
    },
    {
      "epoch": 1.6322547508988188,
      "grad_norm": 2.7955429553985596,
      "learning_rate": 2.5618576322801676e-06,
      "loss": 1.3736,
      "step": 1589
    },
    {
      "epoch": 1.6332819722650231,
      "grad_norm": 2.890211820602417,
      "learning_rate": 2.5599543205177006e-06,
      "loss": 1.4671,
      "step": 1590
    },
    {
      "epoch": 1.6332819722650231,
      "eval_loss": 1.8467649221420288,
      "eval_runtime": 37.284,
      "eval_samples_per_second": 26.821,
      "eval_steps_per_second": 1.126,
      "step": 1590
    },
    {
      "epoch": 1.6343091936312275,
      "grad_norm": 2.698066234588623,
      "learning_rate": 2.558051008755234e-06,
      "loss": 1.4229,
      "step": 1591
    },
    {
      "epoch": 1.635336414997432,
      "grad_norm": 2.8393352031707764,
      "learning_rate": 2.5561476969927677e-06,
      "loss": 1.469,
      "step": 1592
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 2.589914321899414,
      "learning_rate": 2.554244385230301e-06,
      "loss": 1.3853,
      "step": 1593
    },
    {
      "epoch": 1.6373908577298408,
      "grad_norm": 2.630443572998047,
      "learning_rate": 2.5523410734678344e-06,
      "loss": 1.4176,
      "step": 1594
    },
    {
      "epoch": 1.6384180790960452,
      "grad_norm": 2.784212112426758,
      "learning_rate": 2.550437761705368e-06,
      "loss": 1.344,
      "step": 1595
    },
    {
      "epoch": 1.6394453004622496,
      "grad_norm": 2.736009359359741,
      "learning_rate": 2.548534449942901e-06,
      "loss": 1.3867,
      "step": 1596
    },
    {
      "epoch": 1.6404725218284542,
      "grad_norm": 2.9134042263031006,
      "learning_rate": 2.546631138180434e-06,
      "loss": 1.4214,
      "step": 1597
    },
    {
      "epoch": 1.6414997431946583,
      "grad_norm": 2.995575428009033,
      "learning_rate": 2.5447278264179675e-06,
      "loss": 1.409,
      "step": 1598
    },
    {
      "epoch": 1.642526964560863,
      "grad_norm": 2.744499683380127,
      "learning_rate": 2.542824514655501e-06,
      "loss": 1.3727,
      "step": 1599
    },
    {
      "epoch": 1.6435541859270673,
      "grad_norm": 2.835494041442871,
      "learning_rate": 2.540921202893034e-06,
      "loss": 1.4636,
      "step": 1600
    },
    {
      "epoch": 1.6435541859270673,
      "eval_loss": 1.8454105854034424,
      "eval_runtime": 37.4381,
      "eval_samples_per_second": 26.711,
      "eval_steps_per_second": 1.122,
      "step": 1600
    },
    {
      "epoch": 1.6445814072932716,
      "grad_norm": 2.817537307739258,
      "learning_rate": 2.5390178911305675e-06,
      "loss": 1.4416,
      "step": 1601
    },
    {
      "epoch": 1.6456086286594762,
      "grad_norm": 2.868093252182007,
      "learning_rate": 2.5371145793681005e-06,
      "loss": 1.4187,
      "step": 1602
    },
    {
      "epoch": 1.6466358500256806,
      "grad_norm": 2.861865758895874,
      "learning_rate": 2.535211267605634e-06,
      "loss": 1.4067,
      "step": 1603
    },
    {
      "epoch": 1.647663071391885,
      "grad_norm": 2.7015087604522705,
      "learning_rate": 2.533307955843167e-06,
      "loss": 1.3528,
      "step": 1604
    },
    {
      "epoch": 1.6486902927580893,
      "grad_norm": 2.7585439682006836,
      "learning_rate": 2.5314046440807006e-06,
      "loss": 1.4157,
      "step": 1605
    },
    {
      "epoch": 1.6497175141242937,
      "grad_norm": 2.853614330291748,
      "learning_rate": 2.529501332318234e-06,
      "loss": 1.3932,
      "step": 1606
    },
    {
      "epoch": 1.6507447354904983,
      "grad_norm": 2.9736239910125732,
      "learning_rate": 2.5275980205557673e-06,
      "loss": 1.4511,
      "step": 1607
    },
    {
      "epoch": 1.6517719568567026,
      "grad_norm": 2.955389976501465,
      "learning_rate": 2.5256947087933002e-06,
      "loss": 1.4777,
      "step": 1608
    },
    {
      "epoch": 1.652799178222907,
      "grad_norm": 2.9508774280548096,
      "learning_rate": 2.5237913970308336e-06,
      "loss": 1.4523,
      "step": 1609
    },
    {
      "epoch": 1.6538263995891116,
      "grad_norm": 2.88675856590271,
      "learning_rate": 2.521888085268367e-06,
      "loss": 1.4064,
      "step": 1610
    },
    {
      "epoch": 1.6538263995891116,
      "eval_loss": 1.8569127321243286,
      "eval_runtime": 36.9819,
      "eval_samples_per_second": 27.04,
      "eval_steps_per_second": 1.136,
      "step": 1610
    },
    {
      "epoch": 1.6548536209553157,
      "grad_norm": 2.8439440727233887,
      "learning_rate": 2.5199847735059003e-06,
      "loss": 1.4591,
      "step": 1611
    },
    {
      "epoch": 1.6558808423215203,
      "grad_norm": 2.9314420223236084,
      "learning_rate": 2.518081461743434e-06,
      "loss": 1.39,
      "step": 1612
    },
    {
      "epoch": 1.6569080636877247,
      "grad_norm": 2.6517245769500732,
      "learning_rate": 2.5161781499809675e-06,
      "loss": 1.4234,
      "step": 1613
    },
    {
      "epoch": 1.657935285053929,
      "grad_norm": 2.5690908432006836,
      "learning_rate": 2.5142748382185004e-06,
      "loss": 1.3762,
      "step": 1614
    },
    {
      "epoch": 1.6589625064201337,
      "grad_norm": 2.808892011642456,
      "learning_rate": 2.5123715264560338e-06,
      "loss": 1.3727,
      "step": 1615
    },
    {
      "epoch": 1.6599897277863378,
      "grad_norm": 2.777846336364746,
      "learning_rate": 2.510468214693567e-06,
      "loss": 1.3941,
      "step": 1616
    },
    {
      "epoch": 1.6610169491525424,
      "grad_norm": 2.8298423290252686,
      "learning_rate": 2.5085649029311005e-06,
      "loss": 1.3455,
      "step": 1617
    },
    {
      "epoch": 1.6620441705187468,
      "grad_norm": 2.6464784145355225,
      "learning_rate": 2.506661591168634e-06,
      "loss": 1.3061,
      "step": 1618
    },
    {
      "epoch": 1.6630713918849511,
      "grad_norm": 2.7149951457977295,
      "learning_rate": 2.504758279406167e-06,
      "loss": 1.3995,
      "step": 1619
    },
    {
      "epoch": 1.6640986132511557,
      "grad_norm": 3.0328569412231445,
      "learning_rate": 2.5028549676437e-06,
      "loss": 1.4343,
      "step": 1620
    },
    {
      "epoch": 1.6640986132511557,
      "eval_loss": 1.8629075288772583,
      "eval_runtime": 37.1211,
      "eval_samples_per_second": 26.939,
      "eval_steps_per_second": 1.131,
      "step": 1620
    },
    {
      "epoch": 1.66512583461736,
      "grad_norm": 2.8838441371917725,
      "learning_rate": 2.5009516558812335e-06,
      "loss": 1.3817,
      "step": 1621
    },
    {
      "epoch": 1.6661530559835644,
      "grad_norm": 2.843242645263672,
      "learning_rate": 2.499048344118767e-06,
      "loss": 1.4322,
      "step": 1622
    },
    {
      "epoch": 1.667180277349769,
      "grad_norm": 2.642946720123291,
      "learning_rate": 2.4971450323563002e-06,
      "loss": 1.4702,
      "step": 1623
    },
    {
      "epoch": 1.6682074987159732,
      "grad_norm": 2.666194200515747,
      "learning_rate": 2.4952417205938336e-06,
      "loss": 1.3808,
      "step": 1624
    },
    {
      "epoch": 1.6692347200821778,
      "grad_norm": 2.7515366077423096,
      "learning_rate": 2.4933384088313666e-06,
      "loss": 1.3892,
      "step": 1625
    },
    {
      "epoch": 1.6702619414483821,
      "grad_norm": 2.7051661014556885,
      "learning_rate": 2.4914350970689e-06,
      "loss": 1.3809,
      "step": 1626
    },
    {
      "epoch": 1.6712891628145865,
      "grad_norm": 2.7311975955963135,
      "learning_rate": 2.4895317853064337e-06,
      "loss": 1.3952,
      "step": 1627
    },
    {
      "epoch": 1.672316384180791,
      "grad_norm": 2.8365893363952637,
      "learning_rate": 2.4876284735439666e-06,
      "loss": 1.4505,
      "step": 1628
    },
    {
      "epoch": 1.6733436055469952,
      "grad_norm": 2.644507884979248,
      "learning_rate": 2.4857251617815e-06,
      "loss": 1.4114,
      "step": 1629
    },
    {
      "epoch": 1.6743708269131998,
      "grad_norm": 2.7952826023101807,
      "learning_rate": 2.4838218500190334e-06,
      "loss": 1.4223,
      "step": 1630
    },
    {
      "epoch": 1.6743708269131998,
      "eval_loss": 1.8577097654342651,
      "eval_runtime": 37.1645,
      "eval_samples_per_second": 26.907,
      "eval_steps_per_second": 1.13,
      "step": 1630
    },
    {
      "epoch": 1.6753980482794042,
      "grad_norm": 2.787752866744995,
      "learning_rate": 2.4819185382565667e-06,
      "loss": 1.4022,
      "step": 1631
    },
    {
      "epoch": 1.6764252696456086,
      "grad_norm": 2.6963443756103516,
      "learning_rate": 2.4800152264941e-06,
      "loss": 1.3974,
      "step": 1632
    },
    {
      "epoch": 1.6774524910118132,
      "grad_norm": 2.6753835678100586,
      "learning_rate": 2.478111914731633e-06,
      "loss": 1.3639,
      "step": 1633
    },
    {
      "epoch": 1.6784797123780173,
      "grad_norm": 2.7086808681488037,
      "learning_rate": 2.4762086029691664e-06,
      "loss": 1.3987,
      "step": 1634
    },
    {
      "epoch": 1.6795069337442219,
      "grad_norm": 3.2228872776031494,
      "learning_rate": 2.4743052912066998e-06,
      "loss": 1.5202,
      "step": 1635
    },
    {
      "epoch": 1.6805341551104263,
      "grad_norm": 2.8254599571228027,
      "learning_rate": 2.472401979444233e-06,
      "loss": 1.3781,
      "step": 1636
    },
    {
      "epoch": 1.6815613764766306,
      "grad_norm": 2.9406354427337646,
      "learning_rate": 2.4704986676817665e-06,
      "loss": 1.4686,
      "step": 1637
    },
    {
      "epoch": 1.6825885978428352,
      "grad_norm": 2.8287105560302734,
      "learning_rate": 2.4685953559193e-06,
      "loss": 1.4026,
      "step": 1638
    },
    {
      "epoch": 1.6836158192090396,
      "grad_norm": 2.923633337020874,
      "learning_rate": 2.466692044156833e-06,
      "loss": 1.4493,
      "step": 1639
    },
    {
      "epoch": 1.684643040575244,
      "grad_norm": 2.9239513874053955,
      "learning_rate": 2.4647887323943666e-06,
      "loss": 1.3878,
      "step": 1640
    },
    {
      "epoch": 1.684643040575244,
      "eval_loss": 1.853607416152954,
      "eval_runtime": 37.1661,
      "eval_samples_per_second": 26.906,
      "eval_steps_per_second": 1.13,
      "step": 1640
    },
    {
      "epoch": 1.6856702619414485,
      "grad_norm": 2.699998617172241,
      "learning_rate": 2.4628854206319e-06,
      "loss": 1.3637,
      "step": 1641
    },
    {
      "epoch": 1.6866974833076527,
      "grad_norm": 2.901700258255005,
      "learning_rate": 2.460982108869433e-06,
      "loss": 1.3554,
      "step": 1642
    },
    {
      "epoch": 1.6877247046738573,
      "grad_norm": 3.011157512664795,
      "learning_rate": 2.4590787971069662e-06,
      "loss": 1.4487,
      "step": 1643
    },
    {
      "epoch": 1.6887519260400616,
      "grad_norm": 2.672830581665039,
      "learning_rate": 2.4571754853444996e-06,
      "loss": 1.3906,
      "step": 1644
    },
    {
      "epoch": 1.689779147406266,
      "grad_norm": 2.7201497554779053,
      "learning_rate": 2.455272173582033e-06,
      "loss": 1.3747,
      "step": 1645
    },
    {
      "epoch": 1.6908063687724706,
      "grad_norm": 2.979395627975464,
      "learning_rate": 2.4533688618195663e-06,
      "loss": 1.3895,
      "step": 1646
    },
    {
      "epoch": 1.6918335901386747,
      "grad_norm": 2.8048512935638428,
      "learning_rate": 2.4514655500570997e-06,
      "loss": 1.3271,
      "step": 1647
    },
    {
      "epoch": 1.6928608115048793,
      "grad_norm": 2.6457700729370117,
      "learning_rate": 2.449562238294633e-06,
      "loss": 1.3679,
      "step": 1648
    },
    {
      "epoch": 1.6938880328710837,
      "grad_norm": 2.732454299926758,
      "learning_rate": 2.4476589265321664e-06,
      "loss": 1.3332,
      "step": 1649
    },
    {
      "epoch": 1.694915254237288,
      "grad_norm": 3.190871238708496,
      "learning_rate": 2.4457556147696993e-06,
      "loss": 1.4437,
      "step": 1650
    },
    {
      "epoch": 1.694915254237288,
      "eval_loss": 1.8642219305038452,
      "eval_runtime": 37.1865,
      "eval_samples_per_second": 26.891,
      "eval_steps_per_second": 1.129,
      "step": 1650
    },
    {
      "epoch": 1.6959424756034926,
      "grad_norm": 2.85784912109375,
      "learning_rate": 2.4438523030072327e-06,
      "loss": 1.4121,
      "step": 1651
    },
    {
      "epoch": 1.696969696969697,
      "grad_norm": 2.766545534133911,
      "learning_rate": 2.441948991244766e-06,
      "loss": 1.4231,
      "step": 1652
    },
    {
      "epoch": 1.6979969183359014,
      "grad_norm": 2.8962197303771973,
      "learning_rate": 2.4400456794822994e-06,
      "loss": 1.4146,
      "step": 1653
    },
    {
      "epoch": 1.6990241397021058,
      "grad_norm": 2.907231569290161,
      "learning_rate": 2.438142367719833e-06,
      "loss": 1.4375,
      "step": 1654
    },
    {
      "epoch": 1.7000513610683101,
      "grad_norm": 2.8746557235717773,
      "learning_rate": 2.4362390559573657e-06,
      "loss": 1.4002,
      "step": 1655
    },
    {
      "epoch": 1.7010785824345147,
      "grad_norm": 2.741013765335083,
      "learning_rate": 2.434335744194899e-06,
      "loss": 1.3865,
      "step": 1656
    },
    {
      "epoch": 1.702105803800719,
      "grad_norm": 3.118908643722534,
      "learning_rate": 2.432432432432433e-06,
      "loss": 1.4412,
      "step": 1657
    },
    {
      "epoch": 1.7031330251669234,
      "grad_norm": 2.789179563522339,
      "learning_rate": 2.4305291206699662e-06,
      "loss": 1.3446,
      "step": 1658
    },
    {
      "epoch": 1.704160246533128,
      "grad_norm": 2.9241371154785156,
      "learning_rate": 2.428625808907499e-06,
      "loss": 1.4263,
      "step": 1659
    },
    {
      "epoch": 1.7051874678993322,
      "grad_norm": 2.8632888793945312,
      "learning_rate": 2.4267224971450325e-06,
      "loss": 1.4413,
      "step": 1660
    },
    {
      "epoch": 1.7051874678993322,
      "eval_loss": 1.85648512840271,
      "eval_runtime": 37.199,
      "eval_samples_per_second": 26.882,
      "eval_steps_per_second": 1.129,
      "step": 1660
    },
    {
      "epoch": 1.7062146892655368,
      "grad_norm": 2.948183059692383,
      "learning_rate": 2.424819185382566e-06,
      "loss": 1.4058,
      "step": 1661
    },
    {
      "epoch": 1.7072419106317411,
      "grad_norm": 3.04127836227417,
      "learning_rate": 2.4229158736200993e-06,
      "loss": 1.3582,
      "step": 1662
    },
    {
      "epoch": 1.7082691319979455,
      "grad_norm": 2.6956393718719482,
      "learning_rate": 2.4210125618576326e-06,
      "loss": 1.3705,
      "step": 1663
    },
    {
      "epoch": 1.70929635336415,
      "grad_norm": 2.7690250873565674,
      "learning_rate": 2.4191092500951656e-06,
      "loss": 1.3869,
      "step": 1664
    },
    {
      "epoch": 1.7103235747303542,
      "grad_norm": 2.851231813430786,
      "learning_rate": 2.417205938332699e-06,
      "loss": 1.4198,
      "step": 1665
    },
    {
      "epoch": 1.7113507960965588,
      "grad_norm": 2.9241318702697754,
      "learning_rate": 2.4153026265702323e-06,
      "loss": 1.3441,
      "step": 1666
    },
    {
      "epoch": 1.7123780174627632,
      "grad_norm": 2.7888636589050293,
      "learning_rate": 2.4133993148077657e-06,
      "loss": 1.3765,
      "step": 1667
    },
    {
      "epoch": 1.7134052388289676,
      "grad_norm": 2.891366481781006,
      "learning_rate": 2.411496003045299e-06,
      "loss": 1.4393,
      "step": 1668
    },
    {
      "epoch": 1.7144324601951721,
      "grad_norm": 2.8162693977355957,
      "learning_rate": 2.4095926912828324e-06,
      "loss": 1.3811,
      "step": 1669
    },
    {
      "epoch": 1.7154596815613765,
      "grad_norm": 2.6741931438446045,
      "learning_rate": 2.4076893795203658e-06,
      "loss": 1.404,
      "step": 1670
    },
    {
      "epoch": 1.7154596815613765,
      "eval_loss": 1.8440946340560913,
      "eval_runtime": 37.1273,
      "eval_samples_per_second": 26.934,
      "eval_steps_per_second": 1.131,
      "step": 1670
    },
    {
      "epoch": 1.7164869029275809,
      "grad_norm": 2.834951400756836,
      "learning_rate": 2.405786067757899e-06,
      "loss": 1.377,
      "step": 1671
    },
    {
      "epoch": 1.7175141242937855,
      "grad_norm": 2.846113681793213,
      "learning_rate": 2.403882755995432e-06,
      "loss": 1.4689,
      "step": 1672
    },
    {
      "epoch": 1.7185413456599896,
      "grad_norm": 2.6862854957580566,
      "learning_rate": 2.4019794442329654e-06,
      "loss": 1.3979,
      "step": 1673
    },
    {
      "epoch": 1.7195685670261942,
      "grad_norm": 2.7214972972869873,
      "learning_rate": 2.4000761324704988e-06,
      "loss": 1.3738,
      "step": 1674
    },
    {
      "epoch": 1.7205957883923986,
      "grad_norm": 2.7037723064422607,
      "learning_rate": 2.398172820708032e-06,
      "loss": 1.4184,
      "step": 1675
    },
    {
      "epoch": 1.721623009758603,
      "grad_norm": 2.7854912281036377,
      "learning_rate": 2.3962695089455655e-06,
      "loss": 1.3854,
      "step": 1676
    },
    {
      "epoch": 1.7226502311248075,
      "grad_norm": 2.640059471130371,
      "learning_rate": 2.3943661971830984e-06,
      "loss": 1.3656,
      "step": 1677
    },
    {
