{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.750912567356162,
  "eval_steps": 150,
  "global_step": 2160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006952894142186686,
      "grad_norm": 5.413117383066536,
      "learning_rate": 6.944444444444445e-07,
      "loss": 2.3653,
      "step": 2
    },
    {
      "epoch": 0.0013905788284373371,
      "grad_norm": 5.092870612337868,
      "learning_rate": 1.388888888888889e-06,
      "loss": 2.5001,
      "step": 4
    },
    {
      "epoch": 0.0020858682426560054,
      "grad_norm": 3.6285625961711943,
      "learning_rate": 2.0833333333333334e-06,
      "loss": 2.3759,
      "step": 6
    },
    {
      "epoch": 0.0027811576568746743,
      "grad_norm": 2.5025911526151075,
      "learning_rate": 2.777777777777778e-06,
      "loss": 1.944,
      "step": 8
    },
    {
      "epoch": 0.0034764470710933427,
      "grad_norm": 3.077266049542496,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 2.2504,
      "step": 10
    },
    {
      "epoch": 0.004171736485312011,
      "grad_norm": 4.068243939187174,
      "learning_rate": 4.166666666666667e-06,
      "loss": 2.0637,
      "step": 12
    },
    {
      "epoch": 0.00486702589953068,
      "grad_norm": 3.511444738830971,
      "learning_rate": 4.861111111111111e-06,
      "loss": 2.3604,
      "step": 14
    },
    {
      "epoch": 0.0055623153137493485,
      "grad_norm": 5.925744892256934,
      "learning_rate": 5.555555555555556e-06,
      "loss": 2.6528,
      "step": 16
    },
    {
      "epoch": 0.0062576047279680165,
      "grad_norm": 2.7950114871483405,
      "learning_rate": 6.25e-06,
      "loss": 2.3229,
      "step": 18
    },
    {
      "epoch": 0.006952894142186685,
      "grad_norm": 6.586211843766182,
      "learning_rate": 6.944444444444445e-06,
      "loss": 2.342,
      "step": 20
    },
    {
      "epoch": 0.0076481835564053535,
      "grad_norm": 4.902839955269193,
      "learning_rate": 7.63888888888889e-06,
      "loss": 2.4188,
      "step": 22
    },
    {
      "epoch": 0.008343472970624021,
      "grad_norm": 4.257062809771645,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.7957,
      "step": 24
    },
    {
      "epoch": 0.009038762384842691,
      "grad_norm": 4.460352004615699,
      "learning_rate": 9.027777777777777e-06,
      "loss": 2.0726,
      "step": 26
    },
    {
      "epoch": 0.00973405179906136,
      "grad_norm": 5.858061506133739,
      "learning_rate": 9.722222222222223e-06,
      "loss": 2.0455,
      "step": 28
    },
    {
      "epoch": 0.010429341213280027,
      "grad_norm": 4.331946668100709,
      "learning_rate": 1.0416666666666668e-05,
      "loss": 1.7645,
      "step": 30
    },
    {
      "epoch": 0.011124630627498697,
      "grad_norm": 5.428742204187391,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 1.7446,
      "step": 32
    },
    {
      "epoch": 0.011819920041717365,
      "grad_norm": 2.485909286541028,
      "learning_rate": 1.1805555555555555e-05,
      "loss": 1.5885,
      "step": 34
    },
    {
      "epoch": 0.012515209455936033,
      "grad_norm": 3.2602949308730222,
      "learning_rate": 1.25e-05,
      "loss": 1.1014,
      "step": 36
    },
    {
      "epoch": 0.013210498870154701,
      "grad_norm": 4.962187747415964,
      "learning_rate": 1.3194444444444446e-05,
      "loss": 1.3588,
      "step": 38
    },
    {
      "epoch": 0.01390578828437337,
      "grad_norm": 7.627756894198461,
      "learning_rate": 1.388888888888889e-05,
      "loss": 1.4014,
      "step": 40
    },
    {
      "epoch": 0.014601077698592039,
      "grad_norm": 2.821848388410092,
      "learning_rate": 1.4583333333333335e-05,
      "loss": 1.0959,
      "step": 42
    },
    {
      "epoch": 0.015296367112810707,
      "grad_norm": 10.989320944549025,
      "learning_rate": 1.527777777777778e-05,
      "loss": 1.739,
      "step": 44
    },
    {
      "epoch": 0.015991656527029375,
      "grad_norm": 4.300936602591115,
      "learning_rate": 1.597222222222222e-05,
      "loss": 1.1078,
      "step": 46
    },
    {
      "epoch": 0.016686945941248043,
      "grad_norm": 3.7539663261019856,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.2701,
      "step": 48
    },
    {
      "epoch": 0.017382235355466714,
      "grad_norm": 2.589248169352173,
      "learning_rate": 1.736111111111111e-05,
      "loss": 1.4452,
      "step": 50
    },
    {
      "epoch": 0.018077524769685382,
      "grad_norm": 3.6679301322156177,
      "learning_rate": 1.8055555555555555e-05,
      "loss": 1.4243,
      "step": 52
    },
    {
      "epoch": 0.01877281418390405,
      "grad_norm": 2.0885660923860074,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 1.4453,
      "step": 54
    },
    {
      "epoch": 0.01946810359812272,
      "grad_norm": 2.955353237610474,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 1.5388,
      "step": 56
    },
    {
      "epoch": 0.020163393012341387,
      "grad_norm": 3.0527006398487018,
      "learning_rate": 2.013888888888889e-05,
      "loss": 1.4043,
      "step": 58
    },
    {
      "epoch": 0.020858682426560055,
      "grad_norm": 1.9862208864154767,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 1.1007,
      "step": 60
    },
    {
      "epoch": 0.021553971840778723,
      "grad_norm": 2.347361178472164,
      "learning_rate": 2.152777777777778e-05,
      "loss": 0.9291,
      "step": 62
    },
    {
      "epoch": 0.022249261254997394,
      "grad_norm": 2.4812993223105995,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.9592,
      "step": 64
    },
    {
      "epoch": 0.022944550669216062,
      "grad_norm": 1.639333831845777,
      "learning_rate": 2.2916666666666667e-05,
      "loss": 1.0645,
      "step": 66
    },
    {
      "epoch": 0.02363984008343473,
      "grad_norm": 2.92858351082494,
      "learning_rate": 2.361111111111111e-05,
      "loss": 1.4269,
      "step": 68
    },
    {
      "epoch": 0.024335129497653398,
      "grad_norm": 2.8503434812871604,
      "learning_rate": 2.4305555555555558e-05,
      "loss": 0.7829,
      "step": 70
    },
    {
      "epoch": 0.025030418911872066,
      "grad_norm": 2.4875590459354107,
      "learning_rate": 2.5e-05,
      "loss": 0.9733,
      "step": 72
    },
    {
      "epoch": 0.025725708326090734,
      "grad_norm": 5.2567545525905075,
      "learning_rate": 2.5694444444444445e-05,
      "loss": 1.3559,
      "step": 74
    },
    {
      "epoch": 0.026420997740309402,
      "grad_norm": 4.68745753567611,
      "learning_rate": 2.6388888888888892e-05,
      "loss": 1.1499,
      "step": 76
    },
    {
      "epoch": 0.027116287154528074,
      "grad_norm": 2.810345453706711,
      "learning_rate": 2.7083333333333332e-05,
      "loss": 0.8636,
      "step": 78
    },
    {
      "epoch": 0.02781157656874674,
      "grad_norm": 3.0144493626195388,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.9274,
      "step": 80
    },
    {
      "epoch": 0.02850686598296541,
      "grad_norm": 2.3263036535508523,
      "learning_rate": 2.8472222222222223e-05,
      "loss": 1.3979,
      "step": 82
    },
    {
      "epoch": 0.029202155397184078,
      "grad_norm": 1.6009019341419857,
      "learning_rate": 2.916666666666667e-05,
      "loss": 0.9039,
      "step": 84
    },
    {
      "epoch": 0.029897444811402746,
      "grad_norm": 2.0637506891442294,
      "learning_rate": 2.9861111111111113e-05,
      "loss": 1.5187,
      "step": 86
    },
    {
      "epoch": 0.030592734225621414,
      "grad_norm": 1.5533001268209932,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.8937,
      "step": 88
    },
    {
      "epoch": 0.03128802363984008,
      "grad_norm": 1.9974857545819733,
      "learning_rate": 3.125e-05,
      "loss": 0.8218,
      "step": 90
    },
    {
      "epoch": 0.03198331305405875,
      "grad_norm": 1.871899331408953,
      "learning_rate": 3.194444444444444e-05,
      "loss": 1.2451,
      "step": 92
    },
    {
      "epoch": 0.03267860246827742,
      "grad_norm": 1.591601830348497,
      "learning_rate": 3.263888888888889e-05,
      "loss": 0.7485,
      "step": 94
    },
    {
      "epoch": 0.033373891882496086,
      "grad_norm": 2.404166647711005,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.1587,
      "step": 96
    },
    {
      "epoch": 0.03406918129671476,
      "grad_norm": 1.581080788392888,
      "learning_rate": 3.402777777777778e-05,
      "loss": 0.9578,
      "step": 98
    },
    {
      "epoch": 0.03476447071093343,
      "grad_norm": 2.1551207338771547,
      "learning_rate": 3.472222222222222e-05,
      "loss": 1.1305,
      "step": 100
    },
    {
      "epoch": 0.0354597601251521,
      "grad_norm": 4.2246272812704,
      "learning_rate": 3.541666666666667e-05,
      "loss": 1.04,
      "step": 102
    },
    {
      "epoch": 0.036155049539370765,
      "grad_norm": 2.5526898161765588,
      "learning_rate": 3.611111111111111e-05,
      "loss": 1.0743,
      "step": 104
    },
    {
      "epoch": 0.03685033895358943,
      "grad_norm": 1.9784815137623597,
      "learning_rate": 3.6805555555555556e-05,
      "loss": 0.9667,
      "step": 106
    },
    {
      "epoch": 0.0375456283678081,
      "grad_norm": 1.9131648200880944,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.8951,
      "step": 108
    },
    {
      "epoch": 0.03824091778202677,
      "grad_norm": 3.9405296351174575,
      "learning_rate": 3.8194444444444444e-05,
      "loss": 1.222,
      "step": 110
    },
    {
      "epoch": 0.03893620719624544,
      "grad_norm": 13.764861545850291,
      "learning_rate": 3.888888888888889e-05,
      "loss": 1.0497,
      "step": 112
    },
    {
      "epoch": 0.039631496610464105,
      "grad_norm": 3.998123122175411,
      "learning_rate": 3.958333333333333e-05,
      "loss": 1.0901,