      "epoch": 1.7236774524910117,
      "grad_norm": 2.6955630779266357,
      "learning_rate": 2.3924628854206322e-06,
      "loss": 1.3764,
      "step": 1678
    },
    {
      "epoch": 1.7247046738572163,
      "grad_norm": 3.018975257873535,
      "learning_rate": 2.3905595736581656e-06,
      "loss": 1.3922,
      "step": 1679
    },
    {
      "epoch": 1.7257318952234206,
      "grad_norm": 2.7709803581237793,
      "learning_rate": 2.388656261895699e-06,
      "loss": 1.4288,
      "step": 1680
    },
    {
      "epoch": 1.7257318952234206,
      "eval_loss": 1.8474006652832031,
      "eval_runtime": 37.1233,
      "eval_samples_per_second": 26.937,
      "eval_steps_per_second": 1.131,
      "step": 1680
    },
    {
      "epoch": 1.726759116589625,
      "grad_norm": 2.8898143768310547,
      "learning_rate": 2.386752950133232e-06,
      "loss": 1.4143,
      "step": 1681
    },
    {
      "epoch": 1.7277863379558296,
      "grad_norm": 2.9092485904693604,
      "learning_rate": 2.3848496383707653e-06,
      "loss": 1.4097,
      "step": 1682
    },
    {
      "epoch": 1.7288135593220337,
      "grad_norm": 2.886662483215332,
      "learning_rate": 2.3829463266082986e-06,
      "loss": 1.3968,
      "step": 1683
    },
    {
      "epoch": 1.7298407806882383,
      "grad_norm": 2.7398223876953125,
      "learning_rate": 2.381043014845832e-06,
      "loss": 1.4198,
      "step": 1684
    },
    {
      "epoch": 1.7308680020544427,
      "grad_norm": 2.626732587814331,
      "learning_rate": 2.3791397030833653e-06,
      "loss": 1.4116,
      "step": 1685
    },
    {
      "epoch": 1.731895223420647,
      "grad_norm": 2.6434433460235596,
      "learning_rate": 2.3772363913208983e-06,
      "loss": 1.3857,
      "step": 1686
    },
    {
      "epoch": 1.7329224447868516,
      "grad_norm": 2.83876633644104,
      "learning_rate": 2.3753330795584316e-06,
      "loss": 1.3672,
      "step": 1687
    },
    {
      "epoch": 1.733949666153056,
      "grad_norm": 2.925100564956665,
      "learning_rate": 2.3734297677959654e-06,
      "loss": 1.3957,
      "step": 1688
    },
    {
      "epoch": 1.7349768875192604,
      "grad_norm": 2.833876371383667,
      "learning_rate": 2.3715264560334984e-06,
      "loss": 1.442,
      "step": 1689
    },
    {
      "epoch": 1.736004108885465,
      "grad_norm": 2.8355112075805664,
      "learning_rate": 2.3696231442710317e-06,
      "loss": 1.4153,
      "step": 1690
    },
    {
      "epoch": 1.736004108885465,
      "eval_loss": 1.8604445457458496,
      "eval_runtime": 37.2018,
      "eval_samples_per_second": 26.88,
      "eval_steps_per_second": 1.129,
      "step": 1690
    },
    {
      "epoch": 1.7370313302516691,
      "grad_norm": 2.7500364780426025,
      "learning_rate": 2.367719832508565e-06,
      "loss": 1.4165,
      "step": 1691
    },
    {
      "epoch": 1.7380585516178737,
      "grad_norm": 2.87259578704834,
      "learning_rate": 2.3658165207460985e-06,
      "loss": 1.4431,
      "step": 1692
    },
    {
      "epoch": 1.739085772984078,
      "grad_norm": 2.899538993835449,
      "learning_rate": 2.363913208983632e-06,
      "loss": 1.4134,
      "step": 1693
    },
    {
      "epoch": 1.7401129943502824,
      "grad_norm": 2.799593687057495,
      "learning_rate": 2.362009897221165e-06,
      "loss": 1.4249,
      "step": 1694
    },
    {
      "epoch": 1.741140215716487,
      "grad_norm": 2.8966968059539795,
      "learning_rate": 2.360106585458698e-06,
      "loss": 1.4184,
      "step": 1695
    },
    {
      "epoch": 1.7421674370826912,
      "grad_norm": 2.6944682598114014,
      "learning_rate": 2.3582032736962315e-06,
      "loss": 1.3678,
      "step": 1696
    },
    {
      "epoch": 1.7431946584488958,
      "grad_norm": 2.819845199584961,
      "learning_rate": 2.356299961933765e-06,
      "loss": 1.407,
      "step": 1697
    },
    {
      "epoch": 1.7442218798151001,
      "grad_norm": 2.6873581409454346,
      "learning_rate": 2.354396650171298e-06,
      "loss": 1.3771,
      "step": 1698
    },
    {
      "epoch": 1.7452491011813045,
      "grad_norm": 2.6941304206848145,
      "learning_rate": 2.3524933384088316e-06,
      "loss": 1.3743,
      "step": 1699
    },
    {
      "epoch": 1.746276322547509,
      "grad_norm": 2.925926685333252,
      "learning_rate": 2.350590026646365e-06,
      "loss": 1.4549,
      "step": 1700
    },
    {
      "epoch": 1.746276322547509,
      "eval_loss": 1.8619909286499023,
      "eval_runtime": 37.3963,
      "eval_samples_per_second": 26.741,
      "eval_steps_per_second": 1.123,
      "step": 1700
    },
    {
      "epoch": 1.7473035439137135,
      "grad_norm": 2.7275006771087646,
      "learning_rate": 2.3486867148838983e-06,
      "loss": 1.385,
      "step": 1701
    },
    {
      "epoch": 1.7483307652799178,
      "grad_norm": 2.7096338272094727,
      "learning_rate": 2.3467834031214317e-06,
      "loss": 1.3732,
      "step": 1702
    },
    {
      "epoch": 1.7493579866461222,
      "grad_norm": 2.718862295150757,
      "learning_rate": 2.3448800913589646e-06,
      "loss": 1.3889,
      "step": 1703
    },
    {
      "epoch": 1.7503852080123266,
      "grad_norm": 2.8386173248291016,
      "learning_rate": 2.342976779596498e-06,
      "loss": 1.4083,
      "step": 1704
    },
    {
      "epoch": 1.7514124293785311,
      "grad_norm": 2.868790864944458,
      "learning_rate": 2.3410734678340313e-06,
      "loss": 1.3958,
      "step": 1705
    },
    {
      "epoch": 1.7524396507447355,
      "grad_norm": 2.869119882583618,
      "learning_rate": 2.3391701560715647e-06,
      "loss": 1.4344,
      "step": 1706
    },
    {
      "epoch": 1.7534668721109399,
      "grad_norm": 2.702211380004883,
      "learning_rate": 2.337266844309098e-06,
      "loss": 1.3928,
      "step": 1707
    },
    {
      "epoch": 1.7544940934771445,
      "grad_norm": 2.6727139949798584,
      "learning_rate": 2.3353635325466314e-06,
      "loss": 1.3872,
      "step": 1708
    },
    {
      "epoch": 1.7555213148433486,
      "grad_norm": 2.875605583190918,
      "learning_rate": 2.3334602207841648e-06,
      "loss": 1.4097,
      "step": 1709
    },
    {
      "epoch": 1.7565485362095532,
      "grad_norm": 2.7349722385406494,
      "learning_rate": 2.331556909021698e-06,
      "loss": 1.4041,
      "step": 1710
    },
    {
      "epoch": 1.7565485362095532,
      "eval_loss": 1.8533939123153687,
      "eval_runtime": 36.9293,
      "eval_samples_per_second": 27.079,
      "eval_steps_per_second": 1.137,
      "step": 1710
    },
    {
      "epoch": 1.7575757575757576,
      "grad_norm": 2.7475039958953857,
      "learning_rate": 2.3296535972592315e-06,
      "loss": 1.4339,
      "step": 1711
    },
    {
      "epoch": 1.758602978941962,
      "grad_norm": 2.705641746520996,
      "learning_rate": 2.3277502854967644e-06,
      "loss": 1.3763,
      "step": 1712
    },
    {
      "epoch": 1.7596302003081665,
      "grad_norm": 2.77583646774292,
      "learning_rate": 2.325846973734298e-06,
      "loss": 1.4423,
      "step": 1713
    },
    {
      "epoch": 1.7606574216743707,
      "grad_norm": 2.662750482559204,
      "learning_rate": 2.323943661971831e-06,
      "loss": 1.3864,
      "step": 1714
    },
    {
      "epoch": 1.7616846430405753,
      "grad_norm": 2.769313335418701,
      "learning_rate": 2.3220403502093645e-06,
      "loss": 1.4272,
      "step": 1715
    },
    {
      "epoch": 1.7627118644067796,
      "grad_norm": 2.7852976322174072,
      "learning_rate": 2.320137038446898e-06,
      "loss": 1.4398,
      "step": 1716
    },
    {
      "epoch": 1.763739085772984,
      "grad_norm": 2.7452852725982666,
      "learning_rate": 2.318233726684431e-06,
      "loss": 1.4504,
      "step": 1717
    },
    {
      "epoch": 1.7647663071391886,
      "grad_norm": 2.762258291244507,
      "learning_rate": 2.3163304149219646e-06,
      "loss": 1.4325,
      "step": 1718
    },
    {
      "epoch": 1.765793528505393,
      "grad_norm": 2.7612709999084473,
      "learning_rate": 2.314427103159498e-06,
      "loss": 1.4027,
      "step": 1719
    },
    {
      "epoch": 1.7668207498715973,
      "grad_norm": 2.7290871143341064,
      "learning_rate": 2.312523791397031e-06,
      "loss": 1.3919,
      "step": 1720
    },
    {
      "epoch": 1.7668207498715973,
      "eval_loss": 1.8454608917236328,
      "eval_runtime": 37.0509,
      "eval_samples_per_second": 26.99,
      "eval_steps_per_second": 1.134,
      "step": 1720
    },
    {
      "epoch": 1.767847971237802,
      "grad_norm": 2.5960779190063477,
      "learning_rate": 2.3106204796345643e-06,
      "loss": 1.3632,
      "step": 1721
    },
    {
      "epoch": 1.768875192604006,
      "grad_norm": 2.7832717895507812,
      "learning_rate": 2.3087171678720976e-06,
      "loss": 1.3899,
      "step": 1722
    },
    {
      "epoch": 1.7699024139702106,
      "grad_norm": 2.8227648735046387,
      "learning_rate": 2.306813856109631e-06,
      "loss": 1.4517,
      "step": 1723
    },
    {
      "epoch": 1.770929635336415,
      "grad_norm": 2.880059242248535,
      "learning_rate": 2.3049105443471644e-06,
      "loss": 1.4104,
      "step": 1724
    },
    {
      "epoch": 1.7719568567026194,
      "grad_norm": 2.6653501987457275,
      "learning_rate": 2.3030072325846973e-06,
      "loss": 1.3908,
      "step": 1725
    },
    {
      "epoch": 1.772984078068824,
      "grad_norm": 2.844914197921753,
      "learning_rate": 2.3011039208222307e-06,
      "loss": 1.3781,
      "step": 1726
    },
    {
      "epoch": 1.774011299435028,
      "grad_norm": 2.8133163452148438,
      "learning_rate": 2.299200609059764e-06,
      "loss": 1.3775,
      "step": 1727
    },
    {
      "epoch": 1.7750385208012327,
      "grad_norm": 2.7466955184936523,
      "learning_rate": 2.297297297297298e-06,
      "loss": 1.3926,
      "step": 1728
    },
    {
      "epoch": 1.776065742167437,
      "grad_norm": 2.6311862468719482,
      "learning_rate": 2.2953939855348308e-06,
      "loss": 1.3486,
      "step": 1729
    },
    {
      "epoch": 1.7770929635336414,
      "grad_norm": 2.7905211448669434,
      "learning_rate": 2.293490673772364e-06,
      "loss": 1.4369,
      "step": 1730
    },
    {
      "epoch": 1.7770929635336414,
      "eval_loss": 1.8418751955032349,
      "eval_runtime": 37.0448,
      "eval_samples_per_second": 26.994,
      "eval_steps_per_second": 1.134,
      "step": 1730
    },
    {
      "epoch": 1.778120184899846,
      "grad_norm": 2.862295627593994,
      "learning_rate": 2.2915873620098975e-06,
      "loss": 1.4209,
      "step": 1731
    },
    {
      "epoch": 1.7791474062660502,
      "grad_norm": 2.765862464904785,
      "learning_rate": 2.289684050247431e-06,
      "loss": 1.4135,
      "step": 1732
    },
    {
      "epoch": 1.7801746276322548,
      "grad_norm": 2.7296290397644043,
      "learning_rate": 2.287780738484964e-06,
      "loss": 1.3814,
      "step": 1733
    },
    {
      "epoch": 1.7812018489984591,
      "grad_norm": 2.7937533855438232,
      "learning_rate": 2.285877426722497e-06,
      "loss": 1.4203,
      "step": 1734
    },
    {
      "epoch": 1.7822290703646635,
      "grad_norm": 2.7680230140686035,
      "learning_rate": 2.2839741149600305e-06,
      "loss": 1.4354,
      "step": 1735
    },
    {
      "epoch": 1.783256291730868,
      "grad_norm": 2.8785109519958496,
      "learning_rate": 2.282070803197564e-06,
      "loss": 1.4338,
      "step": 1736
    },
    {
      "epoch": 1.7842835130970724,
      "grad_norm": 2.747436046600342,
      "learning_rate": 2.2801674914350972e-06,
      "loss": 1.4087,
      "step": 1737
    },
    {
      "epoch": 1.7853107344632768,
      "grad_norm": 2.6810362339019775,
      "learning_rate": 2.2782641796726306e-06,
      "loss": 1.3993,
      "step": 1738
    },
    {
      "epoch": 1.7863379558294814,
      "grad_norm": 2.7365734577178955,
      "learning_rate": 2.276360867910164e-06,
      "loss": 1.4235,
      "step": 1739
    },
    {
      "epoch": 1.7873651771956856,
      "grad_norm": 2.6946303844451904,
      "learning_rate": 2.2744575561476973e-06,
      "loss": 1.3626,
      "step": 1740
    },
    {
      "epoch": 1.7873651771956856,
      "eval_loss": 1.8456190824508667,
      "eval_runtime": 37.069,
      "eval_samples_per_second": 26.977,
      "eval_steps_per_second": 1.133,
      "step": 1740
    },
    {
      "epoch": 1.7883923985618901,
      "grad_norm": 2.756920337677002,
      "learning_rate": 2.2725542443852307e-06,
      "loss": 1.396,
      "step": 1741
    },
    {
      "epoch": 1.7894196199280945,
      "grad_norm": 2.765428066253662,
      "learning_rate": 2.2706509326227636e-06,
      "loss": 1.4374,
      "step": 1742
    },
    {
      "epoch": 1.7904468412942989,
      "grad_norm": 2.8224971294403076,
      "learning_rate": 2.268747620860297e-06,
      "loss": 1.4113,
      "step": 1743
    },
    {
      "epoch": 1.7914740626605035,
      "grad_norm": 2.90537428855896,
      "learning_rate": 2.2668443090978303e-06,
      "loss": 1.4124,
      "step": 1744
    },
    {
      "epoch": 1.7925012840267076,
      "grad_norm": 2.712862491607666,
      "learning_rate": 2.2649409973353637e-06,
      "loss": 1.4256,
      "step": 1745
    },
    {
      "epoch": 1.7935285053929122,
      "grad_norm": 2.688558340072632,
      "learning_rate": 2.263037685572897e-06,
      "loss": 1.3842,
      "step": 1746
    },
    {
      "epoch": 1.7945557267591166,
      "grad_norm": 2.6504769325256348,
      "learning_rate": 2.26113437381043e-06,
      "loss": 1.3703,
      "step": 1747
    },
    {
      "epoch": 1.795582948125321,
      "grad_norm": 3.0093724727630615,
      "learning_rate": 2.259231062047964e-06,
      "loss": 1.4106,
      "step": 1748
    },
    {
      "epoch": 1.7966101694915255,
      "grad_norm": 2.727687120437622,
      "learning_rate": 2.257327750285497e-06,
      "loss": 1.4286,
      "step": 1749
    },
    {
      "epoch": 1.7976373908577299,
      "grad_norm": 2.679896116256714,
      "learning_rate": 2.2554244385230305e-06,
      "loss": 1.3937,
      "step": 1750
    },
    {
      "epoch": 1.7976373908577299,
      "eval_loss": 1.8520207405090332,
      "eval_runtime": 37.062,
      "eval_samples_per_second": 26.982,
      "eval_steps_per_second": 1.133,
      "step": 1750
    },
    {
      "epoch": 1.7986646122239343,
      "grad_norm": 2.8119945526123047,
      "learning_rate": 2.2535211267605635e-06,
      "loss": 1.3935,
      "step": 1751
    },
    {
      "epoch": 1.7996918335901386,
      "grad_norm": 2.8470423221588135,
      "learning_rate": 2.251617814998097e-06,
      "loss": 1.4138,
      "step": 1752
    },
    {
      "epoch": 1.800719054956343,
      "grad_norm": 2.6278114318847656,
      "learning_rate": 2.24971450323563e-06,
      "loss": 1.328,
      "step": 1753
    },
    {
      "epoch": 1.8017462763225476,
      "grad_norm": 2.861860990524292,
      "learning_rate": 2.2478111914731636e-06,
      "loss": 1.416,
      "step": 1754
    },
    {
      "epoch": 1.802773497688752,
      "grad_norm": 2.559736728668213,
      "learning_rate": 2.245907879710697e-06,
      "loss": 1.2891,
      "step": 1755
    },
    {
      "epoch": 1.8038007190549563,
      "grad_norm": 2.8461058139801025,
      "learning_rate": 2.24400456794823e-06,
      "loss": 1.4194,
      "step": 1756
    },
    {
      "epoch": 1.804827940421161,
      "grad_norm": 2.7069878578186035,
      "learning_rate": 2.2421012561857632e-06,
      "loss": 1.382,
      "step": 1757
    },
    {
      "epoch": 1.805855161787365,
      "grad_norm": 2.727435827255249,
      "learning_rate": 2.240197944423297e-06,
      "loss": 1.3714,
      "step": 1758
    },
    {
      "epoch": 1.8068823831535696,
      "grad_norm": 2.7298061847686768,
      "learning_rate": 2.23829463266083e-06,
      "loss": 1.3834,
      "step": 1759
    },
    {
      "epoch": 1.807909604519774,
      "grad_norm": 2.7419843673706055,
      "learning_rate": 2.2363913208983633e-06,
      "loss": 1.3687,
      "step": 1760
    },
    {
      "epoch": 1.807909604519774,
      "eval_loss": 1.8575356006622314,
      "eval_runtime": 37.0563,
      "eval_samples_per_second": 26.986,
      "eval_steps_per_second": 1.133,
      "step": 1760
    },
    {
      "epoch": 1.8089368258859784,
      "grad_norm": 2.8058817386627197,
      "learning_rate": 2.2344880091358967e-06,
      "loss": 1.415,
      "step": 1761
    },
    {
      "epoch": 1.809964047252183,
      "grad_norm": 2.7287003993988037,
      "learning_rate": 2.23258469737343e-06,
      "loss": 1.3778,
      "step": 1762
    },
    {
      "epoch": 1.810991268618387,
      "grad_norm": 2.7943124771118164,
      "learning_rate": 2.2306813856109634e-06,
      "loss": 1.433,
      "step": 1763
    },
    {
      "epoch": 1.8120184899845917,
      "grad_norm": 2.7254626750946045,
      "learning_rate": 2.2287780738484963e-06,
      "loss": 1.3822,
      "step": 1764
    },
    {
      "epoch": 1.813045711350796,
      "grad_norm": 2.799793004989624,
      "learning_rate": 2.2268747620860297e-06,
      "loss": 1.4595,
      "step": 1765
    },
    {
      "epoch": 1.8140729327170004,
      "grad_norm": 2.79420804977417,
      "learning_rate": 2.224971450323563e-06,
      "loss": 1.3949,
      "step": 1766
    },
    {
      "epoch": 1.815100154083205,
      "grad_norm": 2.861016035079956,
      "learning_rate": 2.2230681385610964e-06,
      "loss": 1.3301,
      "step": 1767
    },
    {
      "epoch": 1.8161273754494094,
      "grad_norm": 2.9270451068878174,
      "learning_rate": 2.2211648267986298e-06,
      "loss": 1.3895,
      "step": 1768
    },
    {
      "epoch": 1.8171545968156138,
      "grad_norm": 2.709031105041504,
      "learning_rate": 2.219261515036163e-06,
      "loss": 1.3757,
      "step": 1769
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.7763030529022217,
      "learning_rate": 2.2173582032736965e-06,
      "loss": 1.4135,
      "step": 1770
    },
    {
      "epoch": 1.8181818181818183,
      "eval_loss": 1.8521159887313843,
      "eval_runtime": 37.1265,
      "eval_samples_per_second": 26.935,
      "eval_steps_per_second": 1.131,
      "step": 1770
    },
    {
      "epoch": 1.8192090395480225,
      "grad_norm": 2.97048282623291,
      "learning_rate": 2.21545489151123e-06,
      "loss": 1.4345,
      "step": 1771
    },
    {
      "epoch": 1.820236260914227,
      "grad_norm": 2.8238744735717773,
      "learning_rate": 2.2135515797487632e-06,
      "loss": 1.3948,
      "step": 1772
    },
    {
      "epoch": 1.8212634822804314,
      "grad_norm": 2.6430611610412598,
      "learning_rate": 2.211648267986296e-06,
      "loss": 1.3878,
      "step": 1773
    },
    {
      "epoch": 1.8222907036466358,
      "grad_norm": 2.602883815765381,
      "learning_rate": 2.2097449562238295e-06,
      "loss": 1.3978,
      "step": 1774
    },
    {
      "epoch": 1.8233179250128404,
      "grad_norm": 2.707813262939453,
      "learning_rate": 2.207841644461363e-06,
      "loss": 1.3937,
      "step": 1775
    },
    {
      "epoch": 1.8243451463790445,
      "grad_norm": 2.662684679031372,
      "learning_rate": 2.2059383326988963e-06,
      "loss": 1.4381,
      "step": 1776
    },
    {
      "epoch": 1.8253723677452491,
      "grad_norm": 2.9407193660736084,
      "learning_rate": 2.2040350209364296e-06,
      "loss": 1.4059,
      "step": 1777
    },
    {
      "epoch": 1.8263995891114535,
      "grad_norm": 2.6678171157836914,
      "learning_rate": 2.2021317091739626e-06,
      "loss": 1.3718,
      "step": 1778
    },
    {
      "epoch": 1.8274268104776579,
      "grad_norm": 2.7536516189575195,
      "learning_rate": 2.2002283974114963e-06,
      "loss": 1.3998,
      "step": 1779
    },
    {
      "epoch": 1.8284540318438625,
      "grad_norm": 2.7340128421783447,
      "learning_rate": 2.1983250856490297e-06,
      "loss": 1.3299,
      "step": 1780
    },
    {
      "epoch": 1.8284540318438625,
      "eval_loss": 1.854264736175537,
      "eval_runtime": 37.0874,
      "eval_samples_per_second": 26.963,
      "eval_steps_per_second": 1.132,
      "step": 1780
    },
    {
      "epoch": 1.8294812532100666,
      "grad_norm": 2.844285726547241,
      "learning_rate": 2.1964217738865626e-06,
      "loss": 1.3746,
      "step": 1781
    },
    {
      "epoch": 1.8305084745762712,
      "grad_norm": 2.9005022048950195,
      "learning_rate": 2.194518462124096e-06,
      "loss": 1.4087,
      "step": 1782
    },
    {
      "epoch": 1.8315356959424756,
      "grad_norm": 2.8825247287750244,
      "learning_rate": 2.1926151503616294e-06,
      "loss": 1.4458,
      "step": 1783
    },
    {
      "epoch": 1.83256291730868,
      "grad_norm": 2.712778091430664,
      "learning_rate": 2.1907118385991627e-06,
      "loss": 1.414,
      "step": 1784
    },
    {
      "epoch": 1.8335901386748845,
      "grad_norm": 2.7036001682281494,
      "learning_rate": 2.188808526836696e-06,
      "loss": 1.3632,
      "step": 1785
    },
    {
      "epoch": 1.8346173600410889,
      "grad_norm": 2.704606771469116,
      "learning_rate": 2.1869052150742295e-06,
      "loss": 1.3992,
      "step": 1786
    },
    {
      "epoch": 1.8356445814072933,
      "grad_norm": 2.7564282417297363,
      "learning_rate": 2.1850019033117624e-06,
      "loss": 1.3331,
      "step": 1787
    },
    {
      "epoch": 1.8366718027734978,
      "grad_norm": 2.898256540298462,
      "learning_rate": 2.1830985915492958e-06,
      "loss": 1.3881,
      "step": 1788
    },
    {
      "epoch": 1.837699024139702,
      "grad_norm": 2.7897069454193115,
      "learning_rate": 2.1811952797868295e-06,
      "loss": 1.4081,
      "step": 1789
    },
    {
      "epoch": 1.8387262455059066,
      "grad_norm": 2.7293193340301514,
      "learning_rate": 2.1792919680243625e-06,
      "loss": 1.3705,
      "step": 1790
    },
    {
      "epoch": 1.8387262455059066,
      "eval_loss": 1.8483723402023315,
      "eval_runtime": 37.4058,
      "eval_samples_per_second": 26.734,
      "eval_steps_per_second": 1.123,
      "step": 1790
    },
    {
      "epoch": 1.839753466872111,
      "grad_norm": 2.9257752895355225,
      "learning_rate": 2.177388656261896e-06,
      "loss": 1.4071,
      "step": 1791
    },
    {
      "epoch": 1.8407806882383153,
      "grad_norm": 3.016033172607422,
      "learning_rate": 2.1754853444994292e-06,
      "loss": 1.4774,
      "step": 1792
    },
    {
      "epoch": 1.84180790960452,
      "grad_norm": 2.74198842048645,
      "learning_rate": 2.1735820327369626e-06,
      "loss": 1.301,
      "step": 1793
    },
    {
      "epoch": 1.842835130970724,
      "grad_norm": 2.809215545654297,
      "learning_rate": 2.171678720974496e-06,
      "loss": 1.4192,
      "step": 1794
    },
    {
      "epoch": 1.8438623523369286,
      "grad_norm": 2.673997402191162,
      "learning_rate": 2.169775409212029e-06,
      "loss": 1.4182,
      "step": 1795
    },
    {
      "epoch": 1.844889573703133,
      "grad_norm": 2.9519944190979004,
      "learning_rate": 2.1678720974495622e-06,
      "loss": 1.4253,
      "step": 1796
    },
    {
      "epoch": 1.8459167950693374,
      "grad_norm": 2.7683424949645996,
      "learning_rate": 2.1659687856870956e-06,
      "loss": 1.4162,
      "step": 1797
    },
    {
      "epoch": 1.846944016435542,
      "grad_norm": 2.861701726913452,
      "learning_rate": 2.164065473924629e-06,
      "loss": 1.4167,
      "step": 1798
    },
    {
      "epoch": 1.8479712378017463,
      "grad_norm": 2.62581467628479,
      "learning_rate": 2.1621621621621623e-06,
      "loss": 1.3568,
      "step": 1799
    },
    {
      "epoch": 1.8489984591679507,
      "grad_norm": 2.782197952270508,
      "learning_rate": 2.1602588503996957e-06,
      "loss": 1.4184,
      "step": 1800
    },
    {
      "epoch": 1.8489984591679507,
      "eval_loss": 1.8527827262878418,
      "eval_runtime": 37.4743,
      "eval_samples_per_second": 26.685,
      "eval_steps_per_second": 1.121,
      "step": 1800
    },
    {
      "epoch": 1.850025680534155,
      "grad_norm": 2.7289304733276367,
      "learning_rate": 2.158355538637229e-06,
      "loss": 1.3979,
      "step": 1801
    },
    {
      "epoch": 1.8510529019003594,
      "grad_norm": 2.7153542041778564,
      "learning_rate": 2.1564522268747624e-06,
      "loss": 1.3863,
      "step": 1802
    },
    {
      "epoch": 1.852080123266564,
      "grad_norm": 2.884798765182495,
      "learning_rate": 2.1545489151122958e-06,
      "loss": 1.4656,
      "step": 1803
    },
    {
      "epoch": 1.8531073446327684,
      "grad_norm": 2.6483845710754395,
      "learning_rate": 2.1526456033498287e-06,
      "loss": 1.4118,
      "step": 1804
    },
    {
      "epoch": 1.8541345659989728,
      "grad_norm": 2.7074198722839355,
      "learning_rate": 2.150742291587362e-06,
      "loss": 1.41,
      "step": 1805
    },
    {
      "epoch": 1.8551617873651773,
      "grad_norm": 2.734137535095215,
      "learning_rate": 2.1488389798248954e-06,
      "loss": 1.3968,
      "step": 1806
    },
    {
      "epoch": 1.8561890087313815,
      "grad_norm": 2.7855992317199707,
      "learning_rate": 2.146935668062429e-06,
      "loss": 1.427,
      "step": 1807
    },
    {
      "epoch": 1.857216230097586,
      "grad_norm": 2.7769339084625244,
      "learning_rate": 2.145032356299962e-06,
      "loss": 1.3476,
      "step": 1808
    },
    {
      "epoch": 1.8582434514637904,
      "grad_norm": 2.6978964805603027,
      "learning_rate": 2.1431290445374955e-06,
      "loss": 1.4047,
      "step": 1809
    },
    {
      "epoch": 1.8592706728299948,
      "grad_norm": 2.806438446044922,
      "learning_rate": 2.141225732775029e-06,
      "loss": 1.3693,
      "step": 1810
    },
    {
      "epoch": 1.8592706728299948,
      "eval_loss": 1.8606462478637695,
      "eval_runtime": 36.9431,
      "eval_samples_per_second": 27.069,
      "eval_steps_per_second": 1.137,
      "step": 1810
    },
    {
      "epoch": 1.8602978941961994,
      "grad_norm": 2.886394500732422,
      "learning_rate": 2.1393224210125623e-06,
      "loss": 1.4022,
      "step": 1811
    },
    {
      "epoch": 1.8613251155624035,
      "grad_norm": 2.7849435806274414,
      "learning_rate": 2.137419109250095e-06,
      "loss": 1.3907,
      "step": 1812
    },
    {
      "epoch": 1.8623523369286081,
      "grad_norm": 2.85810923576355,
      "learning_rate": 2.1355157974876286e-06,
      "loss": 1.4451,
      "step": 1813
    },
    {
      "epoch": 1.8633795582948125,
      "grad_norm": 2.8717806339263916,
      "learning_rate": 2.133612485725162e-06,
      "loss": 1.4303,
      "step": 1814
    },
    {
      "epoch": 1.8644067796610169,
      "grad_norm": 2.6955783367156982,
      "learning_rate": 2.1317091739626953e-06,
      "loss": 1.3454,
      "step": 1815
    },
    {
      "epoch": 1.8654340010272215,
      "grad_norm": 2.835347890853882,
      "learning_rate": 2.1298058622002286e-06,
      "loss": 1.3671,
      "step": 1816
    },
    {
      "epoch": 1.8664612223934258,
      "grad_norm": 2.804835796356201,
      "learning_rate": 2.1279025504377616e-06,
      "loss": 1.388,
      "step": 1817
    },
    {
      "epoch": 1.8674884437596302,
      "grad_norm": 2.69130539894104,
      "learning_rate": 2.125999238675295e-06,
      "loss": 1.3815,
      "step": 1818
    },
    {
      "epoch": 1.8685156651258348,
      "grad_norm": 2.856457233428955,
      "learning_rate": 2.1240959269128287e-06,
      "loss": 1.4323,
      "step": 1819
    },
    {
      "epoch": 1.869542886492039,
      "grad_norm": 2.8100039958953857,
      "learning_rate": 2.122192615150362e-06,
      "loss": 1.3942,
      "step": 1820
    },
    {
      "epoch": 1.869542886492039,
      "eval_loss": 1.860397219657898,
      "eval_runtime": 37.0473,
      "eval_samples_per_second": 26.993,
      "eval_steps_per_second": 1.134,
      "step": 1820
    },
    {
      "epoch": 1.8705701078582435,
      "grad_norm": 2.6977529525756836,
      "learning_rate": 2.120289303387895e-06,
      "loss": 1.3371,
      "step": 1821
    },
    {
      "epoch": 1.8715973292244479,
      "grad_norm": 2.8403844833374023,
      "learning_rate": 2.1183859916254284e-06,
      "loss": 1.3776,
      "step": 1822
    },
    {
      "epoch": 1.8726245505906522,
      "grad_norm": 2.829756498336792,
      "learning_rate": 2.1164826798629618e-06,
      "loss": 1.4229,
      "step": 1823
    },
    {
      "epoch": 1.8736517719568568,
      "grad_norm": 2.6810450553894043,
      "learning_rate": 2.114579368100495e-06,
      "loss": 1.3757,
      "step": 1824
    },
    {
      "epoch": 1.874678993323061,
      "grad_norm": 2.988982677459717,
      "learning_rate": 2.1126760563380285e-06,
      "loss": 1.3671,
      "step": 1825
    },
    {
      "epoch": 1.8757062146892656,
      "grad_norm": 2.661953926086426,
      "learning_rate": 2.1107727445755614e-06,
      "loss": 1.3382,
      "step": 1826
    },
    {
      "epoch": 1.87673343605547,
      "grad_norm": 2.7993199825286865,
      "learning_rate": 2.1088694328130948e-06,
      "loss": 1.4621,
      "step": 1827
    },
    {
      "epoch": 1.8777606574216743,
      "grad_norm": 2.681609869003296,
      "learning_rate": 2.106966121050628e-06,
      "loss": 1.3596,
      "step": 1828
    },
    {
      "epoch": 1.878787878787879,
      "grad_norm": 2.988734722137451,
      "learning_rate": 2.1050628092881615e-06,
      "loss": 1.4269,
      "step": 1829
    },
    {
      "epoch": 1.879815100154083,
      "grad_norm": 2.867772102355957,
      "learning_rate": 2.103159497525695e-06,
      "loss": 1.4618,
      "step": 1830
    },
    {
      "epoch": 1.879815100154083,
      "eval_loss": 1.8595913648605347,
      "eval_runtime": 37.1186,
      "eval_samples_per_second": 26.941,
      "eval_steps_per_second": 1.132,
      "step": 1830
    },
    {
      "epoch": 1.8808423215202876,
      "grad_norm": 2.7212493419647217,
      "learning_rate": 2.1012561857632282e-06,
      "loss": 1.3738,
      "step": 1831
    },
    {
      "epoch": 1.881869542886492,
      "grad_norm": 2.7997097969055176,
      "learning_rate": 2.0993528740007616e-06,
      "loss": 1.4268,
      "step": 1832
    },
    {
      "epoch": 1.8828967642526964,
      "grad_norm": 2.8695220947265625,
      "learning_rate": 2.097449562238295e-06,
      "loss": 1.4397,
      "step": 1833
    },
    {
      "epoch": 1.883923985618901,
      "grad_norm": 2.681015729904175,
      "learning_rate": 2.095546250475828e-06,
      "loss": 1.3873,
      "step": 1834
    },
    {
      "epoch": 1.8849512069851053,
      "grad_norm": 2.787360429763794,
      "learning_rate": 2.0936429387133613e-06,
      "loss": 1.3944,
      "step": 1835
    },
    {
      "epoch": 1.8859784283513097,
      "grad_norm": 2.830164670944214,
      "learning_rate": 2.0917396269508946e-06,
      "loss": 1.4485,
      "step": 1836
    },
    {
      "epoch": 1.8870056497175143,
      "grad_norm": 2.9803659915924072,
      "learning_rate": 2.089836315188428e-06,
      "loss": 1.4259,
      "step": 1837
    },
    {
      "epoch": 1.8880328710837184,
      "grad_norm": 2.846390962600708,
      "learning_rate": 2.0879330034259614e-06,
      "loss": 1.4371,
      "step": 1838
    },
    {
      "epoch": 1.889060092449923,
      "grad_norm": 2.636085033416748,
      "learning_rate": 2.0860296916634947e-06,
      "loss": 1.3679,
      "step": 1839
    },
    {
      "epoch": 1.8900873138161274,
      "grad_norm": 2.806547164916992,
      "learning_rate": 2.084126379901028e-06,
      "loss": 1.4662,
      "step": 1840
    },
    {
      "epoch": 1.8900873138161274,
      "eval_loss": 1.8502073287963867,
      "eval_runtime": 37.1321,
      "eval_samples_per_second": 26.931,
      "eval_steps_per_second": 1.131,
      "step": 1840
    },
    {
      "epoch": 1.8911145351823317,
      "grad_norm": 2.8658089637756348,
      "learning_rate": 2.0822230681385614e-06,
      "loss": 1.3593,
      "step": 1841
    },
    {
      "epoch": 1.8921417565485363,
      "grad_norm": 2.87060284614563,
      "learning_rate": 2.080319756376095e-06,
      "loss": 1.4189,
      "step": 1842
    },
    {
      "epoch": 1.8931689779147405,
      "grad_norm": 2.7965638637542725,
      "learning_rate": 2.0784164446136277e-06,
      "loss": 1.4208,
      "step": 1843
    },
    {
      "epoch": 1.894196199280945,
      "grad_norm": 2.5999045372009277,
      "learning_rate": 2.076513132851161e-06,
      "loss": 1.375,
      "step": 1844
    },
    {
      "epoch": 1.8952234206471494,
      "grad_norm": 2.833935260772705,
      "learning_rate": 2.0746098210886945e-06,
      "loss": 1.4137,
      "step": 1845
    },
    {
      "epoch": 1.8962506420133538,
      "grad_norm": 2.9809396266937256,
      "learning_rate": 2.072706509326228e-06,
      "loss": 1.3874,
      "step": 1846
    },
    {
      "epoch": 1.8972778633795584,
      "grad_norm": 2.9341282844543457,
      "learning_rate": 2.070803197563761e-06,
      "loss": 1.3953,
      "step": 1847
    },
    {
      "epoch": 1.8983050847457628,
      "grad_norm": 2.989488124847412,
      "learning_rate": 2.068899885801294e-06,
      "loss": 1.3913,
      "step": 1848
    },
    {
      "epoch": 1.8993323061119671,
      "grad_norm": 2.8224217891693115,
      "learning_rate": 2.066996574038828e-06,
      "loss": 1.4032,
      "step": 1849
    },
    {
      "epoch": 1.9003595274781715,
      "grad_norm": 2.8850255012512207,
      "learning_rate": 2.0650932622763613e-06,
      "loss": 1.4402,
      "step": 1850
    },
    {
      "epoch": 1.9003595274781715,
      "eval_loss": 1.8440051078796387,
      "eval_runtime": 37.1532,
      "eval_samples_per_second": 26.916,
      "eval_steps_per_second": 1.13,
      "step": 1850
    },
    {
      "epoch": 1.9013867488443759,
      "grad_norm": 2.927288293838501,
      "learning_rate": 2.0631899505138942e-06,
      "loss": 1.3215,
      "step": 1851
    },
    {
      "epoch": 1.9024139702105805,
      "grad_norm": 2.7680368423461914,
      "learning_rate": 2.0612866387514276e-06,
      "loss": 1.3636,
      "step": 1852
    },
    {
      "epoch": 1.9034411915767848,
      "grad_norm": 2.8515689373016357,
      "learning_rate": 2.059383326988961e-06,
      "loss": 1.4049,
      "step": 1853
    },
    {
      "epoch": 1.9044684129429892,
      "grad_norm": 2.7973034381866455,
      "learning_rate": 2.0574800152264943e-06,
      "loss": 1.3876,
      "step": 1854
    },
    {
      "epoch": 1.9054956343091938,
      "grad_norm": 2.7860023975372314,
      "learning_rate": 2.0555767034640277e-06,
      "loss": 1.3687,
      "step": 1855
    },
    {
      "epoch": 1.906522855675398,