      "step": 114
    },
    {
      "epoch": 0.04032678602468277,
      "grad_norm": 3.126193921423756,
      "learning_rate": 4.027777777777778e-05,
      "loss": 1.1143,
      "step": 116
    },
    {
      "epoch": 0.04102207543890144,
      "grad_norm": 2.7839926692610613,
      "learning_rate": 4.0972222222222225e-05,
      "loss": 1.1637,
      "step": 118
    },
    {
      "epoch": 0.04171736485312011,
      "grad_norm": 2.591162621161276,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.0624,
      "step": 120
    },
    {
      "epoch": 0.04241265426733878,
      "grad_norm": 1.4930703711545332,
      "learning_rate": 4.236111111111111e-05,
      "loss": 1.0239,
      "step": 122
    },
    {
      "epoch": 0.043107943681557445,
      "grad_norm": 4.355150070532966,
      "learning_rate": 4.305555555555556e-05,
      "loss": 1.266,
      "step": 124
    },
    {
      "epoch": 0.04380323309577612,
      "grad_norm": 1.8285608876277135,
      "learning_rate": 4.375e-05,
      "loss": 1.0667,
      "step": 126
    },
    {
      "epoch": 0.04449852250999479,
      "grad_norm": 2.1848832742617055,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.7549,
      "step": 128
    },
    {
      "epoch": 0.045193811924213456,
      "grad_norm": 3.547857947451226,
      "learning_rate": 4.5138888888888894e-05,
      "loss": 0.7974,
      "step": 130
    },
    {
      "epoch": 0.045889101338432124,
      "grad_norm": 3.372457028128184,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 1.0482,
      "step": 132
    },
    {
      "epoch": 0.04658439075265079,
      "grad_norm": 2.2871138548091703,
      "learning_rate": 4.652777777777778e-05,
      "loss": 0.8246,
      "step": 134
    },
    {
      "epoch": 0.04727968016686946,
      "grad_norm": 1.7489408054743605,
      "learning_rate": 4.722222222222222e-05,
      "loss": 1.0011,
      "step": 136
    },
    {
      "epoch": 0.04797496958108813,
      "grad_norm": 1.944650349232646,
      "learning_rate": 4.791666666666667e-05,
      "loss": 0.7828,
      "step": 138
    },
    {
      "epoch": 0.048670258995306796,
      "grad_norm": 2.321619459307342,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 1.02,
      "step": 140
    },
    {
      "epoch": 0.049365548409525464,
      "grad_norm": 5.317732559595606,
      "learning_rate": 4.930555555555556e-05,
      "loss": 0.9539,
      "step": 142
    },
    {
      "epoch": 0.05006083782374413,
      "grad_norm": 1.5887292172279854,
      "learning_rate": 5e-05,
      "loss": 0.7961,
      "step": 144
    },
    {
      "epoch": 0.0507561272379628,
      "grad_norm": 2.9987019111854964,
      "learning_rate": 5.069444444444444e-05,
      "loss": 1.0018,
      "step": 146
    },
    {
      "epoch": 0.05145141665218147,
      "grad_norm": 5.63878906132749,
      "learning_rate": 5.138888888888889e-05,
      "loss": 0.9278,
      "step": 148
    },
    {
      "epoch": 0.052146706066400136,
      "grad_norm": 2.4849685204332834,
      "learning_rate": 5.208333333333334e-05,
      "loss": 0.7531,
      "step": 150
    },
    {
      "epoch": 0.052146706066400136,
      "eval_loss": 0.9286500215530396,
      "eval_runtime": 711.2619,
      "eval_samples_per_second": 6.808,
      "eval_steps_per_second": 0.214,
      "step": 150
    },
    {
      "epoch": 0.052841995480618804,
      "grad_norm": 3.4927981580288776,
      "learning_rate": 5.2777777777777784e-05,
      "loss": 0.7422,
      "step": 152
    },
    {
      "epoch": 0.05353728489483748,
      "grad_norm": 2.6593724943984682,
      "learning_rate": 5.3472222222222224e-05,
      "loss": 0.7762,
      "step": 154
    },
    {
      "epoch": 0.05423257430905615,
      "grad_norm": 2.99709354088536,
      "learning_rate": 5.4166666666666664e-05,
      "loss": 0.8658,
      "step": 156
    },
    {
      "epoch": 0.054927863723274815,
      "grad_norm": 3.9196361623391414,
      "learning_rate": 5.486111111111112e-05,
      "loss": 1.1783,
      "step": 158
    },
    {
      "epoch": 0.05562315313749348,
      "grad_norm": 3.022562685608673,
      "learning_rate": 5.555555555555556e-05,
      "loss": 1.0602,
      "step": 160
    },
    {
      "epoch": 0.05631844255171215,
      "grad_norm": 3.292942684053579,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 0.8561,
      "step": 162
    },
    {
      "epoch": 0.05701373196593082,
      "grad_norm": 2.397775023338686,
      "learning_rate": 5.6944444444444445e-05,
      "loss": 0.9561,
      "step": 164
    },
    {
      "epoch": 0.05770902138014949,
      "grad_norm": 3.0546714312119643,
      "learning_rate": 5.7638888888888886e-05,
      "loss": 0.8274,
      "step": 166
    },
    {
      "epoch": 0.058404310794368156,
      "grad_norm": 2.4021883380894393,
      "learning_rate": 5.833333333333334e-05,
      "loss": 0.8492,
      "step": 168
    },
    {
      "epoch": 0.059099600208586824,
      "grad_norm": 2.6052829770116293,
      "learning_rate": 5.902777777777778e-05,
      "loss": 1.1005,
      "step": 170
    },
    {
      "epoch": 0.05979488962280549,
      "grad_norm": 1.6838192892320467,
      "learning_rate": 5.972222222222223e-05,
      "loss": 1.0157,
      "step": 172
    },
    {
      "epoch": 0.06049017903702416,
      "grad_norm": 3.98880214871885,
      "learning_rate": 6.041666666666667e-05,
      "loss": 0.8136,
      "step": 174
    },
    {
      "epoch": 0.06118546845124283,
      "grad_norm": 4.81494345341073,
      "learning_rate": 6.111111111111112e-05,
      "loss": 0.9475,
      "step": 176
    },
    {
      "epoch": 0.061880757865461496,
      "grad_norm": 1.9926278890091862,
      "learning_rate": 6.180555555555556e-05,
      "loss": 0.6549,
      "step": 178
    },
    {
      "epoch": 0.06257604727968016,
      "grad_norm": 1.750353030728397,
      "learning_rate": 6.25e-05,
      "loss": 1.1442,
      "step": 180
    },
    {
      "epoch": 0.06327133669389884,
      "grad_norm": 4.77249782692129,
      "learning_rate": 6.319444444444444e-05,
      "loss": 1.1927,
      "step": 182
    },
    {
      "epoch": 0.0639666261081175,
      "grad_norm": 1.866901788617278,
      "learning_rate": 6.388888888888888e-05,
      "loss": 1.0032,
      "step": 184
    },
    {
      "epoch": 0.06466191552233617,
      "grad_norm": 1.7495859751833545,
      "learning_rate": 6.458333333333334e-05,
      "loss": 0.8439,
      "step": 186
    },
    {
      "epoch": 0.06535720493655484,
      "grad_norm": 2.971674018184174,
      "learning_rate": 6.527777777777778e-05,
      "loss": 0.9817,
      "step": 188
    },
    {
      "epoch": 0.06605249435077351,
      "grad_norm": 2.3753292673540165,
      "learning_rate": 6.597222222222223e-05,
      "loss": 0.9919,
      "step": 190
    },
    {
      "epoch": 0.06674778376499217,
      "grad_norm": 2.7019148741557744,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.1442,
      "step": 192
    },
    {
      "epoch": 0.06744307317921085,
      "grad_norm": 1.8670113287712482,
      "learning_rate": 6.736111111111112e-05,
      "loss": 0.8905,
      "step": 194
    },
    {
      "epoch": 0.06813836259342952,
      "grad_norm": 1.4199145680760579,
      "learning_rate": 6.805555555555556e-05,
      "loss": 0.7223,
      "step": 196
    },
    {
      "epoch": 0.06883365200764818,
      "grad_norm": 3.894200902880186,
      "learning_rate": 6.875e-05,
      "loss": 0.9005,
      "step": 198
    },
    {
      "epoch": 0.06952894142186686,
      "grad_norm": 3.2710376491241955,
      "learning_rate": 6.944444444444444e-05,
      "loss": 1.1605,
      "step": 200
    },
    {
      "epoch": 0.07022423083608552,
      "grad_norm": 1.6459754670035065,
      "learning_rate": 7.013888888888888e-05,
      "loss": 0.8551,
      "step": 202
    },
    {
      "epoch": 0.0709195202503042,
      "grad_norm": 4.361031640374508,
      "learning_rate": 7.083333333333334e-05,
      "loss": 0.7007,
      "step": 204
    },
    {
      "epoch": 0.07161480966452285,
      "grad_norm": 3.573741549123141,
      "learning_rate": 7.152777777777778e-05,
      "loss": 1.1396,
      "step": 206
    },
    {
      "epoch": 0.07231009907874153,
      "grad_norm": 3.296990311359108,
      "learning_rate": 7.222222222222222e-05,
      "loss": 0.8695,
      "step": 208
    },
    {
      "epoch": 0.07300538849296019,
      "grad_norm": 6.769659823038884,
      "learning_rate": 7.291666666666667e-05,
      "loss": 1.0511,
      "step": 210
    },
    {
      "epoch": 0.07370067790717887,
      "grad_norm": 1.6695233666860303,
      "learning_rate": 7.361111111111111e-05,
      "loss": 0.8174,
      "step": 212
    },
    {
      "epoch": 0.07439596732139753,
      "grad_norm": 1.3543755216281146,
      "learning_rate": 7.430555555555557e-05,
      "loss": 0.7137,
      "step": 214
    },
    {
      "epoch": 0.0750912567356162,
      "grad_norm": 4.861063813937456,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0132,
      "step": 216
    },
    {
      "epoch": 0.07578654614983486,
      "grad_norm": 5.715913176528681,
      "learning_rate": 7.569444444444445e-05,
      "loss": 1.1657,
      "step": 218
    },
    {
      "epoch": 0.07648183556405354,
      "grad_norm": 4.0193839358302235,
      "learning_rate": 7.638888888888889e-05,
      "loss": 1.0547,
      "step": 220
    },
    {
      "epoch": 0.0771771249782722,
      "grad_norm": 2.26849134517291,
      "learning_rate": 7.708333333333334e-05,
      "loss": 0.9073,
      "step": 222
    },
    {
      "epoch": 0.07787241439249087,
      "grad_norm": 4.570943697810998,
      "learning_rate": 7.777777777777778e-05,
      "loss": 1.3726,
      "step": 224
    },
    {
      "epoch": 0.07856770380670955,
      "grad_norm": 1.1942436910880105,
      "learning_rate": 7.847222222222222e-05,
      "loss": 0.8146,