      "grad_norm": 2.807511568069458,
      "learning_rate": 2.053673391701561e-06,
      "loss": 1.3382,
      "step": 1856
    },
    {
      "epoch": 1.9075500770416025,
      "grad_norm": 2.9342591762542725,
      "learning_rate": 2.051770079939094e-06,
      "loss": 1.4025,
      "step": 1857
    },
    {
      "epoch": 1.9085772984078069,
      "grad_norm": 2.85943865776062,
      "learning_rate": 2.0498667681766273e-06,
      "loss": 1.3626,
      "step": 1858
    },
    {
      "epoch": 1.9096045197740112,
      "grad_norm": 2.8514609336853027,
      "learning_rate": 2.047963456414161e-06,
      "loss": 1.3675,
      "step": 1859
    },
    {
      "epoch": 1.9106317411402158,
      "grad_norm": 2.8177332878112793,
      "learning_rate": 2.046060144651694e-06,
      "loss": 1.3968,
      "step": 1860
    },
    {
      "epoch": 1.9106317411402158,
      "eval_loss": 1.8451731204986572,
      "eval_runtime": 37.1047,
      "eval_samples_per_second": 26.951,
      "eval_steps_per_second": 1.132,
      "step": 1860
    },
    {
      "epoch": 1.91165896250642,
      "grad_norm": 2.74876070022583,
      "learning_rate": 2.0441568328892274e-06,
      "loss": 1.4022,
      "step": 1861
    },
    {
      "epoch": 1.9126861838726246,
      "grad_norm": 2.6633286476135254,
      "learning_rate": 2.0422535211267608e-06,
      "loss": 1.4004,
      "step": 1862
    },
    {
      "epoch": 1.913713405238829,
      "grad_norm": 2.6457369327545166,
      "learning_rate": 2.040350209364294e-06,
      "loss": 1.3767,
      "step": 1863
    },
    {
      "epoch": 1.9147406266050333,
      "grad_norm": 2.7011449337005615,
      "learning_rate": 2.0384468976018275e-06,
      "loss": 1.373,
      "step": 1864
    },
    {
      "epoch": 1.915767847971238,
      "grad_norm": 2.676337480545044,
      "learning_rate": 2.0365435858393604e-06,
      "loss": 1.3713,
      "step": 1865
    },
    {
      "epoch": 1.9167950693374423,
      "grad_norm": 2.7401342391967773,
      "learning_rate": 2.034640274076894e-06,
      "loss": 1.3644,
      "step": 1866
    },
    {
      "epoch": 1.9178222907036466,
      "grad_norm": 2.7799527645111084,
      "learning_rate": 2.032736962314427e-06,
      "loss": 1.433,
      "step": 1867
    },
    {
      "epoch": 1.9188495120698512,
      "grad_norm": 2.7372307777404785,
      "learning_rate": 2.0308336505519605e-06,
      "loss": 1.3781,
      "step": 1868
    },
    {
      "epoch": 1.9198767334360554,
      "grad_norm": 2.986297845840454,
      "learning_rate": 2.028930338789494e-06,
      "loss": 1.4406,
      "step": 1869
    },
    {
      "epoch": 1.92090395480226,
      "grad_norm": 2.815706491470337,
      "learning_rate": 2.0270270270270273e-06,
      "loss": 1.3555,
      "step": 1870
    },
    {
      "epoch": 1.92090395480226,
      "eval_loss": 1.8498139381408691,
      "eval_runtime": 37.1146,
      "eval_samples_per_second": 26.944,
      "eval_steps_per_second": 1.132,
      "step": 1870
    },
    {
      "epoch": 1.9219311761684643,
      "grad_norm": 2.734189987182617,
      "learning_rate": 2.0251237152645606e-06,
      "loss": 1.373,
      "step": 1871
    },
    {
      "epoch": 1.9229583975346687,
      "grad_norm": 2.7900359630584717,
      "learning_rate": 2.023220403502094e-06,
      "loss": 1.3623,
      "step": 1872
    },
    {
      "epoch": 1.9239856189008733,
      "grad_norm": 2.8590710163116455,
      "learning_rate": 2.0213170917396273e-06,
      "loss": 1.3424,
      "step": 1873
    },
    {
      "epoch": 1.9250128402670774,
      "grad_norm": 2.880814790725708,
      "learning_rate": 2.0194137799771603e-06,
      "loss": 1.3683,
      "step": 1874
    },
    {
      "epoch": 1.926040061633282,
      "grad_norm": 2.7775630950927734,
      "learning_rate": 2.0175104682146937e-06,
      "loss": 1.3879,
      "step": 1875
    },
    {
      "epoch": 1.9270672829994864,
      "grad_norm": 2.713557481765747,
      "learning_rate": 2.015607156452227e-06,
      "loss": 1.3216,
      "step": 1876
    },
    {
      "epoch": 1.9280945043656907,
      "grad_norm": 2.8787734508514404,
      "learning_rate": 2.0137038446897604e-06,
      "loss": 1.3659,
      "step": 1877
    },
    {
      "epoch": 1.9291217257318953,
      "grad_norm": 2.8224005699157715,
      "learning_rate": 2.0118005329272937e-06,
      "loss": 1.3572,
      "step": 1878
    },
    {
      "epoch": 1.9301489470980995,
      "grad_norm": 2.865316390991211,
      "learning_rate": 2.0098972211648267e-06,
      "loss": 1.3817,
      "step": 1879
    },
    {
      "epoch": 1.931176168464304,
      "grad_norm": 2.7488853931427,
      "learning_rate": 2.0079939094023605e-06,
      "loss": 1.3766,
      "step": 1880
    },
    {
      "epoch": 1.931176168464304,
      "eval_loss": 1.8444081544876099,
      "eval_runtime": 37.1079,
      "eval_samples_per_second": 26.948,
      "eval_steps_per_second": 1.132,
      "step": 1880
    },
    {
      "epoch": 1.9322033898305084,
      "grad_norm": 2.902949333190918,
      "learning_rate": 2.006090597639894e-06,
      "loss": 1.4183,
      "step": 1881
    },
    {
      "epoch": 1.9332306111967128,
      "grad_norm": 2.8154122829437256,
      "learning_rate": 2.0041872858774268e-06,
      "loss": 1.4525,
      "step": 1882
    },
    {
      "epoch": 1.9342578325629174,
      "grad_norm": 2.8343429565429688,
      "learning_rate": 2.00228397411496e-06,
      "loss": 1.4099,
      "step": 1883
    },
    {
      "epoch": 1.9352850539291218,
      "grad_norm": 2.863067388534546,
      "learning_rate": 2.0003806623524935e-06,
      "loss": 1.3973,
      "step": 1884
    },
    {
      "epoch": 1.9363122752953261,
      "grad_norm": 2.784085988998413,
      "learning_rate": 1.998477350590027e-06,
      "loss": 1.3117,
      "step": 1885
    },
    {
      "epoch": 1.9373394966615307,
      "grad_norm": 2.7667958736419678,
      "learning_rate": 1.9965740388275602e-06,
      "loss": 1.4056,
      "step": 1886
    },
    {
      "epoch": 1.9383667180277349,
      "grad_norm": 2.747432231903076,
      "learning_rate": 1.994670727065093e-06,
      "loss": 1.3927,
      "step": 1887
    },
    {
      "epoch": 1.9393939393939394,
      "grad_norm": 2.9162001609802246,
      "learning_rate": 1.9927674153026265e-06,
      "loss": 1.3453,
      "step": 1888
    },
    {
      "epoch": 1.9404211607601438,
      "grad_norm": 3.0909414291381836,
      "learning_rate": 1.99086410354016e-06,
      "loss": 1.4148,
      "step": 1889
    },
    {
      "epoch": 1.9414483821263482,
      "grad_norm": 2.774672746658325,
      "learning_rate": 1.9889607917776937e-06,
      "loss": 1.3762,
      "step": 1890
    },
    {
      "epoch": 1.9414483821263482,
      "eval_loss": 1.848209023475647,
      "eval_runtime": 37.2784,
      "eval_samples_per_second": 26.825,
      "eval_steps_per_second": 1.127,
      "step": 1890
    },
    {
      "epoch": 1.9424756034925528,
      "grad_norm": 2.7985880374908447,
      "learning_rate": 1.9870574800152266e-06,
      "loss": 1.3908,
      "step": 1891
    },
    {
      "epoch": 1.943502824858757,
      "grad_norm": 2.749664783477783,
      "learning_rate": 1.98515416825276e-06,
      "loss": 1.4088,
      "step": 1892
    },
    {
      "epoch": 1.9445300462249615,
      "grad_norm": 2.770726442337036,
      "learning_rate": 1.9832508564902933e-06,
      "loss": 1.4054,
      "step": 1893
    },
    {
      "epoch": 1.9455572675911659,
      "grad_norm": 2.634516477584839,
      "learning_rate": 1.9813475447278267e-06,
      "loss": 1.3247,
      "step": 1894
    },
    {
      "epoch": 1.9465844889573702,
      "grad_norm": 2.886481285095215,
      "learning_rate": 1.97944423296536e-06,
      "loss": 1.4025,
      "step": 1895
    },
    {
      "epoch": 1.9476117103235748,
      "grad_norm": 2.8149733543395996,
      "learning_rate": 1.977540921202893e-06,
      "loss": 1.3745,
      "step": 1896
    },
    {
      "epoch": 1.9486389316897792,
      "grad_norm": 2.643970251083374,
      "learning_rate": 1.9756376094404264e-06,
      "loss": 1.3581,
      "step": 1897
    },
    {
      "epoch": 1.9496661530559836,
      "grad_norm": 2.7485244274139404,
      "learning_rate": 1.9737342976779597e-06,
      "loss": 1.3426,
      "step": 1898
    },
    {
      "epoch": 1.950693374422188,
      "grad_norm": 2.930513858795166,
      "learning_rate": 1.971830985915493e-06,
      "loss": 1.4194,
      "step": 1899
    },
    {
      "epoch": 1.9517205957883923,
      "grad_norm": 2.7637267112731934,
      "learning_rate": 1.9699276741530264e-06,
      "loss": 1.3908,
      "step": 1900
    },
    {
      "epoch": 1.9517205957883923,
      "eval_loss": 1.8510984182357788,
      "eval_runtime": 37.4089,
      "eval_samples_per_second": 26.732,
      "eval_steps_per_second": 1.123,
      "step": 1900
    },
    {
      "epoch": 1.9527478171545969,
      "grad_norm": 2.8946938514709473,
      "learning_rate": 1.96802436239056e-06,
      "loss": 1.3892,
      "step": 1901
    },
    {
      "epoch": 1.9537750385208013,
      "grad_norm": 2.7064263820648193,
      "learning_rate": 1.966121050628093e-06,
      "loss": 1.3628,
      "step": 1902
    },
    {
      "epoch": 1.9548022598870056,
      "grad_norm": 2.8588147163391113,
      "learning_rate": 1.9642177388656265e-06,
      "loss": 1.4097,
      "step": 1903
    },
    {
      "epoch": 1.9558294812532102,
      "grad_norm": 2.7746825218200684,
      "learning_rate": 1.9623144271031595e-06,
      "loss": 1.4255,
      "step": 1904
    },
    {
      "epoch": 1.9568567026194144,
      "grad_norm": 2.872587203979492,
      "learning_rate": 1.960411115340693e-06,
      "loss": 1.4308,
      "step": 1905
    },
    {
      "epoch": 1.957883923985619,
      "grad_norm": 2.714769124984741,
      "learning_rate": 1.958507803578226e-06,
      "loss": 1.3847,
      "step": 1906
    },
    {
      "epoch": 1.9589111453518233,
      "grad_norm": 2.7117342948913574,
      "learning_rate": 1.9566044918157596e-06,
      "loss": 1.406,
      "step": 1907
    },
    {
      "epoch": 1.9599383667180277,
      "grad_norm": 2.754995346069336,
      "learning_rate": 1.954701180053293e-06,
      "loss": 1.4216,
      "step": 1908
    },
    {
      "epoch": 1.9609655880842323,
      "grad_norm": 2.6597256660461426,
      "learning_rate": 1.952797868290826e-06,
      "loss": 1.3748,
      "step": 1909
    },
    {
      "epoch": 1.9619928094504364,
      "grad_norm": 2.7966320514678955,
      "learning_rate": 1.9508945565283596e-06,
      "loss": 1.4237,
      "step": 1910
    },
    {
      "epoch": 1.9619928094504364,
      "eval_loss": 1.8453261852264404,
      "eval_runtime": 36.9326,
      "eval_samples_per_second": 27.076,
      "eval_steps_per_second": 1.137,
      "step": 1910
    },
    {
      "epoch": 1.963020030816641,
      "grad_norm": 2.7504498958587646,
      "learning_rate": 1.948991244765893e-06,
      "loss": 1.3832,
      "step": 1911
    },
    {
      "epoch": 1.9640472521828454,
      "grad_norm": 2.7729601860046387,
      "learning_rate": 1.9470879330034264e-06,
      "loss": 1.4176,
      "step": 1912
    },
    {
      "epoch": 1.9650744735490497,
      "grad_norm": 2.6513381004333496,
      "learning_rate": 1.9451846212409593e-06,
      "loss": 1.3524,
      "step": 1913
    },
    {
      "epoch": 1.9661016949152543,
      "grad_norm": 2.8147811889648438,
      "learning_rate": 1.9432813094784927e-06,
      "loss": 1.4119,
      "step": 1914
    },
    {
      "epoch": 1.9671289162814587,
      "grad_norm": 2.7253763675689697,
      "learning_rate": 1.941377997716026e-06,
      "loss": 1.3458,
      "step": 1915
    },
    {
      "epoch": 1.968156137647663,
      "grad_norm": 2.6741955280303955,
      "learning_rate": 1.9394746859535594e-06,
      "loss": 1.3764,
      "step": 1916
    },
    {
      "epoch": 1.9691833590138677,
      "grad_norm": 2.785856008529663,
      "learning_rate": 1.9375713741910928e-06,
      "loss": 1.3484,
      "step": 1917
    },
    {
      "epoch": 1.9702105803800718,
      "grad_norm": 2.743168592453003,
      "learning_rate": 1.9356680624286257e-06,
      "loss": 1.3779,
      "step": 1918
    },
    {
      "epoch": 1.9712378017462764,
      "grad_norm": 2.827883243560791,
      "learning_rate": 1.933764750666159e-06,
      "loss": 1.3246,
      "step": 1919
    },
    {
      "epoch": 1.9722650231124808,
      "grad_norm": 2.7379672527313232,
      "learning_rate": 1.931861438903693e-06,
      "loss": 1.383,
      "step": 1920
    },
    {
      "epoch": 1.9722650231124808,
      "eval_loss": 1.8433737754821777,
      "eval_runtime": 37.0421,
      "eval_samples_per_second": 26.996,
      "eval_steps_per_second": 1.134,
      "step": 1920
    },
    {
      "epoch": 1.9732922444786851,
      "grad_norm": 2.7514965534210205,
      "learning_rate": 1.929958127141226e-06,
      "loss": 1.3513,
      "step": 1921
    },
    {
      "epoch": 1.9743194658448897,
      "grad_norm": 2.6764447689056396,
      "learning_rate": 1.928054815378759e-06,
      "loss": 1.3718,
      "step": 1922
    },
    {
      "epoch": 1.9753466872110939,
      "grad_norm": 2.8592004776000977,
      "learning_rate": 1.9261515036162925e-06,
      "loss": 1.4479,
      "step": 1923
    },
    {
      "epoch": 1.9763739085772984,
      "grad_norm": 2.8035528659820557,
      "learning_rate": 1.924248191853826e-06,
      "loss": 1.4222,
      "step": 1924
    },
    {
      "epoch": 1.9774011299435028,
      "grad_norm": 2.7685303688049316,
      "learning_rate": 1.9223448800913592e-06,
      "loss": 1.3939,
      "step": 1925
    },
    {
      "epoch": 1.9784283513097072,
      "grad_norm": 2.6679201126098633,
      "learning_rate": 1.9204415683288926e-06,
      "loss": 1.385,
      "step": 1926
    },
    {
      "epoch": 1.9794555726759118,
      "grad_norm": 2.7546098232269287,
      "learning_rate": 1.9185382565664255e-06,
      "loss": 1.4134,
      "step": 1927
    },
    {
      "epoch": 1.980482794042116,
      "grad_norm": 2.794991970062256,
      "learning_rate": 1.916634944803959e-06,
      "loss": 1.4212,
      "step": 1928
    },
    {
      "epoch": 1.9815100154083205,
      "grad_norm": 2.767075538635254,
      "learning_rate": 1.9147316330414923e-06,
      "loss": 1.3979,
      "step": 1929
    },
    {
      "epoch": 1.9825372367745249,
      "grad_norm": 2.7712795734405518,
      "learning_rate": 1.9128283212790256e-06,
      "loss": 1.3827,
      "step": 1930
    },
    {
      "epoch": 1.9825372367745249,
      "eval_loss": 1.8437753915786743,
      "eval_runtime": 37.0885,
      "eval_samples_per_second": 26.963,
      "eval_steps_per_second": 1.132,
      "step": 1930
    },
    {
      "epoch": 1.9835644581407292,
      "grad_norm": 2.791165351867676,
      "learning_rate": 1.910925009516559e-06,
      "loss": 1.4328,
      "step": 1931
    },
    {
      "epoch": 1.9845916795069338,
      "grad_norm": 2.7170767784118652,
      "learning_rate": 1.9090216977540924e-06,
      "loss": 1.3958,
      "step": 1932
    },
    {
      "epoch": 1.9856189008731382,
      "grad_norm": 2.8529574871063232,
      "learning_rate": 1.9071183859916257e-06,
      "loss": 1.4379,
      "step": 1933
    },
    {
      "epoch": 1.9866461222393426,
      "grad_norm": 2.8424859046936035,
      "learning_rate": 1.9052150742291589e-06,
      "loss": 1.4066,
      "step": 1934
    },
    {
      "epoch": 1.9876733436055471,
      "grad_norm": 2.8055100440979004,
      "learning_rate": 1.9033117624666922e-06,
      "loss": 1.4,
      "step": 1935
    },
    {
      "epoch": 1.9887005649717513,
      "grad_norm": 2.73415470123291,
      "learning_rate": 1.9014084507042254e-06,
      "loss": 1.3652,
      "step": 1936
    },
    {
      "epoch": 1.9897277863379559,
      "grad_norm": 2.7907590866088867,
      "learning_rate": 1.8995051389417587e-06,
      "loss": 1.3475,
      "step": 1937
    },
    {
      "epoch": 1.9907550077041603,
      "grad_norm": 2.6994218826293945,
      "learning_rate": 1.8976018271792921e-06,
      "loss": 1.3428,
      "step": 1938
    },
    {
      "epoch": 1.9917822290703646,
      "grad_norm": 2.8237955570220947,
      "learning_rate": 1.8956985154168253e-06,
      "loss": 1.3918,
      "step": 1939
    },
    {
      "epoch": 1.9928094504365692,
      "grad_norm": 2.779270648956299,
      "learning_rate": 1.8937952036543588e-06,
      "loss": 1.3726,
      "step": 1940
    },
    {
      "epoch": 1.9928094504365692,
      "eval_loss": 1.8435344696044922,
      "eval_runtime": 37.0797,
      "eval_samples_per_second": 26.969,
      "eval_steps_per_second": 1.133,
      "step": 1940
    },
    {
      "epoch": 1.9938366718027734,
      "grad_norm": 2.7808098793029785,
      "learning_rate": 1.8918918918918922e-06,
      "loss": 1.3368,
      "step": 1941
    },
    {
      "epoch": 1.994863893168978,
      "grad_norm": 2.7355751991271973,
      "learning_rate": 1.8899885801294253e-06,
      "loss": 1.39,
      "step": 1942
    },
    {
      "epoch": 1.9958911145351823,
      "grad_norm": 2.907468318939209,
      "learning_rate": 1.8880852683669587e-06,
      "loss": 1.399,
      "step": 1943
    },
    {
      "epoch": 1.9969183359013867,
      "grad_norm": 2.8597185611724854,
      "learning_rate": 1.886181956604492e-06,
      "loss": 1.3731,
      "step": 1944
    },
    {
      "epoch": 1.9979455572675913,