      "step": 226
    },
    {
      "epoch": 0.07926299322092821,
      "grad_norm": 2.111849588751211,
      "learning_rate": 7.916666666666666e-05,
      "loss": 1.1157,
      "step": 228
    },
    {
      "epoch": 0.07995828263514689,
      "grad_norm": 1.0283088880069582,
      "learning_rate": 7.986111111111112e-05,
      "loss": 0.767,
      "step": 230
    },
    {
      "epoch": 0.08065357204936555,
      "grad_norm": 1.2834055069208525,
      "learning_rate": 8.055555555555556e-05,
      "loss": 0.6625,
      "step": 232
    },
    {
      "epoch": 0.08134886146358422,
      "grad_norm": 1.3772942873595098,
      "learning_rate": 8.125000000000001e-05,
      "loss": 0.8065,
      "step": 234
    },
    {
      "epoch": 0.08204415087780288,
      "grad_norm": 1.5113456549735176,
      "learning_rate": 8.194444444444445e-05,
      "loss": 0.8606,
      "step": 236
    },
    {
      "epoch": 0.08273944029202156,
      "grad_norm": 1.5765846352838255,
      "learning_rate": 8.263888888888889e-05,
      "loss": 0.8335,
      "step": 238
    },
    {
      "epoch": 0.08343472970624022,
      "grad_norm": 2.4873462478329404,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.9705,
      "step": 240
    },
    {
      "epoch": 0.0841300191204589,
      "grad_norm": 1.2369219925635513,
      "learning_rate": 8.402777777777778e-05,
      "loss": 0.6061,
      "step": 242
    },
    {
      "epoch": 0.08482530853467755,
      "grad_norm": 2.542132212473201,
      "learning_rate": 8.472222222222222e-05,
      "loss": 0.9142,
      "step": 244
    },
    {
      "epoch": 0.08552059794889623,
      "grad_norm": 2.0301734217803022,
      "learning_rate": 8.541666666666666e-05,
      "loss": 0.8997,
      "step": 246
    },
    {
      "epoch": 0.08621588736311489,
      "grad_norm": 1.8605316982945626,
      "learning_rate": 8.611111111111112e-05,
      "loss": 1.005,
      "step": 248
    },
    {
      "epoch": 0.08691117677733357,
      "grad_norm": 1.193555257951713,
      "learning_rate": 8.680555555555556e-05,
      "loss": 0.8617,
      "step": 250
    },
    {
      "epoch": 0.08760646619155224,
      "grad_norm": 1.243815428863678,
      "learning_rate": 8.75e-05,
      "loss": 0.6261,
      "step": 252
    },
    {
      "epoch": 0.0883017556057709,
      "grad_norm": 1.6487754861704442,
      "learning_rate": 8.819444444444445e-05,
      "loss": 0.9219,
      "step": 254
    },
    {
      "epoch": 0.08899704501998958,
      "grad_norm": 1.1768410857322613,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.8563,
      "step": 256
    },
    {
      "epoch": 0.08969233443420824,
      "grad_norm": 1.1155265191420587,
      "learning_rate": 8.958333333333335e-05,
      "loss": 0.998,
      "step": 258
    },
    {
      "epoch": 0.09038762384842691,
      "grad_norm": 1.4432456616674065,
      "learning_rate": 9.027777777777779e-05,
      "loss": 0.8755,
      "step": 260
    },
    {
      "epoch": 0.09108291326264557,
      "grad_norm": 1.6013486668654413,
      "learning_rate": 9.097222222222223e-05,
      "loss": 0.6998,
      "step": 262
    },
    {
      "epoch": 0.09177820267686425,
      "grad_norm": 0.7869968770186737,
      "learning_rate": 9.166666666666667e-05,
      "loss": 0.7574,
      "step": 264
    },
    {
      "epoch": 0.09247349209108291,
      "grad_norm": 2.1117903903864566,
      "learning_rate": 9.236111111111112e-05,
      "loss": 0.8436,
      "step": 266
    },
    {
      "epoch": 0.09316878150530158,
      "grad_norm": 1.3582999584721895,
      "learning_rate": 9.305555555555556e-05,
      "loss": 0.8503,
      "step": 268
    },
    {
      "epoch": 0.09386407091952025,
      "grad_norm": 1.5674581009005415,
      "learning_rate": 9.375e-05,
      "loss": 0.9561,
      "step": 270
    },
    {
      "epoch": 0.09455936033373892,
      "grad_norm": 1.0274246934159952,
      "learning_rate": 9.444444444444444e-05,
      "loss": 0.7537,
      "step": 272
    },
    {
      "epoch": 0.09525464974795758,
      "grad_norm": 0.9595343432519174,
      "learning_rate": 9.513888888888888e-05,
      "loss": 0.7855,
      "step": 274
    },
    {
      "epoch": 0.09594993916217626,
      "grad_norm": 1.3313115114367815,
      "learning_rate": 9.583333333333334e-05,
      "loss": 0.7009,
      "step": 276
    },
    {
      "epoch": 0.09664522857639492,
      "grad_norm": 1.4409463331317498,
      "learning_rate": 9.652777777777779e-05,
      "loss": 1.0332,
      "step": 278
    },
    {
      "epoch": 0.09734051799061359,
      "grad_norm": 1.5445848213023137,
      "learning_rate": 9.722222222222223e-05,
      "loss": 0.7352,
      "step": 280
    },
    {
      "epoch": 0.09803580740483227,
      "grad_norm": 1.855518620927316,
      "learning_rate": 9.791666666666667e-05,
      "loss": 0.7191,
      "step": 282
    },
    {
      "epoch": 0.09873109681905093,
      "grad_norm": 1.4033396985161997,
      "learning_rate": 9.861111111111112e-05,
      "loss": 0.7886,
      "step": 284
    },
    {
      "epoch": 0.0994263862332696,
      "grad_norm": 1.9951521438049904,
      "learning_rate": 9.930555555555556e-05,
      "loss": 0.902,
      "step": 286
    },
    {
      "epoch": 0.10012167564748826,
      "grad_norm": 1.1360934228431687,
      "learning_rate": 0.0001,
      "loss": 0.8655,
      "step": 288
    },
    {
      "epoch": 0.10081696506170694,
      "grad_norm": 1.1225735433900375,
      "learning_rate": 0.00010069444444444445,
      "loss": 0.6851,
      "step": 290
    },
    {
      "epoch": 0.1015122544759256,
      "grad_norm": 0.770446891371583,
      "learning_rate": 0.00010138888888888889,
      "loss": 0.7775,
      "step": 292
    },
    {
      "epoch": 0.10220754389014428,
      "grad_norm": 0.797064180835607,
      "learning_rate": 0.00010208333333333333,
      "loss": 0.6567,
      "step": 294
    },
    {
      "epoch": 0.10290283330436294,
      "grad_norm": 0.9876188134326442,
      "learning_rate": 0.00010277777777777778,
      "loss": 0.7299,
      "step": 296
    },
    {
      "epoch": 0.10359812271858161,
      "grad_norm": 0.799601623643537,
      "learning_rate": 0.00010347222222222223,
      "loss": 0.7891,
      "step": 298
    },
    {
      "epoch": 0.10429341213280027,
      "grad_norm": 0.8791512281362982,
      "learning_rate": 0.00010416666666666667,
      "loss": 0.677,
      "step": 300
    },
    {
      "epoch": 0.10429341213280027,
      "eval_loss": 0.6939894556999207,
      "eval_runtime": 709.5268,
      "eval_samples_per_second": 6.824,
      "eval_steps_per_second": 0.214,
      "step": 300
    },
    {
      "epoch": 0.10498870154701895,
      "grad_norm": 1.329564645811689,
      "learning_rate": 0.00010486111111111113,
      "loss": 0.7511,
      "step": 302
    },
    {
      "epoch": 0.10568399096123761,
      "grad_norm": 0.7130415058241142,
      "learning_rate": 0.00010555555555555557,
      "loss": 0.709,
      "step": 304
    },
    {
      "epoch": 0.10637928037545628,
      "grad_norm": 1.1294733411370705,
      "learning_rate": 0.00010625000000000001,
      "loss": 0.7781,
      "step": 306
    },
    {
      "epoch": 0.10707456978967496,
      "grad_norm": 0.581727680778003,
      "learning_rate": 0.00010694444444444445,
      "loss": 0.6296,
      "step": 308
    },
    {
      "epoch": 0.10776985920389362,
      "grad_norm": 0.9855391514311871,
      "learning_rate": 0.00010763888888888889,
      "loss": 0.7852,
      "step": 310
    },
    {
      "epoch": 0.1084651486181123,
      "grad_norm": 0.5708379609857769,
      "learning_rate": 0.00010833333333333333,
      "loss": 0.6759,
      "step": 312
    },
    {
      "epoch": 0.10916043803233096,
      "grad_norm": 0.7243330859752051,
      "learning_rate": 0.00010902777777777777,
      "loss": 0.6326,
      "step": 314
    },
    {
      "epoch": 0.10985572744654963,
      "grad_norm": 1.3922163003264714,
      "learning_rate": 0.00010972222222222224,
      "loss": 0.5406,
      "step": 316
    },
    {
      "epoch": 0.11055101686076829,
      "grad_norm": 2.604851740418153,
      "learning_rate": 0.00011041666666666668,
      "loss": 0.8013,
      "step": 318
    },
    {
      "epoch": 0.11124630627498697,
      "grad_norm": 1.83511037983023,
      "learning_rate": 0.00011111111111111112,
      "loss": 0.8048,
      "step": 320
    },
    {
      "epoch": 0.11194159568920563,
      "grad_norm": 1.6737764614655666,
      "learning_rate": 0.00011180555555555556,
      "loss": 0.6943,
      "step": 322
    },
    {
      "epoch": 0.1126368851034243,
      "grad_norm": 1.077667781888673,
      "learning_rate": 0.00011250000000000001,
      "loss": 0.8054,
      "step": 324
    },
    {
      "epoch": 0.11333217451764296,
      "grad_norm": 0.5597693838209001,
      "learning_rate": 0.00011319444444444445,
      "loss": 0.6251,
      "step": 326
    },
    {
      "epoch": 0.11402746393186164,
      "grad_norm": 0.5636865078063477,
      "learning_rate": 0.00011388888888888889,
      "loss": 0.6832,
      "step": 328
    },
    {
      "epoch": 0.1147227533460803,
      "grad_norm": 0.6445176566556912,
      "learning_rate": 0.00011458333333333333,
      "loss": 0.6661,
      "step": 330
    },
    {
      "epoch": 0.11541804276029898,
      "grad_norm": 0.607532412895966,
      "learning_rate": 0.00011527777777777777,
      "loss": 0.7026,
      "step": 332
    },
    {
      "epoch": 0.11611333217451764,
      "grad_norm": 1.0496949694240345,
      "learning_rate": 0.00011597222222222224,
      "loss": 0.6451,
      "step": 334
    },
    {
      "epoch": 0.11680862158873631,
      "grad_norm": 0.9469233272179363,
      "learning_rate": 0.00011666666666666668,
      "loss": 0.6694,
      "step": 336
    },
    {
      "epoch": 0.11750391100295499,
      "grad_norm": 1.3789882614528595,
      "learning_rate": 0.00011736111111111112,
      "loss": 0.7031,
      "step": 338
    },
    {
      "epoch": 0.11819920041717365,
      "grad_norm": 1.2016589928587922,
      "learning_rate": 0.00011805555555555556,
      "loss": 0.6676,
      "step": 340
    },
    {
      "epoch": 0.11889448983139232,
      "grad_norm": 1.3112845410865746,
      "learning_rate": 0.00011875,
      "loss": 0.7236,
      "step": 342
    },
    {
      "epoch": 0.11958977924561098,
      "grad_norm": 0.6278789052805739,
      "learning_rate": 0.00011944444444444445,
      "loss": 0.6249,
      "step": 344
    },
    {
      "epoch": 0.12028506865982966,
      "grad_norm": 0.5405148902553805,
      "learning_rate": 0.0001201388888888889,
      "loss": 0.6412,
      "step": 346
    },
    {
      "epoch": 0.12098035807404832,
      "grad_norm": 1.6923616138961255,
      "learning_rate": 0.00012083333333333333,
      "loss": 0.7017,
      "step": 348
    },
    {
      "epoch": 0.121675647488267,
      "grad_norm": 1.5994113095813072,
      "learning_rate": 0.00012152777777777777,
      "loss": 0.7075,
      "step": 350
    },
    {
      "epoch": 0.12237093690248566,
      "grad_norm": 2.3083516502461783,
      "learning_rate": 0.00012222222222222224,
      "loss": 0.7755,
      "step": 352
    },
    {
      "epoch": 0.12306622631670433,
      "grad_norm": 0.8769167165306447,
      "learning_rate": 0.00012291666666666668,
      "loss": 0.7428,
      "step": 354
    },
    {
      "epoch": 0.12376151573092299,
      "grad_norm": 0.4456469638797518,
      "learning_rate": 0.00012361111111111112,
      "loss": 0.7002,
      "step": 356
    },
    {
      "epoch": 0.12445680514514167,
      "grad_norm": 0.8199993647792723,
      "learning_rate": 0.00012430555555555556,
      "loss": 0.7222,
      "step": 358
    },
    {
      "epoch": 0.12515209455936033,
      "grad_norm": 0.3741084955970339,
      "learning_rate": 0.000125,
      "loss": 0.6821,
      "step": 360
    },
    {
      "epoch": 0.125847383973579,
      "grad_norm": 0.2972203241099783,
      "learning_rate": 0.00012569444444444444,
      "loss": 0.7083,
      "step": 362
    },
    {
      "epoch": 0.12654267338779768,
      "grad_norm": 0.40651740632246575,
      "learning_rate": 0.00012638888888888888,
      "loss": 0.7144,
      "step": 364
    },
    {
      "epoch": 0.12723796280201635,
      "grad_norm": 0.4388093199704831,
      "learning_rate": 0.00012708333333333332,
      "loss": 0.6863,
      "step": 366
    },
    {
      "epoch": 0.127933252216235,
      "grad_norm": 0.45131267837810835,
      "learning_rate": 0.00012777777777777776,
      "loss": 0.6329,
      "step": 368
    },
    {
      "epoch": 0.12862854163045367,
      "grad_norm": 0.3837830352665476,
      "learning_rate": 0.00012847222222222223,
      "loss": 0.6938,
      "step": 370
    },
    {
      "epoch": 0.12932383104467235,
      "grad_norm": 0.6378624076702905,
      "learning_rate": 0.00012916666666666667,
      "loss": 0.6433,
      "step": 372
    },
    {
      "epoch": 0.13001912045889102,
      "grad_norm": 0.41344403509072675,
      "learning_rate": 0.0001298611111111111,
      "loss": 0.6892,
      "step": 374
    },
    {
      "epoch": 0.13071440987310967,
      "grad_norm": 1.3428471718098582,
      "learning_rate": 0.00013055555555555555,
      "loss": 0.7793,
      "step": 376
    },
    {
      "epoch": 0.13140969928732835,
      "grad_norm": 0.6202275876483008,
      "learning_rate": 0.00013125000000000002,
      "loss": 0.6836,
      "step": 378
    },
    {
      "epoch": 0.13210498870154702,
      "grad_norm": 0.774014383396004,
      "learning_rate": 0.00013194444444444446,
      "loss": 0.701,
      "step": 380
    },
    {
      "epoch": 0.1328002781157657,
      "grad_norm": 1.0650744092292224,
      "learning_rate": 0.0001326388888888889,
      "loss": 0.6562,
      "step": 382
    },
    {
      "epoch": 0.13349556752998434,
      "grad_norm": 0.5540168492927294,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.6267,
      "step": 384
    },
    {
      "epoch": 0.13419085694420302,
      "grad_norm": 0.6403086140920178,
      "learning_rate": 0.00013402777777777778,
      "loss": 0.6268,
      "step": 386
    },
    {
      "epoch": 0.1348861463584217,
      "grad_norm": 0.9197218298114509,
      "learning_rate": 0.00013472222222222225,
      "loss": 0.7208,
      "step": 388
    },
    {
      "epoch": 0.13558143577264037,
      "grad_norm": 0.4235387152144674,
      "learning_rate": 0.0001354166666666667,
      "loss": 0.6057,
      "step": 390
    },
    {
      "epoch": 0.13627672518685904,
      "grad_norm": 0.425648636635043,
      "learning_rate": 0.00013611111111111113,
      "loss": 0.6387,
      "step": 392
    },
    {
      "epoch": 0.1369720146010777,
      "grad_norm": 1.034952396366508,
      "learning_rate": 0.00013680555555555557,
      "loss": 0.7307,
      "step": 394
    },
    {
      "epoch": 0.13766730401529637,
      "grad_norm": 0.8865505732510678,
      "learning_rate": 0.0001375,
      "loss": 0.6547,
      "step": 396
    },
    {
      "epoch": 0.13836259342951504,
      "grad_norm": 0.6490094573088415,
      "learning_rate": 0.00013819444444444445,
      "loss": 0.6909,
      "step": 398
    },
    {
      "epoch": 0.13905788284373372,
      "grad_norm": 0.8592713330921486,
      "learning_rate": 0.0001388888888888889,
      "loss": 0.6442,
      "step": 400
    },
    {
      "epoch": 0.13975317225795236,
      "grad_norm": 1.0279375953862069,
      "learning_rate": 0.00013958333333333333,
      "loss": 0.6907,
      "step": 402
    },
    {
      "epoch": 0.14044846167217104,
      "grad_norm": 0.6151057351983696,
      "learning_rate": 0.00014027777777777777,
      "loss": 0.697,
      "step": 404
    },
    {
      "epoch": 0.1411437510863897,
      "grad_norm": 0.7417879717121144,
      "learning_rate": 0.00014097222222222224,
      "loss": 0.698,
      "step": 406
    },
    {
      "epoch": 0.1418390405006084,
      "grad_norm": 0.5161453091859882,
      "learning_rate": 0.00014166666666666668,
      "loss": 0.6915,
      "step": 408
    },
    {
      "epoch": 0.14253432991482703,
      "grad_norm": 0.5667521096080546,
      "learning_rate": 0.00014236111111111112,
      "loss": 0.6609,
      "step": 410
    },
    {
      "epoch": 0.1432296193290457,
      "grad_norm": 0.7560317583703429,
      "learning_rate": 0.00014305555555555556,
      "loss": 0.6392,
      "step": 412
    },
    {
      "epoch": 0.14392490874326438,
      "grad_norm": 0.5456303981546313,
      "learning_rate": 0.00014375,
      "loss": 0.7561,
      "step": 414
    },
    {
      "epoch": 0.14462019815748306,
      "grad_norm": 0.4218469018151631,
      "learning_rate": 0.00014444444444444444,
      "loss": 0.675,
      "step": 416
    },
    {
      "epoch": 0.14531548757170173,
      "grad_norm": 0.3790517974518404,
      "learning_rate": 0.00014513888888888888,
      "loss": 0.6697,
      "step": 418
    },
    {
      "epoch": 0.14601077698592038,
      "grad_norm": 0.36495171567914964,
      "learning_rate": 0.00014583333333333335,
      "loss": 0.6467,
      "step": 420
    },
    {
      "epoch": 0.14670606640013906,
      "grad_norm": 0.3793520501177419,
      "learning_rate": 0.00014652777777777779,
      "loss": 0.6234,
      "step": 422
    },
    {
      "epoch": 0.14740135581435773,
      "grad_norm": 0.6214905371013544,
      "learning_rate": 0.00014722222222222223,
      "loss": 0.64,
      "step": 424
    },
    {
      "epoch": 0.1480966452285764,
      "grad_norm": 0.4103438113660832,
      "learning_rate": 0.0001479166666666667,
      "loss": 0.6174,
      "step": 426
    },
    {
      "epoch": 0.14879193464279505,
      "grad_norm": 0.5802311029669485,
      "learning_rate": 0.00014861111111111113,
      "loss": 0.6941,
      "step": 428
    },
    {
      "epoch": 0.14948722405701373,
      "grad_norm": 0.4876572424583591,
      "learning_rate": 0.00014930555555555557,
      "loss": 0.6178,
      "step": 430
    },
    {
      "epoch": 0.1501825134712324,
      "grad_norm": 0.6252220373472688,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.692,
      "step": 432
    },
    {
      "epoch": 0.15087780288545108,
      "grad_norm": 1.516758455176553,
      "learning_rate": 0.00015069444444444445,
      "loss": 0.6812,
      "step": 434
    },
    {
      "epoch": 0.15157309229966973,
      "grad_norm": 0.48294685446689867,
      "learning_rate": 0.0001513888888888889,
      "loss": 0.7085,
      "step": 436
    },
    {
      "epoch": 0.1522683817138884,
      "grad_norm": 0.414196721150167,
      "learning_rate": 0.00015208333333333333,
      "loss": 0.673,
      "step": 438
    },
    {
      "epoch": 0.15296367112810708,
      "grad_norm": 0.7294881062503469,
      "learning_rate": 0.00015277777777777777,
      "loss": 0.6415,
      "step": 440
    },
    {
      "epoch": 0.15365896054232575,
      "grad_norm": 0.4846932912054438,
      "learning_rate": 0.00015347222222222224,
      "loss": 0.6644,
      "step": 442
    },
    {
      "epoch": 0.1543542499565444,
      "grad_norm": 0.6067400167426414,
      "learning_rate": 0.00015416666666666668,
      "loss": 0.6816,
      "step": 444
    },
    {
      "epoch": 0.15504953937076307,
      "grad_norm": 0.5586278026791864,
      "learning_rate": 0.00015486111111111112,
      "loss": 0.6223,
      "step": 446
    },
    {
      "epoch": 0.15574482878498175,
      "grad_norm": 0.3947236166856149,
      "learning_rate": 0.00015555555555555556,
      "loss": 0.7297,
      "step": 448
    },
    {
      "epoch": 0.15644011819920042,
      "grad_norm": 0.371457756887947,
      "learning_rate": 0.00015625,
      "loss": 0.6099,
      "step": 450
    },
    {
      "epoch": 0.15644011819920042,
      "eval_loss": 0.6634477376937866,