      "grad_norm": 2.8638176918029785,
      "learning_rate": 1.8842786448420252e-06,
      "loss": 1.3696,
      "step": 1945
    },
    {
      "epoch": 1.9989727786337956,
      "grad_norm": 2.880169153213501,
      "learning_rate": 1.8823753330795586e-06,
      "loss": 1.3971,
      "step": 1946
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.7660279273986816,
      "learning_rate": 1.8804720213170917e-06,
      "loss": 1.3026,
      "step": 1947
    },
    {
      "epoch": 2.0010272213662046,
      "grad_norm": 2.9506170749664307,
      "learning_rate": 1.878568709554625e-06,
      "loss": 1.3298,
      "step": 1948
    },
    {
      "epoch": 2.0020544427324087,
      "grad_norm": 2.8743488788604736,
      "learning_rate": 1.8766653977921585e-06,
      "loss": 1.3808,
      "step": 1949
    },
    {
      "epoch": 2.0030816640986133,
      "grad_norm": 2.8676979541778564,
      "learning_rate": 1.874762086029692e-06,
      "loss": 1.3394,
      "step": 1950
    },
    {
      "epoch": 2.0030816640986133,
      "eval_loss": 1.8473349809646606,
      "eval_runtime": 37.093,
      "eval_samples_per_second": 26.959,
      "eval_steps_per_second": 1.132,
      "step": 1950
    },
    {
      "epoch": 2.0041088854648175,
      "grad_norm": 3.1344428062438965,
      "learning_rate": 1.8728587742672252e-06,
      "loss": 1.3774,
      "step": 1951
    },
    {
      "epoch": 2.005136106831022,
      "grad_norm": 3.1567065715789795,
      "learning_rate": 1.8709554625047585e-06,
      "loss": 1.3863,
      "step": 1952
    },
    {
      "epoch": 2.0061633281972266,
      "grad_norm": 3.109293222427368,
      "learning_rate": 1.8690521507422917e-06,
      "loss": 1.3196,
      "step": 1953
    },
    {
      "epoch": 2.007190549563431,
      "grad_norm": 3.1035029888153076,
      "learning_rate": 1.867148838979825e-06,
      "loss": 1.3473,
      "step": 1954
    },
    {
      "epoch": 2.0082177709296354,
      "grad_norm": 3.02809476852417,
      "learning_rate": 1.8652455272173584e-06,
      "loss": 1.3718,
      "step": 1955
    },
    {
      "epoch": 2.00924499229584,
      "grad_norm": 2.918850898742676,
      "learning_rate": 1.8633422154548916e-06,
      "loss": 1.3794,
      "step": 1956
    },
    {
      "epoch": 2.010272213662044,
      "grad_norm": 2.89725923538208,
      "learning_rate": 1.861438903692425e-06,
      "loss": 1.2861,
      "step": 1957
    },
    {
      "epoch": 2.0112994350282487,
      "grad_norm": 3.051868200302124,
      "learning_rate": 1.859535591929958e-06,
      "loss": 1.3122,
      "step": 1958
    },
    {
      "epoch": 2.012326656394453,
      "grad_norm": 3.1091904640197754,
      "learning_rate": 1.8576322801674915e-06,
      "loss": 1.3941,
      "step": 1959
    },
    {
      "epoch": 2.0133538777606574,
      "grad_norm": 3.046663761138916,
      "learning_rate": 1.855728968405025e-06,
      "loss": 1.3293,
      "step": 1960
    },
    {
      "epoch": 2.0133538777606574,
      "eval_loss": 1.851180911064148,
      "eval_runtime": 37.0842,
      "eval_samples_per_second": 26.966,
      "eval_steps_per_second": 1.133,
      "step": 1960
    },
    {
      "epoch": 2.014381099126862,
      "grad_norm": 3.2143752574920654,
      "learning_rate": 1.8538256566425584e-06,
      "loss": 1.3431,
      "step": 1961
    },
    {
      "epoch": 2.015408320493066,
      "grad_norm": 3.162942409515381,
      "learning_rate": 1.8519223448800915e-06,
      "loss": 1.3883,
      "step": 1962
    },
    {
      "epoch": 2.0164355418592708,
      "grad_norm": 3.1271183490753174,
      "learning_rate": 1.850019033117625e-06,
      "loss": 1.3295,
      "step": 1963
    },
    {
      "epoch": 2.017462763225475,
      "grad_norm": 3.07658314704895,
      "learning_rate": 1.848115721355158e-06,
      "loss": 1.3323,
      "step": 1964
    },
    {
      "epoch": 2.0184899845916795,
      "grad_norm": 3.075946569442749,
      "learning_rate": 1.8462124095926914e-06,
      "loss": 1.3524,
      "step": 1965
    },
    {
      "epoch": 2.019517205957884,
      "grad_norm": 3.0359086990356445,
      "learning_rate": 1.8443090978302248e-06,
      "loss": 1.3578,
      "step": 1966
    },
    {
      "epoch": 2.0205444273240882,
      "grad_norm": 3.1957743167877197,
      "learning_rate": 1.842405786067758e-06,
      "loss": 1.3909,
      "step": 1967
    },
    {
      "epoch": 2.021571648690293,
      "grad_norm": 3.002462863922119,
      "learning_rate": 1.8405024743052913e-06,
      "loss": 1.3466,
      "step": 1968
    },
    {
      "epoch": 2.022598870056497,
      "grad_norm": 3.0507266521453857,
      "learning_rate": 1.8385991625428247e-06,
      "loss": 1.3717,
      "step": 1969
    },
    {
      "epoch": 2.0236260914227016,
      "grad_norm": 3.0719404220581055,
      "learning_rate": 1.8366958507803578e-06,
      "loss": 1.313,
      "step": 1970
    },
    {
      "epoch": 2.0236260914227016,
      "eval_loss": 1.8496923446655273,
      "eval_runtime": 37.1112,
      "eval_samples_per_second": 26.946,
      "eval_steps_per_second": 1.132,
      "step": 1970
    },
    {
      "epoch": 2.024653312788906,
      "grad_norm": 3.068819522857666,
      "learning_rate": 1.8347925390178914e-06,
      "loss": 1.3614,
      "step": 1971
    },
    {
      "epoch": 2.0256805341551103,
      "grad_norm": 3.005855083465576,
      "learning_rate": 1.8328892272554247e-06,
      "loss": 1.3007,
      "step": 1972
    },
    {
      "epoch": 2.026707755521315,
      "grad_norm": 3.068032741546631,
      "learning_rate": 1.8309859154929579e-06,
      "loss": 1.3644,
      "step": 1973
    },
    {
      "epoch": 2.0277349768875195,
      "grad_norm": 3.066153049468994,
      "learning_rate": 1.8290826037304913e-06,
      "loss": 1.3518,
      "step": 1974
    },
    {
      "epoch": 2.0287621982537236,
      "grad_norm": 3.019429922103882,
      "learning_rate": 1.8271792919680246e-06,
      "loss": 1.3208,
      "step": 1975
    },
    {
      "epoch": 2.029789419619928,
      "grad_norm": 3.2104897499084473,
      "learning_rate": 1.8252759802055578e-06,
      "loss": 1.329,
      "step": 1976
    },
    {
      "epoch": 2.0308166409861323,
      "grad_norm": 3.272581100463867,
      "learning_rate": 1.8233726684430911e-06,
      "loss": 1.3665,
      "step": 1977
    },
    {
      "epoch": 2.031843862352337,
      "grad_norm": 3.2155191898345947,
      "learning_rate": 1.8214693566806243e-06,
      "loss": 1.3178,
      "step": 1978
    },
    {
      "epoch": 2.0328710837185415,
      "grad_norm": 3.130934238433838,
      "learning_rate": 1.8195660449181576e-06,
      "loss": 1.3492,
      "step": 1979
    },
    {
      "epoch": 2.0338983050847457,
      "grad_norm": 3.211571216583252,
      "learning_rate": 1.817662733155691e-06,
      "loss": 1.3477,
      "step": 1980
    },
    {
      "epoch": 2.0338983050847457,
      "eval_loss": 1.8516111373901367,
      "eval_runtime": 37.1099,
      "eval_samples_per_second": 26.947,
      "eval_steps_per_second": 1.132,
      "step": 1980
    },
    {
      "epoch": 2.0349255264509503,
      "grad_norm": 2.9799704551696777,
      "learning_rate": 1.8157594213932244e-06,
      "loss": 1.3469,
      "step": 1981
    },
    {
      "epoch": 2.0359527478171544,
      "grad_norm": 3.0847010612487793,
      "learning_rate": 1.8138561096307577e-06,
      "loss": 1.3587,
      "step": 1982
    },
    {
      "epoch": 2.036979969183359,
      "grad_norm": 3.0739681720733643,
      "learning_rate": 1.811952797868291e-06,
      "loss": 1.3053,
      "step": 1983
    },
    {
      "epoch": 2.0380071905495636,
      "grad_norm": 3.0315258502960205,
      "learning_rate": 1.8100494861058242e-06,
      "loss": 1.3381,
      "step": 1984
    },
    {
      "epoch": 2.0390344119157677,
      "grad_norm": 2.9987258911132812,
      "learning_rate": 1.8081461743433576e-06,
      "loss": 1.3545,
      "step": 1985
    },
    {
      "epoch": 2.0400616332819723,
      "grad_norm": 2.9383034706115723,
      "learning_rate": 1.806242862580891e-06,
      "loss": 1.3085,
      "step": 1986
    },
    {
      "epoch": 2.0410888546481765,
      "grad_norm": 3.064450740814209,
      "learning_rate": 1.8043395508184241e-06,
      "loss": 1.3572,
      "step": 1987
    },
    {
      "epoch": 2.042116076014381,
      "grad_norm": 3.102689027786255,
      "learning_rate": 1.8024362390559575e-06,
      "loss": 1.3453,
      "step": 1988
    },
    {
      "epoch": 2.0431432973805856,
      "grad_norm": 3.0369763374328613,
      "learning_rate": 1.8005329272934906e-06,
      "loss": 1.3123,
      "step": 1989
    },
    {
      "epoch": 2.04417051874679,
      "grad_norm": 3.2758162021636963,
      "learning_rate": 1.798629615531024e-06,
      "loss": 1.4068,
      "step": 1990
    },
    {
      "epoch": 2.04417051874679,
      "eval_loss": 1.8500858545303345,
      "eval_runtime": 37.334,
      "eval_samples_per_second": 26.785,
      "eval_steps_per_second": 1.125,
      "step": 1990
    },
    {
      "epoch": 2.0451977401129944,
      "grad_norm": 3.1648876667022705,
      "learning_rate": 1.7967263037685576e-06,
      "loss": 1.3328,
      "step": 1991
    },
    {
      "epoch": 2.046224961479199,
      "grad_norm": 3.104123830795288,
      "learning_rate": 1.794822992006091e-06,
      "loss": 1.3297,
      "step": 1992
    },
    {
      "epoch": 2.047252182845403,
      "grad_norm": 3.198483943939209,
      "learning_rate": 1.792919680243624e-06,
      "loss": 1.3851,
      "step": 1993
    },
    {
      "epoch": 2.0482794042116077,
      "grad_norm": 3.212456226348877,
      "learning_rate": 1.7910163684811574e-06,
      "loss": 1.3062,
      "step": 1994
    },
    {
      "epoch": 2.049306625577812,
      "grad_norm": 3.058353900909424,
      "learning_rate": 1.7891130567186906e-06,
      "loss": 1.3347,
      "step": 1995
    },
    {
      "epoch": 2.0503338469440164,
      "grad_norm": 2.963183879852295,
      "learning_rate": 1.787209744956224e-06,
      "loss": 1.3163,
      "step": 1996
    },
    {
      "epoch": 2.051361068310221,
      "grad_norm": 3.1013290882110596,
      "learning_rate": 1.7853064331937573e-06,
      "loss": 1.3377,
      "step": 1997
    },
    {
      "epoch": 2.052388289676425,
      "grad_norm": 3.068808078765869,
      "learning_rate": 1.7834031214312905e-06,
      "loss": 1.3772,
      "step": 1998
    },
    {
      "epoch": 2.0534155110426298,
      "grad_norm": 3.0556447505950928,
      "learning_rate": 1.7814998096688238e-06,
      "loss": 1.3113,
      "step": 1999
    },
    {
      "epoch": 2.054442732408834,
      "grad_norm": 3.115323543548584,
      "learning_rate": 1.779596497906357e-06,
      "loss": 1.3499,
      "step": 2000
    },
    {
      "epoch": 2.054442732408834,
      "eval_loss": 1.8396267890930176,
      "eval_runtime": 37.4474,
      "eval_samples_per_second": 26.704,
      "eval_steps_per_second": 1.122,
      "step": 2000
    },
    {
      "epoch": 2.0554699537750385,
      "grad_norm": 3.133096694946289,
      "learning_rate": 1.7776931861438906e-06,
      "loss": 1.3938,
      "step": 2001
    },
    {
      "epoch": 2.056497175141243,
      "grad_norm": 3.259174108505249,
      "learning_rate": 1.775789874381424e-06,
      "loss": 1.382,
      "step": 2002
    },
    {
      "epoch": 2.0575243965074472,
      "grad_norm": 3.17948317527771,
      "learning_rate": 1.7738865626189573e-06,
      "loss": 1.3643,
      "step": 2003
    },
    {
      "epoch": 2.058551617873652,
      "grad_norm": 3.0918309688568115,
      "learning_rate": 1.7719832508564904e-06,
      "loss": 1.3067,
      "step": 2004
    },
    {
      "epoch": 2.0595788392398564,
      "grad_norm": 2.9949889183044434,
      "learning_rate": 1.7700799390940238e-06,
      "loss": 1.3323,
      "step": 2005
    },
    {
      "epoch": 2.0606060606060606,
      "grad_norm": 3.044935464859009,
      "learning_rate": 1.768176627331557e-06,
      "loss": 1.359,
      "step": 2006
    },
    {
      "epoch": 2.061633281972265,
      "grad_norm": 2.9965357780456543,
      "learning_rate": 1.7662733155690903e-06,
      "loss": 1.3456,
      "step": 2007
    },
    {
      "epoch": 2.0626605033384693,
      "grad_norm": 3.092984676361084,
      "learning_rate": 1.7643700038066237e-06,
      "loss": 1.3831,
      "step": 2008
    },
    {
      "epoch": 2.063687724704674,
      "grad_norm": 3.004263162612915,
      "learning_rate": 1.7624666920441568e-06,
      "loss": 1.3087,
      "step": 2009
    },
    {
      "epoch": 2.0647149460708785,
      "grad_norm": 3.0408692359924316,
      "learning_rate": 1.7605633802816902e-06,
      "loss": 1.3058,
      "step": 2010
    },
    {
      "epoch": 2.0647149460708785,
      "eval_loss": 1.8413820266723633,
      "eval_runtime": 36.9354,
      "eval_samples_per_second": 27.074,
      "eval_steps_per_second": 1.137,
      "step": 2010
    },
    {
      "epoch": 2.0657421674370826,
      "grad_norm": 3.155033826828003,
      "learning_rate": 1.7586600685192238e-06,
      "loss": 1.394,
      "step": 2011
    },
    {
      "epoch": 2.066769388803287,
      "grad_norm": 3.3466567993164062,
      "learning_rate": 1.756756756756757e-06,
      "loss": 1.401,
      "step": 2012
    },
    {
      "epoch": 2.0677966101694913,
      "grad_norm": 3.219165802001953,
      "learning_rate": 1.7548534449942903e-06,
      "loss": 1.3564,
      "step": 2013
    },
    {
      "epoch": 2.068823831535696,
      "grad_norm": 3.1733009815216064,
      "learning_rate": 1.7529501332318236e-06,
      "loss": 1.3446,
      "step": 2014
    },
    {
      "epoch": 2.0698510529019005,
      "grad_norm": 3.1392745971679688,
      "learning_rate": 1.7510468214693568e-06,
      "loss": 1.3212,
      "step": 2015
    },
    {
      "epoch": 2.0708782742681047,
      "grad_norm": 2.9845967292785645,
      "learning_rate": 1.7491435097068902e-06,
      "loss": 1.3117,
      "step": 2016
    },
    {
      "epoch": 2.0719054956343093,
      "grad_norm": 3.2182602882385254,
      "learning_rate": 1.7472401979444233e-06,
      "loss": 1.3539,
      "step": 2017
    },
    {
      "epoch": 2.0729327170005134,
      "grad_norm": 3.0809013843536377,
      "learning_rate": 1.7453368861819567e-06,
      "loss": 1.2891,
      "step": 2018
    },
    {
      "epoch": 2.073959938366718,
      "grad_norm": 3.0690958499908447,
      "learning_rate": 1.74343357441949e-06,
      "loss": 1.3381,
      "step": 2019
    },
    {
      "epoch": 2.0749871597329226,
      "grad_norm": 3.091454267501831,
      "learning_rate": 1.7415302626570232e-06,
      "loss": 1.3874,
      "step": 2020
    },
    {
      "epoch": 2.0749871597329226,
      "eval_loss": 1.8516818284988403,
      "eval_runtime": 37.0619,
      "eval_samples_per_second": 26.982,
      "eval_steps_per_second": 1.133,
      "step": 2020
    },
    {
      "epoch": 2.0760143810991267,
      "grad_norm": 3.039327621459961,
      "learning_rate": 1.7396269508945568e-06,
      "loss": 1.3295,
      "step": 2021
    },
    {
      "epoch": 2.0770416024653313,
      "grad_norm": 3.0967671871185303,
      "learning_rate": 1.7377236391320901e-06,
      "loss": 1.3122,
      "step": 2022
    },
    {
      "epoch": 2.078068823831536,
      "grad_norm": 3.228410005569458,
      "learning_rate": 1.7358203273696233e-06,
      "loss": 1.3766,
      "step": 2023
    },
    {
      "epoch": 2.07909604519774,
      "grad_norm": 3.1973841190338135,
      "learning_rate": 1.7339170156071566e-06,
      "loss": 1.3325,
      "step": 2024
    },
    {
      "epoch": 2.0801232665639446,
      "grad_norm": 3.131504535675049,
      "learning_rate": 1.73201370384469e-06,
      "loss": 1.3355,
      "step": 2025
    },
    {
      "epoch": 2.081150487930149,
      "grad_norm": 3.132685661315918,
      "learning_rate": 1.7301103920822231e-06,
      "loss": 1.3446,
      "step": 2026
    },
    {
      "epoch": 2.0821777092963534,
      "grad_norm": 3.2344162464141846,
      "learning_rate": 1.7282070803197565e-06,
      "loss": 1.3525,
      "step": 2027
    },
    {
      "epoch": 2.083204930662558,
      "grad_norm": 3.1193928718566895,
      "learning_rate": 1.7263037685572897e-06,
      "loss": 1.3415,
      "step": 2028
    },
    {
      "epoch": 2.084232152028762,
      "grad_norm": 3.164365291595459,
      "learning_rate": 1.724400456794823e-06,
      "loss": 1.3652,
      "step": 2029
    },
    {
      "epoch": 2.0852593733949667,
      "grad_norm": 3.0825226306915283,
      "learning_rate": 1.7224971450323564e-06,
      "loss": 1.3242,
      "step": 2030
    },
    {
      "epoch": 2.0852593733949667,
      "eval_loss": 1.8433457612991333,
      "eval_runtime": 37.0921,
      "eval_samples_per_second": 26.96,
      "eval_steps_per_second": 1.132,
      "step": 2030
    },
    {
      "epoch": 2.086286594761171,
      "grad_norm": 3.1651787757873535,
      "learning_rate": 1.72059383326989e-06,
      "loss": 1.333,
      "step": 2031
    },
    {
      "epoch": 2.0873138161273754,
      "grad_norm": 3.295436382293701,
      "learning_rate": 1.7186905215074231e-06,
      "loss": 1.4064,
      "step": 2032
    },
    {
      "epoch": 2.08834103749358,
      "grad_norm": 3.177628517150879,
      "learning_rate": 1.7167872097449565e-06,
      "loss": 1.3173,
      "step": 2033
    },
    {
      "epoch": 2.089368258859784,
      "grad_norm": 3.112971544265747,