      "eval_runtime": 706.8027,
      "eval_samples_per_second": 6.851,
      "eval_steps_per_second": 0.215,
      "step": 450
    },
    {
      "epoch": 0.1571354076134191,
      "grad_norm": 0.4157928192338373,
      "learning_rate": 0.00015694444444444444,
      "loss": 0.7163,
      "step": 452
    },
    {
      "epoch": 0.15783069702763775,
      "grad_norm": 0.6318513446857751,
      "learning_rate": 0.00015763888888888888,
      "loss": 0.658,
      "step": 454
    },
    {
      "epoch": 0.15852598644185642,
      "grad_norm": 0.3188170086808347,
      "learning_rate": 0.00015833333333333332,
      "loss": 0.5966,
      "step": 456
    },
    {
      "epoch": 0.1592212758560751,
      "grad_norm": 0.4873146052629784,
      "learning_rate": 0.00015902777777777776,
      "loss": 0.687,
      "step": 458
    },
    {
      "epoch": 0.15991656527029377,
      "grad_norm": 0.3939618900754228,
      "learning_rate": 0.00015972222222222223,
      "loss": 0.6704,
      "step": 460
    },
    {
      "epoch": 0.16061185468451242,
      "grad_norm": 0.43442241310529234,
      "learning_rate": 0.00016041666666666667,
      "loss": 0.6941,
      "step": 462
    },
    {
      "epoch": 0.1613071440987311,
      "grad_norm": 0.5944395304786771,
      "learning_rate": 0.0001611111111111111,
      "loss": 0.5934,
      "step": 464
    },
    {
      "epoch": 0.16200243351294977,
      "grad_norm": 0.6186667715273749,
      "learning_rate": 0.00016180555555555555,
      "loss": 0.6604,
      "step": 466
    },
    {
      "epoch": 0.16269772292716844,
      "grad_norm": 0.8945416923537124,
      "learning_rate": 0.00016250000000000002,
      "loss": 0.6224,
      "step": 468
    },
    {
      "epoch": 0.1633930123413871,
      "grad_norm": 1.0526200270890014,
      "learning_rate": 0.00016319444444444446,
      "loss": 0.616,
      "step": 470
    },
    {
      "epoch": 0.16408830175560576,
      "grad_norm": 0.4575985966727194,
      "learning_rate": 0.0001638888888888889,
      "loss": 0.6591,
      "step": 472
    },
    {
      "epoch": 0.16478359116982444,
      "grad_norm": 0.4499931204645402,
      "learning_rate": 0.00016458333333333334,
      "loss": 0.6644,
      "step": 474
    },
    {
      "epoch": 0.16547888058404311,
      "grad_norm": 0.8365574384741885,
      "learning_rate": 0.00016527777777777778,
      "loss": 0.6503,
      "step": 476
    },
    {
      "epoch": 0.1661741699982618,
      "grad_norm": 0.5265284650577726,
      "learning_rate": 0.00016597222222222225,
      "loss": 0.557,
      "step": 478
    },
    {
      "epoch": 0.16686945941248044,
      "grad_norm": 0.6004742828917694,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.696,
      "step": 480
    },
    {
      "epoch": 0.1675647488266991,
      "grad_norm": 0.8653134209549845,
      "learning_rate": 0.00016736111111111113,
      "loss": 0.6544,
      "step": 482
    },
    {
      "epoch": 0.1682600382409178,
      "grad_norm": 1.033230873418064,
      "learning_rate": 0.00016805555555555557,
      "loss": 0.641,
      "step": 484
    },
    {
      "epoch": 0.16895532765513646,
      "grad_norm": 0.45457193405665913,
      "learning_rate": 0.00016875,
      "loss": 0.5952,
      "step": 486
    },
    {
      "epoch": 0.1696506170693551,
      "grad_norm": 0.7413902971651539,
      "learning_rate": 0.00016944444444444445,
      "loss": 0.6306,
      "step": 488
    },
    {
      "epoch": 0.17034590648357378,
      "grad_norm": 0.5808764628223875,
      "learning_rate": 0.0001701388888888889,
      "loss": 0.6714,
      "step": 490
    },
    {
      "epoch": 0.17104119589779246,
      "grad_norm": 0.42505370383777924,
      "learning_rate": 0.00017083333333333333,
      "loss": 0.6432,
      "step": 492
    },
    {
      "epoch": 0.17173648531201113,
      "grad_norm": 0.9403542369255028,
      "learning_rate": 0.00017152777777777777,
      "loss": 0.626,
      "step": 494
    },
    {
      "epoch": 0.17243177472622978,
      "grad_norm": 0.41714707593148775,
      "learning_rate": 0.00017222222222222224,
      "loss": 0.6132,
      "step": 496
    },
    {
      "epoch": 0.17312706414044846,
      "grad_norm": 0.6660896849726371,
      "learning_rate": 0.00017291666666666668,
      "loss": 0.7009,
      "step": 498
    },
    {
      "epoch": 0.17382235355466713,
      "grad_norm": 0.5079993412341118,
      "learning_rate": 0.00017361111111111112,
      "loss": 0.6808,
      "step": 500
    },
    {
      "epoch": 0.1745176429688858,
      "grad_norm": 0.3426487947129772,
      "learning_rate": 0.00017430555555555556,
      "loss": 0.687,
      "step": 502
    },
    {
      "epoch": 0.17521293238310448,
      "grad_norm": 0.7316051797581208,
      "learning_rate": 0.000175,
      "loss": 0.686,
      "step": 504
    },
    {
      "epoch": 0.17590822179732313,
      "grad_norm": 0.4619785653282459,
      "learning_rate": 0.00017569444444444444,
      "loss": 0.7019,
      "step": 506
    },
    {
      "epoch": 0.1766035112115418,
      "grad_norm": 0.47786977912472967,
      "learning_rate": 0.0001763888888888889,
      "loss": 0.7097,
      "step": 508
    },
    {
      "epoch": 0.17729880062576048,
      "grad_norm": 0.4204398291864812,
      "learning_rate": 0.00017708333333333335,
      "loss": 0.6715,
      "step": 510
    },
    {
      "epoch": 0.17799409003997915,
      "grad_norm": 0.5257761541709909,
      "learning_rate": 0.00017777777777777779,
      "loss": 0.6675,
      "step": 512
    },
    {
      "epoch": 0.1786893794541978,
      "grad_norm": 0.5752239420884688,
      "learning_rate": 0.00017847222222222225,
      "loss": 0.5847,
      "step": 514
    },
    {
      "epoch": 0.17938466886841647,
      "grad_norm": 0.5611432017880666,
      "learning_rate": 0.0001791666666666667,
      "loss": 0.5978,
      "step": 516
    },
    {
      "epoch": 0.18007995828263515,
      "grad_norm": 0.5147097358785945,
      "learning_rate": 0.00017986111111111113,
      "loss": 0.641,
      "step": 518
    },
    {
      "epoch": 0.18077524769685382,
      "grad_norm": 0.4387330962078066,
      "learning_rate": 0.00018055555555555557,
      "loss": 0.5771,
      "step": 520
    },
    {
      "epoch": 0.18147053711107247,
      "grad_norm": 0.5712526440788663,
      "learning_rate": 0.00018125000000000001,
      "loss": 0.6212,
      "step": 522
    },
    {
      "epoch": 0.18216582652529115,
      "grad_norm": 0.6357133907029916,
      "learning_rate": 0.00018194444444444445,
      "loss": 0.5981,
      "step": 524
    },
    {
      "epoch": 0.18286111593950982,
      "grad_norm": 1.2129620643054042,
      "learning_rate": 0.0001826388888888889,
      "loss": 0.5848,
      "step": 526
    },
    {
      "epoch": 0.1835564053537285,
      "grad_norm": 1.651599276688714,
      "learning_rate": 0.00018333333333333334,
      "loss": 0.7089,
      "step": 528
    },
    {
      "epoch": 0.18425169476794717,
      "grad_norm": 0.9051688342434142,
      "learning_rate": 0.00018402777777777778,
      "loss": 0.7255,
      "step": 530
    },
    {
      "epoch": 0.18494698418216582,
      "grad_norm": 0.6932441010863684,
      "learning_rate": 0.00018472222222222224,
      "loss": 0.6497,
      "step": 532
    },
    {
      "epoch": 0.1856422735963845,
      "grad_norm": 1.2163006408955903,
      "learning_rate": 0.00018541666666666668,
      "loss": 0.7585,
      "step": 534
    },
    {
      "epoch": 0.18633756301060317,
      "grad_norm": 0.4387074048377486,
      "learning_rate": 0.00018611111111111112,
      "loss": 0.6541,
      "step": 536
    },
    {
      "epoch": 0.18703285242482184,
      "grad_norm": 0.49430581216479236,
      "learning_rate": 0.00018680555555555556,
      "loss": 0.6489,
      "step": 538
    },
    {
      "epoch": 0.1877281418390405,
      "grad_norm": 0.47561442420926275,
      "learning_rate": 0.0001875,
      "loss": 0.6406,
      "step": 540
    },
    {
      "epoch": 0.18842343125325917,
      "grad_norm": 0.7046092923664611,
      "learning_rate": 0.00018819444444444444,
      "loss": 0.6099,
      "step": 542
    },
    {
      "epoch": 0.18911872066747784,
      "grad_norm": 0.46256136269201026,
      "learning_rate": 0.00018888888888888888,
      "loss": 0.6992,
      "step": 544
    },
    {
      "epoch": 0.18981401008169652,
      "grad_norm": 0.5489877082982352,
      "learning_rate": 0.00018958333333333332,
      "loss": 0.5817,
      "step": 546
    },
    {
      "epoch": 0.19050929949591516,
      "grad_norm": 0.46129441798908893,
      "learning_rate": 0.00019027777777777776,
      "loss": 0.645,
      "step": 548
    },
    {
      "epoch": 0.19120458891013384,
      "grad_norm": 0.3724226655450749,
      "learning_rate": 0.00019097222222222223,
      "loss": 0.6177,
      "step": 550
    },
    {
      "epoch": 0.1918998783243525,
      "grad_norm": 0.4625796323902992,
      "learning_rate": 0.00019166666666666667,
      "loss": 0.6702,
      "step": 552
    },
    {
      "epoch": 0.1925951677385712,
      "grad_norm": 1.0142124821047231,
      "learning_rate": 0.0001923611111111111,
      "loss": 0.7173,
      "step": 554
    },
    {
      "epoch": 0.19329045715278984,
      "grad_norm": 0.5452830720753045,
      "learning_rate": 0.00019305555555555558,
      "loss": 0.6897,
      "step": 556
    },
    {
      "epoch": 0.1939857465670085,
      "grad_norm": 0.5723214484723252,
      "learning_rate": 0.00019375000000000002,
      "loss": 0.6495,
      "step": 558
    },
    {
      "epoch": 0.19468103598122719,
      "grad_norm": 0.39925844595054966,
      "learning_rate": 0.00019444444444444446,
      "loss": 0.6479,
      "step": 560
    },
    {
      "epoch": 0.19537632539544586,
      "grad_norm": 0.3575565088751118,
      "learning_rate": 0.0001951388888888889,
      "loss": 0.6593,
      "step": 562
    },
    {
      "epoch": 0.19607161480966454,
      "grad_norm": 0.6119892689536569,
      "learning_rate": 0.00019583333333333334,
      "loss": 0.6831,
      "step": 564
    },
    {
      "epoch": 0.19676690422388318,
      "grad_norm": 0.5076569073121309,
      "learning_rate": 0.00019652777777777778,
      "loss": 0.6294,
      "step": 566
    },
    {
      "epoch": 0.19746219363810186,
      "grad_norm": 0.622226505121079,
      "learning_rate": 0.00019722222222222225,
      "loss": 0.5551,
      "step": 568
    },
    {
      "epoch": 0.19815748305232053,
      "grad_norm": 0.4106526912254415,
      "learning_rate": 0.0001979166666666667,
      "loss": 0.6503,
      "step": 570
    },
    {
      "epoch": 0.1988527724665392,
      "grad_norm": 0.47164751383243125,
      "learning_rate": 0.00019861111111111113,
      "loss": 0.6992,
      "step": 572
    },
    {
      "epoch": 0.19954806188075785,
      "grad_norm": 0.492328312395826,
      "learning_rate": 0.00019930555555555557,
      "loss": 0.7183,
      "step": 574
    },
    {
      "epoch": 0.20024335129497653,
      "grad_norm": 0.5087240677439067,
      "learning_rate": 0.0002,
      "loss": 0.6429,
      "step": 576
    },
    {
      "epoch": 0.2009386407091952,
      "grad_norm": 0.9311216098504759,
      "learning_rate": 0.00019999992632143608,
      "loss": 0.6586,
      "step": 578
    },
    {
      "epoch": 0.20163393012341388,
      "grad_norm": 0.38840935170189844,
      "learning_rate": 0.00019999970528585288,
      "loss": 0.5905,
      "step": 580
    },
    {
      "epoch": 0.20232921953763253,
      "grad_norm": 0.5966472765038273,
      "learning_rate": 0.0001999993368935761,
      "loss": 0.7112,
      "step": 582
    },
    {
      "epoch": 0.2030245089518512,
      "grad_norm": 0.535232520705813,
      "learning_rate": 0.00019999882114514863,
      "loss": 0.7036,
      "step": 584
    },
    {
      "epoch": 0.20371979836606988,
      "grad_norm": 0.5039916125175515,
      "learning_rate": 0.0001999981580413304,
      "loss": 0.7485,
      "step": 586
    },
    {
      "epoch": 0.20441508778028855,
      "grad_norm": 0.6733262892097686,
      "learning_rate": 0.0001999973475830986,
      "loss": 0.7312,
      "step": 588
    },
    {
      "epoch": 0.20511037719450723,
      "grad_norm": 0.42805740684614807,
      "learning_rate": 0.00019999638977164747,
      "loss": 0.6356,
      "step": 590
    },
    {
      "epoch": 0.20580566660872587,
      "grad_norm": 0.976918546573707,
      "learning_rate": 0.00019999528460838844,
      "loss": 0.666,
      "step": 592
    },
    {
      "epoch": 0.20650095602294455,
      "grad_norm": 0.674463735133944,
      "learning_rate": 0.00019999403209495,
      "loss": 0.6735,
      "step": 594
    },
    {
      "epoch": 0.20719624543716322,
      "grad_norm": 0.412232119296174,
      "learning_rate": 0.00019999263223317786,
      "loss": 0.6364,
      "step": 596
    },
    {
      "epoch": 0.2078915348513819,
      "grad_norm": 0.3611034413524545,
      "learning_rate": 0.0001999910850251348,
      "loss": 0.6368,
      "step": 598
    },
    {
      "epoch": 0.20858682426560055,
      "grad_norm": 0.542506398091656,
      "learning_rate": 0.0001999893904731007,
      "loss": 0.7292,
      "step": 600
    },
    {
      "epoch": 0.20858682426560055,
      "eval_loss": 0.6517693400382996,
      "eval_runtime": 707.4674,
      "eval_samples_per_second": 6.844,
      "eval_steps_per_second": 0.215,
      "step": 600
    },
    {
      "epoch": 0.20928211367981922,
      "grad_norm": 0.4799286290563759,
      "learning_rate": 0.0001999875485795727,
      "loss": 0.6609,
      "step": 602
    },
    {
      "epoch": 0.2099774030940379,
      "grad_norm": 0.41830446575524177,
      "learning_rate": 0.0001999855593472649,
      "loss": 0.6084,
      "step": 604
    },
    {
      "epoch": 0.21067269250825657,
      "grad_norm": 0.8865562086549428,
      "learning_rate": 0.00019998342277910856,
      "loss": 0.6204,
      "step": 606
    },
    {
      "epoch": 0.21136798192247522,
      "grad_norm": 0.5567410910026902,
      "learning_rate": 0.00019998113887825206,
      "loss": 0.7087,
      "step": 608
    },
    {
      "epoch": 0.2120632713366939,
      "grad_norm": 0.4229098163428897,
      "learning_rate": 0.000199978707648061,
      "loss": 0.6403,
      "step": 610
    },
    {
      "epoch": 0.21275856075091257,
      "grad_norm": 0.7793555918286379,
      "learning_rate": 0.00019997612909211784,
      "loss": 0.6147,
      "step": 612
    },
    {
      "epoch": 0.21345385016513124,
      "grad_norm": 0.49302383416102263,
      "learning_rate": 0.00019997340321422228,
      "loss": 0.5771,
      "step": 614
    },
    {
      "epoch": 0.21414913957934992,
      "grad_norm": 1.0807551778698754,
      "learning_rate": 0.00019997053001839115,
      "loss": 0.6538,
      "step": 616
    },
    {
      "epoch": 0.21484442899356856,
      "grad_norm": 0.5935042706034646,
      "learning_rate": 0.0001999675095088583,
      "loss": 0.6058,
      "step": 618
    },
    {
      "epoch": 0.21553971840778724,
      "grad_norm": 0.5151203960605611,
      "learning_rate": 0.00019996434169007468,
      "loss": 0.6138,
      "step": 620
    },
    {
      "epoch": 0.21623500782200591,
      "grad_norm": 0.7837579499467073,
      "learning_rate": 0.00019996102656670824,
      "loss": 0.657,
      "step": 622
    },
    {
      "epoch": 0.2169302972362246,
      "grad_norm": 0.5438202413527335,
      "learning_rate": 0.0001999575641436441,
      "loss": 0.6642,
      "step": 624
    },
    {
      "epoch": 0.21762558665044324,
      "grad_norm": 0.5566641749580342,
      "learning_rate": 0.00019995395442598435,
      "loss": 0.6813,
      "step": 626
    },
    {
      "epoch": 0.2183208760646619,
      "grad_norm": 0.9345190838089675,
      "learning_rate": 0.0001999501974190482,
      "loss": 0.684,
      "step": 628
    },
    {
      "epoch": 0.2190161654788806,
      "grad_norm": 0.48677420225080537,
      "learning_rate": 0.00019994629312837186,
      "loss": 0.6481,
      "step": 630
    },
    {
      "epoch": 0.21971145489309926,
      "grad_norm": 1.3826373483976924,
      "learning_rate": 0.00019994224155970856,
      "loss": 0.7222,
      "step": 632
    },
    {
      "epoch": 0.2204067443073179,
      "grad_norm": 0.9031189580678588,
      "learning_rate": 0.00019993804271902857,
      "loss": 0.6809,
      "step": 634
    },
    {
      "epoch": 0.22110203372153658,
      "grad_norm": 0.5674519795217096,
      "learning_rate": 0.00019993369661251923,
      "loss": 0.6184,
      "step": 636
    },
    {
      "epoch": 0.22179732313575526,
      "grad_norm": 2.5294344265719135,
      "learning_rate": 0.0001999292032465848,
      "loss": 0.7686,
      "step": 638
    },
    {
      "epoch": 0.22249261254997393,
      "grad_norm": 2.8855656605632047,
      "learning_rate": 0.00019992456262784658,
      "loss": 0.8215,
      "step": 640
    },
    {
      "epoch": 0.2231879019641926,
      "grad_norm": 1.5331524159142917,
      "learning_rate": 0.00019991977476314286,
      "loss": 0.7057,
      "step": 642
    },
    {
      "epoch": 0.22388319137841126,
      "grad_norm": 0.8086145005161695,
      "learning_rate": 0.0001999148396595289,
      "loss": 0.6289,
      "step": 644
    },
    {
      "epoch": 0.22457848079262993,
      "grad_norm": 0.767389193699688,
      "learning_rate": 0.00019990975732427693,
      "loss": 0.6117,
      "step": 646
    },
    {
      "epoch": 0.2252737702068486,
      "grad_norm": 0.9275190697426144,
      "learning_rate": 0.0001999045277648761,
      "loss": 0.6582,
      "step": 648
    },
    {
      "epoch": 0.22596905962106728,
      "grad_norm": 0.8004230202091956,
      "learning_rate": 0.00019989915098903257,
      "loss": 0.5387,
      "step": 650
    },
    {
      "epoch": 0.22666434903528593,
      "grad_norm": 0.8210833181388095,
      "learning_rate": 0.0001998936270046694,
      "loss": 0.5941,
      "step": 652
    },
    {
      "epoch": 0.2273596384495046,
      "grad_norm": 2.1189675938584616,
      "learning_rate": 0.00019988795581992656,
      "loss": 0.7504,
      "step": 654
    },
    {
      "epoch": 0.22805492786372328,
      "grad_norm": 0.5952032026226816,
      "learning_rate": 0.000199882137443161,
      "loss": 0.6643,
      "step": 656
    },
    {
      "epoch": 0.22875021727794195,
      "grad_norm": 0.40481972159485846,
      "learning_rate": 0.00019987617188294642,
      "loss": 0.5225,
      "step": 658
    },
    {
      "epoch": 0.2294455066921606,
      "grad_norm": 0.7478677085785272,
      "learning_rate": 0.00019987005914807356,
      "loss": 0.6561,
      "step": 660
    },
    {
      "epoch": 0.23014079610637928,
      "grad_norm": 0.4468692254604696,
      "learning_rate": 0.00019986379924754997,
      "loss": 0.6262,
      "step": 662
    },
    {
      "epoch": 0.23083608552059795,
      "grad_norm": 0.46697718493106893,
      "learning_rate": 0.00019985739219060002,
      "loss": 0.6178,
      "step": 664
    },
    {
      "epoch": 0.23153137493481663,
      "grad_norm": 0.388728867903269,
      "learning_rate": 0.000199850837986665,
      "loss": 0.5544,
      "step": 666
    },
    {
      "epoch": 0.23222666434903527,
      "grad_norm": 0.5731403139849517,
      "learning_rate": 0.000199844136645403,
      "loss": 0.5906,
      "step": 668
    },
    {
      "epoch": 0.23292195376325395,
      "grad_norm": 0.7601215827849259,
      "learning_rate": 0.0001998372881766889,
      "loss": 0.6577,
      "step": 670
    },
    {
      "epoch": 0.23361724317747262,
      "grad_norm": 0.8997786413980279,
      "learning_rate": 0.00019983029259061446,
      "loss": 0.6388,
      "step": 672
    },
    {
      "epoch": 0.2343125325916913,