      "learning_rate": 1.7148838979824896e-06,
      "loss": 1.3401,
      "step": 2034
    },
    {
      "epoch": 2.0903954802259888,
      "grad_norm": 3.150045871734619,
      "learning_rate": 1.712980586220023e-06,
      "loss": 1.3371,
      "step": 2035
    },
    {
      "epoch": 2.091422701592193,
      "grad_norm": 3.0645205974578857,
      "learning_rate": 1.7110772744575563e-06,
      "loss": 1.3213,
      "step": 2036
    },
    {
      "epoch": 2.0924499229583975,
      "grad_norm": 3.230185031890869,
      "learning_rate": 1.7091739626950895e-06,
      "loss": 1.3317,
      "step": 2037
    },
    {
      "epoch": 2.093477144324602,
      "grad_norm": 3.1509249210357666,
      "learning_rate": 1.7072706509326229e-06,
      "loss": 1.3706,
      "step": 2038
    },
    {
      "epoch": 2.0945043656908062,
      "grad_norm": 3.185981273651123,
      "learning_rate": 1.7053673391701562e-06,
      "loss": 1.3398,
      "step": 2039
    },
    {
      "epoch": 2.095531587057011,
      "grad_norm": 3.000025510787964,
      "learning_rate": 1.7034640274076894e-06,
      "loss": 1.3197,
      "step": 2040
    },
    {
      "epoch": 2.095531587057011,
      "eval_loss": 1.8413926362991333,
      "eval_runtime": 37.0971,
      "eval_samples_per_second": 26.956,
      "eval_steps_per_second": 1.132,
      "step": 2040
    },
    {
      "epoch": 2.0965588084232154,
      "grad_norm": 3.156095027923584,
      "learning_rate": 1.701560715645223e-06,
      "loss": 1.3864,
      "step": 2041
    },
    {
      "epoch": 2.0975860297894195,
      "grad_norm": 2.970560073852539,
      "learning_rate": 1.6996574038827563e-06,
      "loss": 1.3016,
      "step": 2042
    },
    {
      "epoch": 2.098613251155624,
      "grad_norm": 3.141657829284668,
      "learning_rate": 1.6977540921202895e-06,
      "loss": 1.3488,
      "step": 2043
    },
    {
      "epoch": 2.0996404725218283,
      "grad_norm": 3.125731945037842,
      "learning_rate": 1.6958507803578228e-06,
      "loss": 1.3275,
      "step": 2044
    },
    {
      "epoch": 2.100667693888033,
      "grad_norm": 3.2703487873077393,
      "learning_rate": 1.693947468595356e-06,
      "loss": 1.3577,
      "step": 2045
    },
    {
      "epoch": 2.1016949152542375,
      "grad_norm": 3.1855714321136475,
      "learning_rate": 1.6920441568328893e-06,
      "loss": 1.3671,
      "step": 2046
    },
    {
      "epoch": 2.1027221366204416,
      "grad_norm": 3.0699303150177,
      "learning_rate": 1.6901408450704227e-06,
      "loss": 1.3066,
      "step": 2047
    },
    {
      "epoch": 2.103749357986646,
      "grad_norm": 3.2417423725128174,
      "learning_rate": 1.6882375333079559e-06,
      "loss": 1.3274,
      "step": 2048
    },
    {
      "epoch": 2.1047765793528503,
      "grad_norm": 3.265108108520508,
      "learning_rate": 1.6863342215454892e-06,
      "loss": 1.3924,
      "step": 2049
    },
    {
      "epoch": 2.105803800719055,
      "grad_norm": 3.2078635692596436,
      "learning_rate": 1.6844309097830226e-06,
      "loss": 1.3782,
      "step": 2050
    },
    {
      "epoch": 2.105803800719055,
      "eval_loss": 1.8441476821899414,
      "eval_runtime": 37.1042,
      "eval_samples_per_second": 26.951,
      "eval_steps_per_second": 1.132,
      "step": 2050
    },
    {
      "epoch": 2.1068310220852595,
      "grad_norm": 3.180485248565674,
      "learning_rate": 1.682527598020556e-06,
      "loss": 1.3613,
      "step": 2051
    },
    {
      "epoch": 2.1078582434514637,
      "grad_norm": 3.110530376434326,
      "learning_rate": 1.6806242862580893e-06,
      "loss": 1.376,
      "step": 2052
    },
    {
      "epoch": 2.1088854648176683,
      "grad_norm": 2.997999429702759,
      "learning_rate": 1.6787209744956227e-06,
      "loss": 1.3172,
      "step": 2053
    },
    {
      "epoch": 2.109912686183873,
      "grad_norm": 3.0154783725738525,
      "learning_rate": 1.6768176627331558e-06,
      "loss": 1.3015,
      "step": 2054
    },
    {
      "epoch": 2.110939907550077,
      "grad_norm": 2.992138624191284,
      "learning_rate": 1.6749143509706892e-06,
      "loss": 1.3517,
      "step": 2055
    },
    {
      "epoch": 2.1119671289162816,
      "grad_norm": 3.0572457313537598,
      "learning_rate": 1.6730110392082225e-06,
      "loss": 1.3301,
      "step": 2056
    },
    {
      "epoch": 2.1129943502824857,
      "grad_norm": 2.9924540519714355,
      "learning_rate": 1.6711077274457557e-06,
      "loss": 1.3534,
      "step": 2057
    },
    {
      "epoch": 2.1140215716486903,
      "grad_norm": 2.9569015502929688,
      "learning_rate": 1.669204415683289e-06,
      "loss": 1.2972,
      "step": 2058
    },
    {
      "epoch": 2.115048793014895,
      "grad_norm": 3.064371109008789,
      "learning_rate": 1.6673011039208222e-06,
      "loss": 1.3553,
      "step": 2059
    },
    {
      "epoch": 2.116076014381099,
      "grad_norm": 3.0954525470733643,
      "learning_rate": 1.6653977921583556e-06,
      "loss": 1.3421,
      "step": 2060
    },
    {
      "epoch": 2.116076014381099,
      "eval_loss": 1.8407723903656006,
      "eval_runtime": 37.1122,
      "eval_samples_per_second": 26.945,
      "eval_steps_per_second": 1.132,
      "step": 2060
    },
    {
      "epoch": 2.1171032357473036,
      "grad_norm": 3.2661712169647217,
      "learning_rate": 1.663494480395889e-06,
      "loss": 1.3582,
      "step": 2061
    },
    {
      "epoch": 2.118130457113508,
      "grad_norm": 3.1663613319396973,
      "learning_rate": 1.6615911686334225e-06,
      "loss": 1.3203,
      "step": 2062
    },
    {
      "epoch": 2.1191576784797124,
      "grad_norm": 3.0864570140838623,
      "learning_rate": 1.6596878568709557e-06,
      "loss": 1.3431,
      "step": 2063
    },
    {
      "epoch": 2.120184899845917,
      "grad_norm": 3.1531989574432373,
      "learning_rate": 1.657784545108489e-06,
      "loss": 1.3275,
      "step": 2064
    },
    {
      "epoch": 2.121212121212121,
      "grad_norm": 3.2258596420288086,
      "learning_rate": 1.6558812333460222e-06,
      "loss": 1.3711,
      "step": 2065
    },
    {
      "epoch": 2.1222393425783257,
      "grad_norm": 3.2881453037261963,
      "learning_rate": 1.6539779215835555e-06,
      "loss": 1.3688,
      "step": 2066
    },
    {
      "epoch": 2.12326656394453,
      "grad_norm": 3.0734169483184814,
      "learning_rate": 1.652074609821089e-06,
      "loss": 1.3277,
      "step": 2067
    },
    {
      "epoch": 2.1242937853107344,
      "grad_norm": 2.9943618774414062,
      "learning_rate": 1.650171298058622e-06,
      "loss": 1.3268,
      "step": 2068
    },
    {
      "epoch": 2.125321006676939,
      "grad_norm": 3.034625291824341,
      "learning_rate": 1.6482679862961554e-06,
      "loss": 1.3265,
      "step": 2069
    },
    {
      "epoch": 2.126348228043143,
      "grad_norm": 3.0360612869262695,
      "learning_rate": 1.6463646745336886e-06,
      "loss": 1.3256,
      "step": 2070
    },
    {
      "epoch": 2.126348228043143,
      "eval_loss": 1.8443471193313599,
      "eval_runtime": 37.0917,
      "eval_samples_per_second": 26.96,
      "eval_steps_per_second": 1.132,
      "step": 2070
    },
    {
      "epoch": 2.1273754494093478,
      "grad_norm": 3.0917110443115234,
      "learning_rate": 1.644461362771222e-06,
      "loss": 1.3099,
      "step": 2071
    },
    {
      "epoch": 2.128402670775552,
      "grad_norm": 3.1142187118530273,
      "learning_rate": 1.6425580510087555e-06,
      "loss": 1.3421,
      "step": 2072
    },
    {
      "epoch": 2.1294298921417565,
      "grad_norm": 3.2128524780273438,
      "learning_rate": 1.6406547392462889e-06,
      "loss": 1.3123,
      "step": 2073
    },
    {
      "epoch": 2.130457113507961,
      "grad_norm": 3.060520887374878,
      "learning_rate": 1.638751427483822e-06,
      "loss": 1.3069,
      "step": 2074
    },
    {
      "epoch": 2.131484334874165,
      "grad_norm": 3.092120885848999,
      "learning_rate": 1.6368481157213554e-06,
      "loss": 1.3537,
      "step": 2075
    },
    {
      "epoch": 2.13251155624037,
      "grad_norm": 3.2490074634552,
      "learning_rate": 1.6349448039588885e-06,
      "loss": 1.3324,
      "step": 2076
    },
    {
      "epoch": 2.1335387776065744,
      "grad_norm": 3.2517645359039307,
      "learning_rate": 1.6330414921964219e-06,
      "loss": 1.385,
      "step": 2077
    },
    {
      "epoch": 2.1345659989727785,
      "grad_norm": 3.1669421195983887,
      "learning_rate": 1.6311381804339553e-06,
      "loss": 1.3054,
      "step": 2078
    },
    {
      "epoch": 2.135593220338983,
      "grad_norm": 3.2323193550109863,
      "learning_rate": 1.6292348686714884e-06,
      "loss": 1.3488,
      "step": 2079
    },
    {
      "epoch": 2.1366204417051873,
      "grad_norm": 3.1058294773101807,
      "learning_rate": 1.6273315569090218e-06,
      "loss": 1.3256,
      "step": 2080
    },
    {
      "epoch": 2.1366204417051873,
      "eval_loss": 1.8446909189224243,
      "eval_runtime": 37.1231,
      "eval_samples_per_second": 26.937,
      "eval_steps_per_second": 1.131,
      "step": 2080
    },
    {
      "epoch": 2.137647663071392,
      "grad_norm": 3.1662075519561768,
      "learning_rate": 1.625428245146555e-06,
      "loss": 1.4073,
      "step": 2081
    },
    {
      "epoch": 2.1386748844375965,
      "grad_norm": 3.083482503890991,
      "learning_rate": 1.6235249333840885e-06,
      "loss": 1.3897,
      "step": 2082
    },
    {
      "epoch": 2.1397021058038006,
      "grad_norm": 3.040388345718384,
      "learning_rate": 1.6216216216216219e-06,
      "loss": 1.349,
      "step": 2083
    },
    {
      "epoch": 2.140729327170005,
      "grad_norm": 3.124114513397217,
      "learning_rate": 1.6197183098591552e-06,
      "loss": 1.3732,
      "step": 2084
    },
    {
      "epoch": 2.1417565485362093,
      "grad_norm": 3.2662036418914795,
      "learning_rate": 1.6178149980966884e-06,
      "loss": 1.3522,
      "step": 2085
    },
    {
      "epoch": 2.142783769902414,
      "grad_norm": 3.149472951889038,
      "learning_rate": 1.6159116863342217e-06,
      "loss": 1.3179,
      "step": 2086
    },
    {
      "epoch": 2.1438109912686185,
      "grad_norm": 3.256852865219116,
      "learning_rate": 1.6140083745717549e-06,
      "loss": 1.3897,
      "step": 2087
    },
    {
      "epoch": 2.1448382126348227,
      "grad_norm": 3.1227493286132812,
      "learning_rate": 1.6121050628092882e-06,
      "loss": 1.3104,
      "step": 2088
    },
    {
      "epoch": 2.1458654340010272,
      "grad_norm": 3.3213047981262207,
      "learning_rate": 1.6102017510468216e-06,
      "loss": 1.3829,
      "step": 2089
    },
    {
      "epoch": 2.146892655367232,
      "grad_norm": 3.160428047180176,
      "learning_rate": 1.6082984392843548e-06,
      "loss": 1.2939,
      "step": 2090
    },
    {
      "epoch": 2.146892655367232,
      "eval_loss": 1.8534585237503052,
      "eval_runtime": 37.3333,
      "eval_samples_per_second": 26.786,
      "eval_steps_per_second": 1.125,
      "step": 2090
    },
    {
      "epoch": 2.147919876733436,
      "grad_norm": 3.224741220474243,
      "learning_rate": 1.6063951275218881e-06,
      "loss": 1.3418,
      "step": 2091
    },
    {
      "epoch": 2.1489470980996406,
      "grad_norm": 3.3884503841400146,
      "learning_rate": 1.6044918157594217e-06,
      "loss": 1.395,
      "step": 2092
    },
    {
      "epoch": 2.1499743194658447,
      "grad_norm": 3.1761722564697266,
      "learning_rate": 1.6025885039969548e-06,
      "loss": 1.3208,
      "step": 2093
    },
    {
      "epoch": 2.1510015408320493,
      "grad_norm": 3.0476465225219727,
      "learning_rate": 1.6006851922344882e-06,
      "loss": 1.3465,
      "step": 2094
    },
    {
      "epoch": 2.152028762198254,
      "grad_norm": 3.1662650108337402,
      "learning_rate": 1.5987818804720216e-06,
      "loss": 1.3645,
      "step": 2095
    },
    {
      "epoch": 2.153055983564458,
      "grad_norm": 3.1860709190368652,
      "learning_rate": 1.5968785687095547e-06,
      "loss": 1.361,
      "step": 2096
    },
    {
      "epoch": 2.1540832049306626,
      "grad_norm": 3.104501962661743,
      "learning_rate": 1.594975256947088e-06,
      "loss": 1.3135,
      "step": 2097
    },
    {
      "epoch": 2.1551104262968668,
      "grad_norm": 3.067155599594116,
      "learning_rate": 1.5930719451846212e-06,
      "loss": 1.3693,
      "step": 2098
    },
    {
      "epoch": 2.1561376476630714,
      "grad_norm": 3.195204019546509,
      "learning_rate": 1.5911686334221546e-06,
      "loss": 1.3143,
      "step": 2099
    },
    {
      "epoch": 2.157164869029276,
      "grad_norm": 3.159278154373169,
      "learning_rate": 1.589265321659688e-06,
      "loss": 1.3607,
      "step": 2100
    },
    {
      "epoch": 2.157164869029276,
      "eval_loss": 1.8502479791641235,
      "eval_runtime": 37.4583,
      "eval_samples_per_second": 26.696,
      "eval_steps_per_second": 1.121,
      "step": 2100
    },
    {
      "epoch": 2.15819209039548,
      "grad_norm": 2.973191261291504,
      "learning_rate": 1.5873620098972211e-06,
      "loss": 1.2776,
      "step": 2101
    },
    {
      "epoch": 2.1592193117616847,
      "grad_norm": 3.0078418254852295,
      "learning_rate": 1.5854586981347547e-06,
      "loss": 1.3173,
      "step": 2102
    },
    {
      "epoch": 2.1602465331278893,
      "grad_norm": 3.2115561962127686,
      "learning_rate": 1.583555386372288e-06,
      "loss": 1.3621,
      "step": 2103
    },
    {
      "epoch": 2.1612737544940934,
      "grad_norm": 3.234210968017578,
      "learning_rate": 1.5816520746098212e-06,
      "loss": 1.374,
      "step": 2104
    },
    {
      "epoch": 2.162300975860298,
      "grad_norm": 3.195405960083008,
      "learning_rate": 1.5797487628473546e-06,
      "loss": 1.3402,
      "step": 2105
    },
    {
      "epoch": 2.163328197226502,
      "grad_norm": 3.016078233718872,
      "learning_rate": 1.577845451084888e-06,
      "loss": 1.3036,
      "step": 2106
    },
    {
      "epoch": 2.1643554185927067,
      "grad_norm": 3.2147345542907715,
      "learning_rate": 1.575942139322421e-06,
      "loss": 1.3511,
      "step": 2107
    },
    {
      "epoch": 2.1653826399589113,
      "grad_norm": 2.980320692062378,
      "learning_rate": 1.5740388275599544e-06,
      "loss": 1.2635,
      "step": 2108
    },
    {
      "epoch": 2.1664098613251155,
      "grad_norm": 3.0693414211273193,
      "learning_rate": 1.5721355157974876e-06,
      "loss": 1.3435,
      "step": 2109
    },
    {
      "epoch": 2.16743708269132,
      "grad_norm": 3.1445682048797607,
      "learning_rate": 1.570232204035021e-06,
      "loss": 1.3294,
      "step": 2110
    },
    {
      "epoch": 2.16743708269132,
      "eval_loss": 1.8479558229446411,
      "eval_runtime": 36.9467,
      "eval_samples_per_second": 27.066,
      "eval_steps_per_second": 1.137,
      "step": 2110
    },
    {
      "epoch": 2.168464304057524,
      "grad_norm": 3.1309633255004883,
      "learning_rate": 1.5683288922725543e-06,
      "loss": 1.3312,
      "step": 2111
    },
    {
      "epoch": 2.169491525423729,
      "grad_norm": 3.0906994342803955,
      "learning_rate": 1.5664255805100879e-06,
      "loss": 1.342,
      "step": 2112
    },
    {
      "epoch": 2.1705187467899334,
      "grad_norm": 3.1987171173095703,
      "learning_rate": 1.564522268747621e-06,
      "loss": 1.3358,
      "step": 2113
    },
    {
      "epoch": 2.1715459681561375,
      "grad_norm": 3.221889019012451,
      "learning_rate": 1.5626189569851544e-06,
      "loss": 1.3484,
      "step": 2114
    },
    {
      "epoch": 2.172573189522342,
      "grad_norm": 3.1508629322052,
      "learning_rate": 1.5607156452226875e-06,
      "loss": 1.2785,
      "step": 2115
    },
    {
      "epoch": 2.1736004108885463,
      "grad_norm": 3.115274667739868,
      "learning_rate": 1.558812333460221e-06,
      "loss": 1.3463,
      "step": 2116
    },
    {
      "epoch": 2.174627632254751,
      "grad_norm": 3.0454249382019043,
      "learning_rate": 1.5569090216977543e-06,
      "loss": 1.3534,
      "step": 2117
    },
    {
      "epoch": 2.1756548536209555,
      "grad_norm": 3.20078182220459,
      "learning_rate": 1.5550057099352874e-06,
      "loss": 1.3469,
      "step": 2118
    },
    {
      "epoch": 2.1766820749871596,
      "grad_norm": 3.091568946838379,
      "learning_rate": 1.5531023981728208e-06,
      "loss": 1.3547,
      "step": 2119
    },
    {
      "epoch": 2.177709296353364,
      "grad_norm": 3.0242369174957275,
      "learning_rate": 1.5511990864103542e-06,
      "loss": 1.3198,
      "step": 2120
    },
    {
      "epoch": 2.177709296353364,
      "eval_loss": 1.8469057083129883,
      "eval_runtime": 37.0881,
      "eval_samples_per_second": 26.963,
      "eval_steps_per_second": 1.132,
      "step": 2120
    },
    {
      "epoch": 2.1787365177195683,
      "grad_norm": 3.1775739192962646,
      "learning_rate": 1.5492957746478873e-06,
      "loss": 1.3491,
      "step": 2121
    },
    {
      "epoch": 2.179763739085773,
      "grad_norm": 3.239668846130371,
      "learning_rate": 1.5473924628854209e-06,
      "loss": 1.3641,
      "step": 2122
    },
    {
      "epoch": 2.1807909604519775,
      "grad_norm": 3.194678544998169,