      "grad_norm": 0.519349430301096,
      "learning_rate": 0.00019982314989748813,
      "loss": 0.6561,
      "step": 674
    },
    {
      "epoch": 0.23500782200590997,
      "grad_norm": 0.4918913590728881,
      "learning_rate": 0.00019981586010783513,
      "loss": 0.6176,
      "step": 676
    },
    {
      "epoch": 0.23570311142012862,
      "grad_norm": 0.5324672662478441,
      "learning_rate": 0.00019980842323239756,
      "loss": 0.5801,
      "step": 678
    },
    {
      "epoch": 0.2363984008343473,
      "grad_norm": 0.6375745879180794,
      "learning_rate": 0.00019980083928213417,
      "loss": 0.6885,
      "step": 680
    },
    {
      "epoch": 0.23709369024856597,
      "grad_norm": 0.4317997282790841,
      "learning_rate": 0.00019979310826822046,
      "loss": 0.5727,
      "step": 682
    },
    {
      "epoch": 0.23778897966278464,
      "grad_norm": 0.7933828103238656,
      "learning_rate": 0.00019978523020204857,
      "loss": 0.6703,
      "step": 684
    },
    {
      "epoch": 0.2384842690770033,
      "grad_norm": 0.5465102255190097,
      "learning_rate": 0.00019977720509522747,
      "loss": 0.6675,
      "step": 686
    },
    {
      "epoch": 0.23917955849122197,
      "grad_norm": 0.8356137363612883,
      "learning_rate": 0.00019976903295958269,
      "loss": 0.6558,
      "step": 688
    },
    {
      "epoch": 0.23987484790544064,
      "grad_norm": 0.5190073187676806,
      "learning_rate": 0.00019976071380715645,
      "loss": 0.6779,
      "step": 690
    },
    {
      "epoch": 0.24057013731965932,
      "grad_norm": 0.40561527338007225,
      "learning_rate": 0.0001997522476502076,
      "loss": 0.6687,
      "step": 692
    },
    {
      "epoch": 0.24126542673387796,
      "grad_norm": 0.592821589706298,
      "learning_rate": 0.0001997436345012117,
      "loss": 0.6807,
      "step": 694
    },
    {
      "epoch": 0.24196071614809664,
      "grad_norm": 0.686918345062074,
      "learning_rate": 0.00019973487437286073,
      "loss": 0.6831,
      "step": 696
    },
    {
      "epoch": 0.2426560055623153,
      "grad_norm": 0.964075508243471,
      "learning_rate": 0.00019972596727806346,
      "loss": 0.665,
      "step": 698
    },
    {
      "epoch": 0.243351294976534,
      "grad_norm": 0.739137434656244,
      "learning_rate": 0.00019971691322994507,
      "loss": 0.6387,
      "step": 700
    },
    {
      "epoch": 0.24404658439075266,
      "grad_norm": 0.3978099419540589,
      "learning_rate": 0.00019970771224184737,
      "loss": 0.6143,
      "step": 702
    },
    {
      "epoch": 0.2447418738049713,
      "grad_norm": 0.4997978607153984,
      "learning_rate": 0.0001996983643273287,
      "loss": 0.6083,
      "step": 704
    },
    {
      "epoch": 0.24543716321918999,
      "grad_norm": 1.3443518380439796,
      "learning_rate": 0.0001996888695001638,
      "loss": 0.6711,
      "step": 706
    },
    {
      "epoch": 0.24613245263340866,
      "grad_norm": 0.5424379336505594,
      "learning_rate": 0.000199679227774344,
      "loss": 0.6321,
      "step": 708
    },
    {
      "epoch": 0.24682774204762734,
      "grad_norm": 0.7689235498835658,
      "learning_rate": 0.00019966943916407712,
      "loss": 0.6918,
      "step": 710
    },
    {
      "epoch": 0.24752303146184598,
      "grad_norm": 0.6467240394468301,
      "learning_rate": 0.00019965950368378734,
      "loss": 0.6368,
      "step": 712
    },
    {
      "epoch": 0.24821832087606466,
      "grad_norm": 0.48819289451999526,
      "learning_rate": 0.00019964942134811532,
      "loss": 0.5875,
      "step": 714
    },
    {
      "epoch": 0.24891361029028333,
      "grad_norm": 1.1929427463467126,
      "learning_rate": 0.00019963919217191807,
      "loss": 0.6795,
      "step": 716
    },
    {
      "epoch": 0.249608899704502,
      "grad_norm": 0.4182713825442439,
      "learning_rate": 0.00019962881617026902,
      "loss": 0.5835,
      "step": 718
    },
    {
      "epoch": 0.25030418911872065,
      "grad_norm": 0.5770262516296942,
      "learning_rate": 0.00019961829335845795,
      "loss": 0.6299,
      "step": 720
    },
    {
      "epoch": 0.25099947853293936,
      "grad_norm": 0.9135250550485389,
      "learning_rate": 0.00019960762375199095,
      "loss": 0.6844,
      "step": 722
    },
    {
      "epoch": 0.251694767947158,
      "grad_norm": 0.5511234805517417,
      "learning_rate": 0.0001995968073665905,
      "loss": 0.5745,
      "step": 724
    },
    {
      "epoch": 0.25239005736137665,
      "grad_norm": 0.49137872896877094,
      "learning_rate": 0.00019958584421819528,
      "loss": 0.698,
      "step": 726
    },
    {
      "epoch": 0.25308534677559535,
      "grad_norm": 0.4437253322782162,
      "learning_rate": 0.00019957473432296026,
      "loss": 0.6864,
      "step": 728
    },
    {
      "epoch": 0.253780636189814,
      "grad_norm": 0.5064381489074585,
      "learning_rate": 0.0001995634776972567,
      "loss": 0.5758,
      "step": 730
    },
    {
      "epoch": 0.2544759256040327,
      "grad_norm": 0.5464655664741616,
      "learning_rate": 0.00019955207435767201,
      "loss": 0.6351,
      "step": 732
    },
    {
      "epoch": 0.25517121501825135,
      "grad_norm": 0.44060524677853835,
      "learning_rate": 0.00019954052432100982,
      "loss": 0.6837,
      "step": 734
    },
    {
      "epoch": 0.25586650443247,
      "grad_norm": 0.942390696980061,
      "learning_rate": 0.00019952882760428998,
      "loss": 0.6378,
      "step": 736
    },
    {
      "epoch": 0.2565617938466887,
      "grad_norm": 0.4710023299938762,
      "learning_rate": 0.00019951698422474836,
      "loss": 0.6968,
      "step": 738
    },
    {
      "epoch": 0.25725708326090735,
      "grad_norm": 0.4179809222387651,
      "learning_rate": 0.00019950499419983707,
      "loss": 0.5643,
      "step": 740
    },
    {
      "epoch": 0.257952372675126,
      "grad_norm": 0.4446278865199289,
      "learning_rate": 0.00019949285754722426,
      "loss": 0.5089,
      "step": 742
    },
    {
      "epoch": 0.2586476620893447,
      "grad_norm": 0.914728851966275,
      "learning_rate": 0.00019948057428479418,
      "loss": 0.7933,
      "step": 744
    },
    {
      "epoch": 0.25934295150356335,
      "grad_norm": 0.47142275185055055,
      "learning_rate": 0.00019946814443064703,
      "loss": 0.6384,
      "step": 746
    },
    {
      "epoch": 0.26003824091778205,
      "grad_norm": 0.720916103725334,
      "learning_rate": 0.00019945556800309917,
      "loss": 0.5957,
      "step": 748
    },
    {
      "epoch": 0.2607335303320007,
      "grad_norm": 0.8550360126036364,
      "learning_rate": 0.00019944284502068275,
      "loss": 0.6454,
      "step": 750
    },
    {
      "epoch": 0.2607335303320007,
      "eval_loss": 0.6373963952064514,
      "eval_runtime": 728.7951,
      "eval_samples_per_second": 6.644,
      "eval_steps_per_second": 0.209,
      "step": 750
    },
    {
      "epoch": 0.26142881974621934,
      "grad_norm": 0.5467512313200382,
      "learning_rate": 0.0001994299755021461,
      "loss": 0.5781,
      "step": 752
    },
    {
      "epoch": 0.26212410916043805,
      "grad_norm": 0.41230384162224254,
      "learning_rate": 0.0001994169594664533,
      "loss": 0.6467,
      "step": 754
    },
    {
      "epoch": 0.2628193985746567,
      "grad_norm": 0.5635895964904281,
      "learning_rate": 0.00019940379693278448,
      "loss": 0.6182,
      "step": 756
    },
    {
      "epoch": 0.2635146879888754,
      "grad_norm": 0.6456168984233462,
      "learning_rate": 0.0001993904879205355,
      "loss": 0.6086,
      "step": 758
    },
    {
      "epoch": 0.26420997740309404,
      "grad_norm": 0.5060399335833471,
      "learning_rate": 0.00019937703244931815,
      "loss": 0.6407,
      "step": 760
    },
    {
      "epoch": 0.2649052668173127,
      "grad_norm": 0.5033297464646784,
      "learning_rate": 0.00019936343053896004,
      "loss": 0.6157,
      "step": 762
    },
    {
      "epoch": 0.2656005562315314,
      "grad_norm": 0.3911665523252255,
      "learning_rate": 0.00019934968220950458,
      "loss": 0.6227,
      "step": 764
    },
    {
      "epoch": 0.26629584564575004,
      "grad_norm": 1.7130937466256138,
      "learning_rate": 0.00019933578748121086,
      "loss": 0.5994,
      "step": 766
    },
    {
      "epoch": 0.2669911350599687,
      "grad_norm": 0.5747338886603724,
      "learning_rate": 0.00019932174637455382,
      "loss": 0.6238,
      "step": 768
    },
    {
      "epoch": 0.2676864244741874,
      "grad_norm": 0.6226324357252033,
      "learning_rate": 0.00019930755891022398,
      "loss": 0.6255,
      "step": 770
    },
    {
      "epoch": 0.26838171388840604,
      "grad_norm": 0.5282488794229044,
      "learning_rate": 0.00019929322510912756,
      "loss": 0.5808,
      "step": 772
    },
    {
      "epoch": 0.26907700330262474,
      "grad_norm": 0.5596904577167024,
      "learning_rate": 0.0001992787449923865,
      "loss": 0.5361,
      "step": 774
    },
    {
      "epoch": 0.2697722927168434,
      "grad_norm": 1.0477422618379753,
      "learning_rate": 0.00019926411858133824,
      "loss": 0.54,
      "step": 776
    },
    {
      "epoch": 0.27046758213106203,
      "grad_norm": 1.1314704495635823,
      "learning_rate": 0.00019924934589753582,
      "loss": 0.5869,
      "step": 778
    },
    {
      "epoch": 0.27116287154528074,
      "grad_norm": 0.5426548544381972,
      "learning_rate": 0.00019923442696274794,
      "loss": 0.5951,
      "step": 780
    },
    {
      "epoch": 0.2718581609594994,
      "grad_norm": 0.8480025293040686,
      "learning_rate": 0.00019921936179895862,
      "loss": 0.6003,
      "step": 782
    },
    {
      "epoch": 0.2725534503737181,
      "grad_norm": 0.5729452380906337,