      "learning_rate": 1.5454891511229542e-06,
      "loss": 1.3317,
      "step": 2123
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 3.129948377609253,
      "learning_rate": 1.5435858393604874e-06,
      "loss": 1.3405,
      "step": 2124
    },
    {
      "epoch": 2.1828454031843862,
      "grad_norm": 3.2040436267852783,
      "learning_rate": 1.5416825275980208e-06,
      "loss": 1.3596,
      "step": 2125
    },
    {
      "epoch": 2.183872624550591,
      "grad_norm": 3.2529456615448,
      "learning_rate": 1.539779215835554e-06,
      "loss": 1.3171,
      "step": 2126
    },
    {
      "epoch": 2.184899845916795,
      "grad_norm": 3.2071499824523926,
      "learning_rate": 1.5378759040730873e-06,
      "loss": 1.318,
      "step": 2127
    },
    {
      "epoch": 2.1859270672829996,
      "grad_norm": 3.2709555625915527,
      "learning_rate": 1.5359725923106206e-06,
      "loss": 1.3811,
      "step": 2128
    },
    {
      "epoch": 2.1869542886492037,
      "grad_norm": 3.244680643081665,
      "learning_rate": 1.5340692805481538e-06,
      "loss": 1.3539,
      "step": 2129
    },
    {
      "epoch": 2.1879815100154083,
      "grad_norm": 3.068683624267578,
      "learning_rate": 1.5321659687856871e-06,
      "loss": 1.3207,
      "step": 2130
    },
    {
      "epoch": 2.1879815100154083,
      "eval_loss": 1.8478527069091797,
      "eval_runtime": 37.138,
      "eval_samples_per_second": 26.927,
      "eval_steps_per_second": 1.131,
      "step": 2130
    },
    {
      "epoch": 2.189008731381613,
      "grad_norm": 3.1829938888549805,
      "learning_rate": 1.5302626570232205e-06,
      "loss": 1.3127,
      "step": 2131
    },
    {
      "epoch": 2.190035952747817,
      "grad_norm": 3.060347080230713,
      "learning_rate": 1.5283593452607539e-06,
      "loss": 1.287,
      "step": 2132
    },
    {
      "epoch": 2.1910631741140216,
      "grad_norm": 3.3016905784606934,
      "learning_rate": 1.5264560334982872e-06,
      "loss": 1.3838,
      "step": 2133
    },
    {
      "epoch": 2.1920903954802258,
      "grad_norm": 3.1945807933807373,
      "learning_rate": 1.5245527217358206e-06,
      "loss": 1.3311,
      "step": 2134
    },
    {
      "epoch": 2.1931176168464304,
      "grad_norm": 3.182347059249878,
      "learning_rate": 1.5226494099733537e-06,
      "loss": 1.3544,
      "step": 2135
    },
    {
      "epoch": 2.194144838212635,
      "grad_norm": 3.069206476211548,
      "learning_rate": 1.520746098210887e-06,
      "loss": 1.3148,
      "step": 2136
    },
    {
      "epoch": 2.195172059578839,
      "grad_norm": 3.1632168292999268,
      "learning_rate": 1.5188427864484205e-06,
      "loss": 1.3857,
      "step": 2137
    },
    {
      "epoch": 2.1961992809450437,
      "grad_norm": 3.055793523788452,
      "learning_rate": 1.5169394746859536e-06,
      "loss": 1.3116,
      "step": 2138
    },
    {
      "epoch": 2.1972265023112483,
      "grad_norm": 3.1330678462982178,
      "learning_rate": 1.515036162923487e-06,
      "loss": 1.3214,
      "step": 2139
    },
    {
      "epoch": 2.1982537236774524,
      "grad_norm": 3.118450880050659,
      "learning_rate": 1.5131328511610201e-06,
      "loss": 1.3479,
      "step": 2140
    },
    {
      "epoch": 2.1982537236774524,
      "eval_loss": 1.8538312911987305,
      "eval_runtime": 37.122,
      "eval_samples_per_second": 26.938,
      "eval_steps_per_second": 1.131,
      "step": 2140
    },
    {
      "epoch": 2.199280945043657,
      "grad_norm": 3.249545097351074,
      "learning_rate": 1.5112295393985535e-06,
      "loss": 1.379,
      "step": 2141
    },
    {
      "epoch": 2.200308166409861,
      "grad_norm": 3.1331286430358887,
      "learning_rate": 1.509326227636087e-06,
      "loss": 1.3621,
      "step": 2142
    },
    {
      "epoch": 2.2013353877760657,
      "grad_norm": 3.150987148284912,
      "learning_rate": 1.5074229158736204e-06,
      "loss": 1.291,
      "step": 2143
    },
    {
      "epoch": 2.2023626091422703,
      "grad_norm": 3.1020572185516357,
      "learning_rate": 1.5055196041111536e-06,
      "loss": 1.3381,
      "step": 2144
    },
    {
      "epoch": 2.2033898305084745,
      "grad_norm": 3.129390239715576,
      "learning_rate": 1.503616292348687e-06,
      "loss": 1.3405,
      "step": 2145
    },
    {
      "epoch": 2.204417051874679,
      "grad_norm": 3.2193870544433594,
      "learning_rate": 1.50171298058622e-06,
      "loss": 1.3744,
      "step": 2146
    },
    {
      "epoch": 2.205444273240883,
      "grad_norm": 3.043828248977661,
      "learning_rate": 1.4998096688237535e-06,
      "loss": 1.2782,
      "step": 2147
    },
    {
      "epoch": 2.206471494607088,
      "grad_norm": 3.2130300998687744,
      "learning_rate": 1.4979063570612868e-06,
      "loss": 1.3625,
      "step": 2148
    },
    {
      "epoch": 2.2074987159732924,
      "grad_norm": 3.250370979309082,
      "learning_rate": 1.49600304529882e-06,
      "loss": 1.3396,
      "step": 2149
    },
    {
      "epoch": 2.2085259373394965,
      "grad_norm": 3.0954599380493164,
      "learning_rate": 1.4940997335363533e-06,
      "loss": 1.3187,
      "step": 2150
    },
    {
      "epoch": 2.2085259373394965,
      "eval_loss": 1.8469802141189575,
      "eval_runtime": 37.2109,
      "eval_samples_per_second": 26.874,
      "eval_steps_per_second": 1.129,
      "step": 2150
    },
    {
      "epoch": 2.209553158705701,
      "grad_norm": 3.153545379638672,
      "learning_rate": 1.4921964217738865e-06,
      "loss": 1.3488,
      "step": 2151
    },
    {
      "epoch": 2.2105803800719057,
      "grad_norm": 3.1206071376800537,
      "learning_rate": 1.49029311001142e-06,
      "loss": 1.336,
      "step": 2152
    },
    {
      "epoch": 2.21160760143811,
      "grad_norm": 3.151784896850586,
      "learning_rate": 1.4883897982489534e-06,
      "loss": 1.3613,
      "step": 2153
    },
    {
      "epoch": 2.2126348228043144,
      "grad_norm": 3.18965744972229,
      "learning_rate": 1.4864864864864868e-06,
      "loss": 1.3698,
      "step": 2154
    },
    {
      "epoch": 2.2136620441705186,
      "grad_norm": 3.108281135559082,
      "learning_rate": 1.48458317472402e-06,
      "loss": 1.3287,
      "step": 2155
    },
    {
      "epoch": 2.214689265536723,
      "grad_norm": 3.2309465408325195,
      "learning_rate": 1.4826798629615533e-06,
      "loss": 1.3425,
      "step": 2156
    },
    {
      "epoch": 2.2157164869029278,
      "grad_norm": 3.0829577445983887,
      "learning_rate": 1.4807765511990864e-06,
      "loss": 1.2917,
      "step": 2157
    },
    {
      "epoch": 2.216743708269132,
      "grad_norm": 3.1417033672332764,
      "learning_rate": 1.4788732394366198e-06,
      "loss": 1.3113,
      "step": 2158
    },
    {
      "epoch": 2.2177709296353365,
      "grad_norm": 3.1996822357177734,
      "learning_rate": 1.4769699276741532e-06,
      "loss": 1.3248,
      "step": 2159
    },
    {
      "epoch": 2.2187981510015407,
      "grad_norm": 3.262814998626709,
      "learning_rate": 1.4750666159116863e-06,
      "loss": 1.3186,
      "step": 2160
    },
    {
      "epoch": 2.2187981510015407,
      "eval_loss": 1.844687819480896,
      "eval_runtime": 37.2211,
      "eval_samples_per_second": 26.866,
      "eval_steps_per_second": 1.128,
      "step": 2160
    },
    {
      "epoch": 2.2198253723677452,
      "grad_norm": 3.0836942195892334,
      "learning_rate": 1.4731633041492197e-06,
      "loss": 1.2935,
      "step": 2161
    },
    {
      "epoch": 2.22085259373395,
      "grad_norm": 3.1720023155212402,
      "learning_rate": 1.4712599923867528e-06,
      "loss": 1.3622,
      "step": 2162
    },
    {
      "epoch": 2.221879815100154,
      "grad_norm": 3.2829599380493164,
      "learning_rate": 1.4693566806242864e-06,
      "loss": 1.3794,
      "step": 2163
    },
    {
      "epoch": 2.2229070364663586,
      "grad_norm": 3.045949935913086,
      "learning_rate": 1.4674533688618198e-06,
      "loss": 1.3054,
      "step": 2164
    },
    {
      "epoch": 2.2239342578325627,
      "grad_norm": 3.1561572551727295,
      "learning_rate": 1.4655500570993531e-06,
      "loss": 1.3289,
      "step": 2165
    },
    {
      "epoch": 2.2249614791987673,
      "grad_norm": 3.0877439975738525,
      "learning_rate": 1.4636467453368863e-06,
      "loss": 1.329,
      "step": 2166
    },
    {
      "epoch": 2.225988700564972,
      "grad_norm": 3.1330180168151855,
      "learning_rate": 1.4617434335744197e-06,
      "loss": 1.3853,
      "step": 2167
    },
    {
      "epoch": 2.227015921931176,
      "grad_norm": 3.231693744659424,
      "learning_rate": 1.4598401218119528e-06,
      "loss": 1.378,
      "step": 2168
    },
    {
      "epoch": 2.2280431432973806,
      "grad_norm": 3.202930212020874,
      "learning_rate": 1.4579368100494862e-06,
      "loss": 1.3549,
      "step": 2169
    },
    {
      "epoch": 2.2290703646635848,
      "grad_norm": 3.164461135864258,
      "learning_rate": 1.4560334982870195e-06,
      "loss": 1.3365,
      "step": 2170
    },
    {
      "epoch": 2.2290703646635848,
      "eval_loss": 1.8483937978744507,
      "eval_runtime": 37.1839,
      "eval_samples_per_second": 26.893,
      "eval_steps_per_second": 1.13,
      "step": 2170
    },
    {
      "epoch": 2.2300975860297894,
      "grad_norm": 3.1866867542266846,
      "learning_rate": 1.4541301865245527e-06,
      "loss": 1.3503,
      "step": 2171
    },
    {
      "epoch": 2.231124807395994,
      "grad_norm": 3.2202508449554443,
      "learning_rate": 1.452226874762086e-06,
      "loss": 1.354,
      "step": 2172
    },
    {
      "epoch": 2.232152028762198,
      "grad_norm": 3.190744161605835,
      "learning_rate": 1.4503235629996196e-06,
      "loss": 1.3073,
      "step": 2173
    },
    {
      "epoch": 2.2331792501284027,
      "grad_norm": 3.2158405780792236,
      "learning_rate": 1.4484202512371528e-06,
      "loss": 1.3731,
      "step": 2174
    },
    {
      "epoch": 2.2342064714946073,
      "grad_norm": 3.0294954776763916,
      "learning_rate": 1.4465169394746861e-06,
      "loss": 1.2772,
      "step": 2175
    },
    {
      "epoch": 2.2352336928608114,
      "grad_norm": 3.2871217727661133,
      "learning_rate": 1.4446136277122195e-06,
      "loss": 1.3888,
      "step": 2176
    },
    {
      "epoch": 2.236260914227016,
      "grad_norm": 3.0494141578674316,
      "learning_rate": 1.4427103159497526e-06,
      "loss": 1.2991,
      "step": 2177
    },
    {
      "epoch": 2.23728813559322,
      "grad_norm": 3.122699022293091,
      "learning_rate": 1.440807004187286e-06,
      "loss": 1.3186,
      "step": 2178
    },
    {
      "epoch": 2.2383153569594247,
      "grad_norm": 3.119854211807251,
      "learning_rate": 1.4389036924248192e-06,
      "loss": 1.3421,
      "step": 2179
    },
    {
      "epoch": 2.2393425783256293,
      "grad_norm": 3.2588119506835938,
      "learning_rate": 1.4370003806623525e-06,
      "loss": 1.3657,
      "step": 2180
    },
    {
      "epoch": 2.2393425783256293,
      "eval_loss": 1.8473975658416748,
      "eval_runtime": 37.1561,
      "eval_samples_per_second": 26.914,
      "eval_steps_per_second": 1.13,
      "step": 2180
    },
    {
      "epoch": 2.2403697996918335,
      "grad_norm": 3.138901948928833,
      "learning_rate": 1.4350970688998859e-06,
      "loss": 1.3273,
      "step": 2181
    },
    {
      "epoch": 2.241397021058038,
      "grad_norm": 3.2792508602142334,
      "learning_rate": 1.433193757137419e-06,
      "loss": 1.3763,
      "step": 2182
    },
    {
      "epoch": 2.242424242424242,
      "grad_norm": 3.065627098083496,
      "learning_rate": 1.4312904453749526e-06,
      "loss": 1.3292,
      "step": 2183
    },
    {
      "epoch": 2.243451463790447,
      "grad_norm": 3.101377010345459,
      "learning_rate": 1.429387133612486e-06,
      "loss": 1.2957,
      "step": 2184
    },
    {
      "epoch": 2.2444786851566514,
      "grad_norm": 3.140158176422119,
      "learning_rate": 1.4274838218500191e-06,
      "loss": 1.3406,
      "step": 2185
    },
    {
      "epoch": 2.2455059065228555,
      "grad_norm": 3.2498116493225098,
      "learning_rate": 1.4255805100875525e-06,
      "loss": 1.3499,
      "step": 2186
    },
    {
      "epoch": 2.24653312788906,
      "grad_norm": 3.2681965827941895,
      "learning_rate": 1.4236771983250858e-06,
      "loss": 1.407,
      "step": 2187
    },
    {
      "epoch": 2.2475603492552647,
      "grad_norm": 3.2457778453826904,
      "learning_rate": 1.421773886562619e-06,
      "loss": 1.3278,
      "step": 2188
    },
    {
      "epoch": 2.248587570621469,
      "grad_norm": 3.099942207336426,
      "learning_rate": 1.4198705748001524e-06,
      "loss": 1.3114,
      "step": 2189
    },
    {
      "epoch": 2.2496147919876734,
      "grad_norm": 3.090271234512329,
      "learning_rate": 1.4179672630376855e-06,
      "loss": 1.3285,
      "step": 2190
    },
    {
      "epoch": 2.2496147919876734,
      "eval_loss": 1.8479118347167969,
      "eval_runtime": 37.2331,
      "eval_samples_per_second": 26.858,
      "eval_steps_per_second": 1.128,
      "step": 2190
    },
    {
      "epoch": 2.2506420133538776,
      "grad_norm": 3.0344860553741455,
      "learning_rate": 1.4160639512752189e-06,
      "loss": 1.3104,
      "step": 2191
    },
    {
      "epoch": 2.251669234720082,
      "grad_norm": 3.087860584259033,
      "learning_rate": 1.4141606395127522e-06,
      "loss": 1.3167,
      "step": 2192
    },
    {
      "epoch": 2.2526964560862868,
      "grad_norm": 3.0974831581115723,
      "learning_rate": 1.4122573277502858e-06,
      "loss": 1.2926,
      "step": 2193
    },
    {
      "epoch": 2.253723677452491,
      "grad_norm": 3.228492498397827,
      "learning_rate": 1.410354015987819e-06,
      "loss": 1.3541,
      "step": 2194
    },
    {
      "epoch": 2.2547508988186955,
      "grad_norm": 3.1761956214904785,
      "learning_rate": 1.4084507042253523e-06,
      "loss": 1.3342,
      "step": 2195
    },
    {
      "epoch": 2.2557781201848996,
      "grad_norm": 3.1890695095062256,
      "learning_rate": 1.4065473924628855e-06,
      "loss": 1.362,
      "step": 2196
    },
    {
      "epoch": 2.2568053415511042,
      "grad_norm": 3.1543149948120117,
      "learning_rate": 1.4046440807004188e-06,
      "loss": 1.329,
      "step": 2197
    },
    {
      "epoch": 2.257832562917309,
      "grad_norm": 3.2313342094421387,
      "learning_rate": 1.4027407689379522e-06,
      "loss": 1.3256,
      "step": 2198
    },
    {
      "epoch": 2.258859784283513,
      "grad_norm": 3.130805253982544,
      "learning_rate": 1.4008374571754854e-06,
      "loss": 1.3047,
      "step": 2199
    },
    {
      "epoch": 2.2598870056497176,
      "grad_norm": 3.0488216876983643,
      "learning_rate": 1.3989341454130187e-06,
      "loss": 1.3461,
      "step": 2200
    },
    {
      "epoch": 2.2598870056497176,
      "eval_loss": 1.8517944812774658,
      "eval_runtime": 37.4528,
      "eval_samples_per_second": 26.7,
      "eval_steps_per_second": 1.121,
      "step": 2200
    },
    {
      "epoch": 2.260914227015922,
      "grad_norm": 3.0915918350219727,
      "learning_rate": 1.397030833650552e-06,
      "loss": 1.3391,
      "step": 2201
    },
    {
      "epoch": 2.2619414483821263,
      "grad_norm": 3.156053066253662,
      "learning_rate": 1.3951275218880852e-06,
      "loss": 1.3525,
      "step": 2202
    },
    {
      "epoch": 2.262968669748331,
      "grad_norm": 3.4035868644714355,
      "learning_rate": 1.3932242101256188e-06,
      "loss": 1.4122,
      "step": 2203
    },
    {
      "epoch": 2.263995891114535,
      "grad_norm": 2.9787867069244385,
      "learning_rate": 1.3913208983631522e-06,
      "loss": 1.3096,
      "step": 2204
    },
    {
      "epoch": 2.2650231124807396,
      "grad_norm": 3.154271364212036,
      "learning_rate": 1.3894175866006853e-06,
      "loss": 1.3421,
      "step": 2205
    },
    {
      "epoch": 2.266050333846944,
      "grad_norm": 3.1899831295013428,
      "learning_rate": 1.3875142748382187e-06,
      "loss": 1.2949,
      "step": 2206
    },
    {
      "epoch": 2.2670775552131484,
      "grad_norm": 3.126694679260254,
      "learning_rate": 1.385610963075752e-06,
      "loss": 1.2898,
      "step": 2207
    },
    {
      "epoch": 2.268104776579353,
      "grad_norm": 3.149989128112793,
      "learning_rate": 1.3837076513132852e-06,
      "loss": 1.3369,
      "step": 2208
    },
    {
      "epoch": 2.269131997945557,
      "grad_norm": 3.250481128692627,
      "learning_rate": 1.3818043395508186e-06,
      "loss": 1.3605,
      "step": 2209
    },
    {
      "epoch": 2.2701592193117617,
      "grad_norm": 3.2549924850463867,
      "learning_rate": 1.3799010277883517e-06,
      "loss": 1.3428,
      "step": 2210
    },
    {
      "epoch": 2.2701592193117617,
      "eval_loss": 1.8562498092651367,
      "eval_runtime": 37.0713,
      "eval_samples_per_second": 26.975,
      "eval_steps_per_second": 1.133,
      "step": 2210
    },
    {
      "epoch": 2.2711864406779663,
      "grad_norm": 3.1369433403015137,
      "learning_rate": 1.377997716025885e-06,
      "loss": 1.333,
      "step": 2211
    },
    {
      "epoch": 2.2722136620441704,
      "grad_norm": 3.2522025108337402,
      "learning_rate": 1.3760944042634184e-06,
      "loss": 1.3496,
      "step": 2212
    },
    {
      "epoch": 2.273240883410375,
      "grad_norm": 3.225821018218994,
      "learning_rate": 1.3741910925009518e-06,
      "loss": 1.3716,
      "step": 2213
    },
    {
      "epoch": 2.2742681047765796,
      "grad_norm": 3.1488454341888428,
      "learning_rate": 1.3722877807384852e-06,
      "loss": 1.3508,
      "step": 2214
    },
    {
      "epoch": 2.2752953261427837,
      "grad_norm": 3.258491039276123,
      "learning_rate": 1.3703844689760185e-06,
      "loss": 1.3565,
      "step": 2215
    },
    {
      "epoch": 2.2763225475089883,
      "grad_norm": 3.1634750366210938,
      "learning_rate": 1.3684811572135517e-06,
      "loss": 1.3269,
      "step": 2216
    },
    {
      "epoch": 2.2773497688751925,
      "grad_norm": 3.1604928970336914,
      "learning_rate": 1.366577845451085e-06,
      "loss": 1.2832,
      "step": 2217
    },
    {
      "epoch": 2.278376990241397,
      "grad_norm": 3.2551920413970947,
      "learning_rate": 1.3646745336886184e-06,
      "loss": 1.3567,
      "step": 2218
    },
    {
      "epoch": 2.279404211607601,
      "grad_norm": 3.392820119857788,
      "learning_rate": 1.3627712219261515e-06,
      "loss": 1.3601,
      "step": 2219
    },
    {
      "epoch": 2.280431432973806,
      "grad_norm": 3.057724952697754,
      "learning_rate": 1.360867910163685e-06,
      "loss": 1.3138,
      "step": 2220
    },
    {
      "epoch": 2.280431432973806,
      "eval_loss": 1.848382830619812,
      "eval_runtime": 37.2255,
      "eval_samples_per_second": 26.863,
      "eval_steps_per_second": 1.128,
      "step": 2220
    },
    {
      "epoch": 2.2814586543400104,
      "grad_norm": 3.2199649810791016,
      "learning_rate": 1.358964598401218e-06,
      "loss": 1.3338,
      "step": 2221
    },
    {
      "epoch": 2.2824858757062145,
      "grad_norm": 3.252066135406494,
      "learning_rate": 1.3570612866387514e-06,
      "loss": 1.3413,
      "step": 2222
    },
    {
      "epoch": 2.283513097072419,
      "grad_norm": 3.038052558898926,
      "learning_rate": 1.355157974876285e-06,
      "loss": 1.3268,
      "step": 2223
    },
    {
      "epoch": 2.2845403184386237,
      "grad_norm": 3.2332992553710938,
      "learning_rate": 1.3532546631138184e-06,
      "loss": 1.3866,
      "step": 2224
    },
    {
      "epoch": 2.285567539804828,
      "grad_norm": 3.14321231842041,
      "learning_rate": 1.3513513513513515e-06,
      "loss": 1.3466,
      "step": 2225
    },
    {
      "epoch": 2.2865947611710324,
      "grad_norm": 3.0714094638824463,
      "learning_rate": 1.3494480395888849e-06,
      "loss": 1.331,
      "step": 2226
    },
    {
      "epoch": 2.2876219825372366,
      "grad_norm": 3.1167569160461426,
      "learning_rate": 1.347544727826418e-06,
      "loss": 1.3242,
      "step": 2227
    },
    {
      "epoch": 2.288649203903441,
      "grad_norm": 3.072951316833496,
      "learning_rate": 1.3456414160639514e-06,
      "loss": 1.3139,
      "step": 2228
    },
    {
      "epoch": 2.2896764252696458,
      "grad_norm": 3.104212999343872,
      "learning_rate": 1.3437381043014847e-06,
      "loss": 1.3524,
      "step": 2229
    },
    {
      "epoch": 2.29070364663585,
      "grad_norm": 3.104212999343872,
      "learning_rate": 1.3437381043014847e-06,
      "loss": 1.334,
      "step": 2230
    },
    {
      "epoch": 2.29070364663585,
      "eval_loss": 1.848712682723999,
      "eval_runtime": 37.2687,
      "eval_samples_per_second": 26.832,
      "eval_steps_per_second": 1.127,
      "step": 2230
    },
    {
      "epoch": 2.2917308680020545,
      "grad_norm": 3.056426525115967,
      "learning_rate": 1.341834792539018e-06,
      "loss": 1.3344,
      "step": 2231
    },
    {
      "epoch": 2.2927580893682586,
      "grad_norm": 3.2412500381469727,
      "learning_rate": 1.3399314807765513e-06,
      "loss": 1.3416,
      "step": 2232
    },
    {
      "epoch": 2.2937853107344632,
      "grad_norm": 3.178572416305542,
      "learning_rate": 1.3380281690140844e-06,
      "loss": 1.3472,
      "step": 2233
    },
    {
      "epoch": 2.294812532100668,
      "grad_norm": 3.2567203044891357,
      "learning_rate": 1.336124857251618e-06,
      "loss": 1.311,
      "step": 2234
    },
    {
      "epoch": 2.295839753466872,
      "grad_norm": 3.2018561363220215,
      "learning_rate": 1.3342215454891513e-06,
      "loss": 1.301,
      "step": 2235
    },
    {
      "epoch": 2.2968669748330766,
      "grad_norm": 3.0722200870513916,
      "learning_rate": 1.3323182337266847e-06,
      "loss": 1.3325,
      "step": 2236
    },
    {
      "epoch": 2.297894196199281,
      "grad_norm": 3.2170779705047607,
      "learning_rate": 1.3304149219642179e-06,
      "loss": 1.3752,
      "step": 2237
    },
    {
      "epoch": 2.2989214175654853,
      "grad_norm": 3.1309115886688232,
      "learning_rate": 1.3285116102017512e-06,
      "loss": 1.3261,
      "step": 2238
    },
    {
      "epoch": 2.29994863893169,
      "grad_norm": 3.1463022232055664,
      "learning_rate": 1.3266082984392844e-06,
      "loss": 1.3316,
      "step": 2239
    },
    {
      "epoch": 2.300975860297894,
      "grad_norm": 3.174701690673828,
      "learning_rate": 1.3247049866768177e-06,
      "loss": 1.3148,
      "step": 2240
    },
    {
      "epoch": 2.300975860297894,
      "eval_loss": 1.8436999320983887,
      "eval_runtime": 37.2577,
      "eval_samples_per_second": 26.84,
      "eval_steps_per_second": 1.127,
      "step": 2240
    },
    {
      "epoch": 2.3020030816640986,
      "grad_norm": 3.18149471282959,
      "learning_rate": 1.322801674914351e-06,
      "loss": 1.312,
      "step": 2241
    },
    {
      "epoch": 2.303030303030303,
      "grad_norm": 3.213361978530884,
      "learning_rate": 1.3208983631518843e-06,
      "loss": 1.3406,
      "step": 2242
    },
    {
      "epoch": 2.3040575243965074,
      "grad_norm": 3.128354787826538,
      "learning_rate": 1.3189950513894176e-06,
      "loss": 1.3286,
      "step": 2243
    },
    {
      "epoch": 2.305084745762712,
      "grad_norm": 3.210330009460449,
      "learning_rate": 1.3170917396269512e-06,
      "loss": 1.3406,
      "step": 2244
    },
    {
      "epoch": 2.306111967128916,
      "grad_norm": 3.187161445617676,
      "learning_rate": 1.3151884278644843e-06,
      "loss": 1.3621,
      "step": 2245
    },
    {
      "epoch": 2.3071391884951207,
      "grad_norm": 3.108806848526001,
      "learning_rate": 1.3132851161020177e-06,
      "loss": 1.3094,
      "step": 2246
    },
    {
      "epoch": 2.3081664098613253,
      "grad_norm": 3.336510181427002,
      "learning_rate": 1.311381804339551e-06,
      "loss": 1.3998,
      "step": 2247
    },
    {
      "epoch": 2.3091936312275294,
      "grad_norm": 3.159184455871582,
      "learning_rate": 1.3094784925770842e-06,
      "loss": 1.3082,
      "step": 2248
    },
    {
      "epoch": 2.310220852593734,
      "grad_norm": 3.128321409225464,
      "learning_rate": 1.3075751808146176e-06,
      "loss": 1.3494,
      "step": 2249
    },
    {
      "epoch": 2.3112480739599386,
      "grad_norm": 3.122690439224243,
      "learning_rate": 1.3056718690521507e-06,
      "loss": 1.3064,
      "step": 2250
    },
    {
      "epoch": 2.3112480739599386,
      "eval_loss": 1.848455786705017,
      "eval_runtime": 37.2647,
      "eval_samples_per_second": 26.835,
      "eval_steps_per_second": 1.127,
      "step": 2250
    },
    {
      "epoch": 2.3122752953261427,
      "grad_norm": 3.1642186641693115,
      "learning_rate": 1.303768557289684e-06,
      "loss": 1.3134,
      "step": 2251
    },
    {
      "epoch": 2.3133025166923473,
      "grad_norm": 3.1710383892059326,
      "learning_rate": 1.3018652455272175e-06,
      "loss": 1.3126,
      "step": 2252
    },
    {
      "epoch": 2.3143297380585515,
      "grad_norm": 3.1513214111328125,
      "learning_rate": 1.2999619337647506e-06,
      "loss": 1.3218,
      "step": 2253
    },
    {
      "epoch": 2.315356959424756,
      "grad_norm": 3.1538784503936768,
      "learning_rate": 1.2980586220022842e-06,
      "loss": 1.3473,
      "step": 2254
    },
    {
      "epoch": 2.3163841807909606,
      "grad_norm": 3.2409846782684326,
      "learning_rate": 1.2961553102398175e-06,
      "loss": 1.371,
      "step": 2255
    },
    {
      "epoch": 2.317411402157165,
      "grad_norm": 3.3224987983703613,
      "learning_rate": 1.2942519984773507e-06,
      "loss": 1.3506,
      "step": 2256
    },
    {
      "epoch": 2.3184386235233694,
      "grad_norm": 3.1684653759002686,
      "learning_rate": 1.292348686714884e-06,
      "loss": 1.3127,
      "step": 2257
    },
    {
      "epoch": 2.3194658448895735,
      "grad_norm": 3.1784262657165527,
      "learning_rate": 1.2904453749524174e-06,
      "loss": 1.3541,
      "step": 2258
    },
    {
      "epoch": 2.320493066255778,
      "grad_norm": 3.1121561527252197,
      "learning_rate": 1.2885420631899506e-06,
      "loss": 1.2843,
      "step": 2259
    },
    {
      "epoch": 2.3215202876219827,
      "grad_norm": 3.200490951538086,
      "learning_rate": 1.286638751427484e-06,
      "loss": 1.3191,
      "step": 2260
    },
    {
      "epoch": 2.3215202876219827,
      "eval_loss": 1.8469436168670654,
      "eval_runtime": 37.2695,
      "eval_samples_per_second": 26.832,
      "eval_steps_per_second": 1.127,
      "step": 2260
    },
    {
      "epoch": 2.322547508988187,
      "grad_norm": 3.143329620361328,
      "learning_rate": 1.284735439665017e-06,
      "loss": 1.3454,
      "step": 2261
    },
    {
      "epoch": 2.3235747303543914,
      "grad_norm": 3.1955461502075195,
      "learning_rate": 1.2828321279025504e-06,
      "loss": 1.3505,
      "step": 2262
    },
    {
      "epoch": 2.324601951720596,
      "grad_norm": 3.3297348022460938,
      "learning_rate": 1.2809288161400838e-06,
      "loss": 1.3374,
      "step": 2263
    },
    {
      "epoch": 2.3256291730868,
      "grad_norm": 3.144385814666748,
      "learning_rate": 1.279025504377617e-06,
      "loss": 1.3198,
      "step": 2264
    },
    {
      "epoch": 2.3266563944530048,
      "grad_norm": 3.2952210903167725,
      "learning_rate": 1.2771221926151505e-06,
      "loss": 1.3744,
      "step": 2265
    },
    {
      "epoch": 2.327683615819209,
      "grad_norm": 3.212287187576294,
      "learning_rate": 1.275218880852684e-06,
      "loss": 1.3173,
      "step": 2266
    },
    {
      "epoch": 2.3287108371854135,
      "grad_norm": 3.235272169113159,
      "learning_rate": 1.273315569090217e-06,
      "loss": 1.3231,
      "step": 2267
    },
    {
      "epoch": 2.3297380585516176,
      "grad_norm": 3.2961113452911377,
      "learning_rate": 1.2714122573277504e-06,
      "loss": 1.3909,
      "step": 2268
    },
    {
      "epoch": 2.3307652799178222,
      "grad_norm": 3.13678240776062,
      "learning_rate": 1.2695089455652838e-06,
      "loss": 1.3911,
      "step": 2269
    },
    {
      "epoch": 2.331792501284027,
      "grad_norm": 3.166139841079712,
      "learning_rate": 1.267605633802817e-06,
      "loss": 1.3496,
      "step": 2270
    },
    {
      "epoch": 2.331792501284027,
      "eval_loss": 1.8424044847488403,
      "eval_runtime": 37.128,
      "eval_samples_per_second": 26.934,
      "eval_steps_per_second": 1.131,
      "step": 2270
    },
    {
      "epoch": 2.332819722650231,
      "grad_norm": 3.20859432220459,
      "learning_rate": 1.2657023220403503e-06,
      "loss": 1.3402,
      "step": 2271
    },
    {
      "epoch": 2.3338469440164356,
      "grad_norm": 3.18941593170166,
      "learning_rate": 1.2637990102778836e-06,
      "loss": 1.3272,
      "step": 2272
    },
    {
      "epoch": 2.33487416538264,
      "grad_norm": 3.2382333278656006,
      "learning_rate": 1.2618956985154168e-06,
      "loss": 1.3242,
      "step": 2273
    },
    {
      "epoch": 2.3359013867488443,
      "grad_norm": 3.2496516704559326,
      "learning_rate": 1.2599923867529502e-06,
      "loss": 1.3675,
      "step": 2274
    },
    {
      "epoch": 2.336928608115049,
      "grad_norm": 3.1282081604003906,
      "learning_rate": 1.2580890749904837e-06,
      "loss": 1.3343,
      "step": 2275
    },
    {
      "epoch": 2.337955829481253,
      "grad_norm": 3.34047794342041,
      "learning_rate": 1.2561857632280169e-06,
      "loss": 1.3651,
      "step": 2276
    },
    {
      "epoch": 2.3389830508474576,
      "grad_norm": 3.3369057178497314,
      "learning_rate": 1.2542824514655502e-06,
      "loss": 1.3944,
      "step": 2277
    },
    {
      "epoch": 2.340010272213662,
      "grad_norm": 3.282226085662842,
      "learning_rate": 1.2523791397030834e-06,
      "loss": 1.3595,
      "step": 2278
    },
    {
      "epoch": 2.3410374935798663,
      "grad_norm": 3.111595630645752,
      "learning_rate": 1.2504758279406168e-06,
      "loss": 1.3325,
      "step": 2279
    },
    {
      "epoch": 2.342064714946071,
      "grad_norm": 3.306988477706909,
      "learning_rate": 1.2485725161781501e-06,
      "loss": 1.3595,
      "step": 2280
    },
    {
      "epoch": 2.342064714946071,
      "eval_loss": 1.8465285301208496,
      "eval_runtime": 37.1211,
      "eval_samples_per_second": 26.939,
      "eval_steps_per_second": 1.131,
      "step": 2280
    },
    {
      "epoch": 2.343091936312275,
      "grad_norm": 3.4089088439941406,
      "learning_rate": 1.2466692044156833e-06,
      "loss": 1.4337,
      "step": 2281
    },
    {
      "epoch": 2.3441191576784797,
      "grad_norm": 3.1354665756225586,
      "learning_rate": 1.2447658926532168e-06,
      "loss": 1.2996,
      "step": 2282
    },
    {
      "epoch": 2.3451463790446843,
      "grad_norm": 3.067383289337158,
      "learning_rate": 1.24286258089075e-06,
      "loss": 1.3446,
      "step": 2283
    },
    {
      "epoch": 2.3461736004108884,
      "grad_norm": 3.162349224090576,
      "learning_rate": 1.2409592691282834e-06,
      "loss": 1.3068,
      "step": 2284
    },
    {
      "epoch": 2.347200821777093,
      "grad_norm": 3.2846035957336426,
      "learning_rate": 1.2390559573658165e-06,
      "loss": 1.3496,
      "step": 2285
    },
    {
      "epoch": 2.3482280431432976,
      "grad_norm": 3.290997266769409,
      "learning_rate": 1.2371526456033499e-06,
      "loss": 1.3471,
      "step": 2286
    },
    {
      "epoch": 2.3492552645095017,
      "grad_norm": 3.1327056884765625,
      "learning_rate": 1.2352493338408832e-06,
      "loss": 1.3108,
      "step": 2287
    },
    {
      "epoch": 2.3502824858757063,
      "grad_norm": 3.295562744140625,
      "learning_rate": 1.2333460220784166e-06,
      "loss": 1.4226,
      "step": 2288
    },
    {
      "epoch": 2.3513097072419105,
      "grad_norm": 3.3461313247680664,
      "learning_rate": 1.23144271031595e-06,
      "loss": 1.3494,
      "step": 2289
    },
    {
      "epoch": 2.352336928608115,
      "grad_norm": 3.4178872108459473,
      "learning_rate": 1.2295393985534831e-06,
      "loss": 1.3876,
      "step": 2290
    },
    {
      "epoch": 2.352336928608115,
      "eval_loss": 1.855657935142517,
      "eval_runtime": 37.2052,
      "eval_samples_per_second": 26.878,
      "eval_steps_per_second": 1.129,
      "step": 2290
    },
    {
      "epoch": 2.3533641499743196,
      "grad_norm": 3.0909383296966553,
      "learning_rate": 1.2276360867910165e-06,
      "loss": 1.2625,
      "step": 2291
    },
    {
      "epoch": 2.354391371340524,
      "grad_norm": 3.2940239906311035,
      "learning_rate": 1.2257327750285498e-06,
      "loss": 1.3836,
      "step": 2292
    },
    {
      "epoch": 2.3554185927067284,
      "grad_norm": 3.166198253631592,
      "learning_rate": 1.2238294632660832e-06,
      "loss": 1.3014,
      "step": 2293
    },
    {
      "epoch": 2.3564458140729325,
      "grad_norm": 3.1372265815734863,
      "learning_rate": 1.2219261515036164e-06,
      "loss": 1.3133,
      "step": 2294
    },
    {
      "epoch": 2.357473035439137,
      "grad_norm": 3.2904369831085205,
      "learning_rate": 1.2200228397411497e-06,
      "loss": 1.3578,
      "step": 2295
    },
    {
      "epoch": 2.3585002568053417,
      "grad_norm": 3.1268763542175293,
      "learning_rate": 1.2181195279786829e-06,
      "loss": 1.391,
      "step": 2296
    },
    {
      "epoch": 2.359527478171546,
      "grad_norm": 3.1357498168945312,
      "learning_rate": 1.2162162162162164e-06,
      "loss": 1.327,
      "step": 2297
    },
    {
      "epoch": 2.3605546995377504,
      "grad_norm": 3.1766469478607178,
      "learning_rate": 1.2143129044537496e-06,
      "loss": 1.3146,
      "step": 2298
    },
    {
      "epoch": 2.361581920903955,
      "grad_norm": 3.1096513271331787,
      "learning_rate": 1.212409592691283e-06,
      "loss": 1.3235,
      "step": 2299
    },
    {
      "epoch": 2.362609142270159,
      "grad_norm": 3.161182403564453,
      "learning_rate": 1.2105062809288163e-06,
      "loss": 1.3411,
      "step": 2300
    },
    {
      "epoch": 2.362609142270159,
      "eval_loss": 1.852128267288208,
      "eval_runtime": 37.4484,
      "eval_samples_per_second": 26.703,
      "eval_steps_per_second": 1.122,
      "step": 2300
    }
  ],
  "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": 7.169811641532416e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}