      "learning_rate": 0.0001992041504283675,
      "loss": 0.6526,
      "step": 784
    },
    {
      "epoch": 0.27324873978793673,
      "grad_norm": 0.7362085286496177,
      "learning_rate": 0.00019918879287338957,
      "loss": 0.6776,
      "step": 786
    },
    {
      "epoch": 0.2739440292021554,
      "grad_norm": 0.7440059372391256,
      "learning_rate": 0.00019917328915665535,
      "loss": 0.6486,
      "step": 788
    },
    {
      "epoch": 0.2746393186163741,
      "grad_norm": 0.6110842206790659,
      "learning_rate": 0.0001991576393010106,
      "loss": 0.6226,
      "step": 790
    },
    {
      "epoch": 0.27533460803059273,
      "grad_norm": 0.8002336898560896,
      "learning_rate": 0.00019914184332951655,
      "loss": 0.6215,
      "step": 792
    },
    {
      "epoch": 0.2760298974448114,
      "grad_norm": 0.6761736356192558,
      "learning_rate": 0.00019912590126544964,
      "loss": 0.5988,
      "step": 794
    },
    {
      "epoch": 0.2767251868590301,
      "grad_norm": 1.1005016630123619,
      "learning_rate": 0.00019910981313230172,
      "loss": 0.6213,
      "step": 796
    },
    {
      "epoch": 0.27742047627324873,
      "grad_norm": 0.6513452605734942,
      "learning_rate": 0.00019909357895377973,
      "loss": 0.682,
      "step": 798
    },
    {
      "epoch": 0.27811576568746743,
      "grad_norm": 0.8536492677008404,
      "learning_rate": 0.0001990771987538059,
      "loss": 0.707,
      "step": 800
    },
    {
      "epoch": 0.2788110551016861,
      "grad_norm": 0.3362293750538688,
      "learning_rate": 0.00019906067255651765,
      "loss": 0.7108,
      "step": 802
    },
    {
      "epoch": 0.2795063445159047,
      "grad_norm": 0.4792899916449288,
      "learning_rate": 0.0001990440003862675,
      "loss": 0.6055,
      "step": 804
    },
    {
      "epoch": 0.28020163393012343,
      "grad_norm": 1.272937677307704,
      "learning_rate": 0.00019902718226762304,
      "loss": 0.6816,
      "step": 806
    },
    {
      "epoch": 0.2808969233443421,
      "grad_norm": 0.30646692495778655,
      "learning_rate": 0.00019901021822536704,
      "loss": 0.688,
      "step": 808
    },
    {
      "epoch": 0.2815922127585608,
      "grad_norm": 0.3035733218665055,
      "learning_rate": 0.00019899310828449713,
      "loss": 0.6746,
      "step": 810
    },
    {
      "epoch": 0.2822875021727794,
      "grad_norm": 0.2763276026826459,
      "learning_rate": 0.00019897585247022613,
      "loss": 0.6632,
      "step": 812
    },
    {
      "epoch": 0.2829827915869981,
      "grad_norm": 1.173118337900773,
      "learning_rate": 0.00019895845080798166,
      "loss": 0.6306,
      "step": 814
    },
    {
      "epoch": 0.2836780810012168,
      "grad_norm": 0.4335278252167635,
      "learning_rate": 0.0001989409033234063,
      "loss": 0.6147,
      "step": 816
    },
    {
      "epoch": 0.2843733704154354,
      "grad_norm": 0.3556694503062785,
      "learning_rate": 0.00019892321004235755,
      "loss": 0.5771,
      "step": 818
    },
    {
      "epoch": 0.28506865982965407,
      "grad_norm": 0.9083240381897224,
      "learning_rate": 0.00019890537099090768,
      "loss": 0.5729,
      "step": 820
    },
    {
      "epoch": 0.2857639492438728,
      "grad_norm": 0.5433346815141633,
      "learning_rate": 0.00019888738619534385,
      "loss": 0.5554,
      "step": 822
    },
    {
      "epoch": 0.2864592386580914,
      "grad_norm": 0.5244705156138804,
      "learning_rate": 0.0001988692556821679,
      "loss": 0.6525,
      "step": 824
    },
    {
      "epoch": 0.2871545280723101,
      "grad_norm": 0.7580866792170871,
      "learning_rate": 0.00019885097947809648,
      "loss": 0.6512,
      "step": 826
    },
    {
      "epoch": 0.28784981748652877,
      "grad_norm": 0.7034554538404351,
      "learning_rate": 0.00019883255761006082,
      "loss": 0.5414,
      "step": 828
    },
    {
      "epoch": 0.2885451069007474,
      "grad_norm": 0.6915575597289163,
      "learning_rate": 0.00019881399010520688,
      "loss": 0.6036,
      "step": 830
    },
    {
      "epoch": 0.2892403963149661,
      "grad_norm": 0.6895372001781882,
      "learning_rate": 0.00019879527699089524,
      "loss": 0.5894,
      "step": 832
    },
    {
      "epoch": 0.28993568572918477,
      "grad_norm": 0.7762412863407715,
      "learning_rate": 0.00019877641829470094,
      "loss": 0.7115,
      "step": 834
    },
    {
      "epoch": 0.29063097514340347,
      "grad_norm": 0.5761220663639801,
      "learning_rate": 0.00019875741404441367,
      "loss": 0.6108,
      "step": 836
    },
    {
      "epoch": 0.2913262645576221,
      "grad_norm": 0.47176943357070505,
      "learning_rate": 0.00019873826426803755,
      "loss": 0.634,
      "step": 838
    },
    {
      "epoch": 0.29202155397184076,
      "grad_norm": 0.5985873148196751,
      "learning_rate": 0.00019871896899379107,
      "loss": 0.6528,
      "step": 840
    },
    {
      "epoch": 0.29271684338605947,
      "grad_norm": 0.39514741111190665,
      "learning_rate": 0.00019869952825010727,
      "loss": 0.6034,
      "step": 842
    },
    {
      "epoch": 0.2934121328002781,
      "grad_norm": 0.29787585594263405,
      "learning_rate": 0.00019867994206563343,
      "loss": 0.6063,
      "step": 844
    },
    {
      "epoch": 0.29410742221449676,
      "grad_norm": 0.303320759302155,
      "learning_rate": 0.00019866021046923118,
      "loss": 0.6343,
      "step": 846
    },
    {
      "epoch": 0.29480271162871546,
      "grad_norm": 0.33135450527244925,
      "learning_rate": 0.00019864033348997645,
      "loss": 0.6421,
      "step": 848
    },
    {
      "epoch": 0.2954980010429341,
      "grad_norm": 0.553668190192523,
      "learning_rate": 0.0001986203111571594,
      "loss": 0.6503,
      "step": 850
    },
    {
      "epoch": 0.2961932904571528,
      "grad_norm": 0.31948016623126946,
      "learning_rate": 0.00019860014350028438,
      "loss": 0.6259,
      "step": 852
    },
    {
      "epoch": 0.29688857987137146,
      "grad_norm": 0.5325237443938606,
      "learning_rate": 0.0001985798305490698,
      "loss": 0.6207,
      "step": 854
    },
    {
      "epoch": 0.2975838692855901,
      "grad_norm": 0.5093186589927414,
      "learning_rate": 0.00019855937233344831,
      "loss": 0.5397,
      "step": 856
    },
    {
      "epoch": 0.2982791586998088,
      "grad_norm": 0.5220573948537062,
      "learning_rate": 0.00019853876888356652,
      "loss": 0.6237,
      "step": 858
    },
    {
      "epoch": 0.29897444811402746,
      "grad_norm": 0.9861332892020862,
      "learning_rate": 0.00019851802022978506,
      "loss": 0.689,
      "step": 860
    },
    {
      "epoch": 0.2996697375282461,
      "grad_norm": 0.464669721879274,
      "learning_rate": 0.00019849712640267861,
      "loss": 0.522,
      "step": 862
    },
    {
      "epoch": 0.3003650269424648,
      "grad_norm": 0.9223435358714303,
      "learning_rate": 0.00019847608743303567,
      "loss": 0.7491,
      "step": 864
    },
    {
      "epoch": 0.30106031635668346,
      "grad_norm": 0.5058775377593727,
      "learning_rate": 0.00019845490335185866,
      "loss": 0.562,
      "step": 866
    },
    {
      "epoch": 0.30175560577090216,
      "grad_norm": 0.5517767811356443,
      "learning_rate": 0.00019843357419036382,
      "loss": 0.6162,
      "step": 868
    },
    {
      "epoch": 0.3024508951851208,
      "grad_norm": 0.49255497197537723,
      "learning_rate": 0.00019841209997998127,
      "loss": 0.6803,
      "step": 870
    },
    {
      "epoch": 0.30314618459933945,
      "grad_norm": 0.41553745561512617,
      "learning_rate": 0.0001983904807523547,
      "loss": 0.6415,
      "step": 872
    },
    {
      "epoch": 0.30384147401355815,
      "grad_norm": 0.49551628457734653,
      "learning_rate": 0.00019836871653934162,
      "loss": 0.6176,
      "step": 874
    },
    {
      "epoch": 0.3045367634277768,
      "grad_norm": 0.7489091107060393,
      "learning_rate": 0.00019834680737301313,
      "loss": 0.6337,
      "step": 876
    },
    {
      "epoch": 0.3052320528419955,
      "grad_norm": 0.32312869533576805,
      "learning_rate": 0.00019832475328565398,
      "loss": 0.6135,
      "step": 878
    },
    {
      "epoch": 0.30592734225621415,
      "grad_norm": 0.304002075332943,
      "learning_rate": 0.00019830255430976242,
      "loss": 0.5533,
      "step": 880
    },
    {
      "epoch": 0.3066226316704328,
      "grad_norm": 0.4137621036041215,
      "learning_rate": 0.00019828021047805022,
      "loss": 0.573,
      "step": 882
    },
    {
      "epoch": 0.3073179210846515,
      "grad_norm": 0.7506870255042438,
      "learning_rate": 0.00019825772182344262,
      "loss": 0.6971,
      "step": 884
    },
    {
      "epoch": 0.30801321049887015,
      "grad_norm": 0.7069489041589112,
      "learning_rate": 0.00019823508837907828,
      "loss": 0.5848,
      "step": 886
    },
    {
      "epoch": 0.3087084999130888,
      "grad_norm": 0.49383355752727304,
      "learning_rate": 0.00019821231017830914,
      "loss": 0.6349,
      "step": 888
    },
    {
      "epoch": 0.3094037893273075,
      "grad_norm": 0.7893505446859834,
      "learning_rate": 0.0001981893872547005,
      "loss": 0.6335,
      "step": 890
    },
    {
      "epoch": 0.31009907874152615,
      "grad_norm": 1.0962653113728835,
      "learning_rate": 0.00019816631964203097,
      "loss": 0.6438,
      "step": 892
    },
    {
      "epoch": 0.31079436815574485,
      "grad_norm": 0.40606329821748216,
      "learning_rate": 0.0001981431073742923,
      "loss": 0.557,
      "step": 894
    },
    {
      "epoch": 0.3114896575699635,
      "grad_norm": 0.8061837126172193,
      "learning_rate": 0.00019811975048568943,
      "loss": 0.6334,
      "step": 896
    },
    {
      "epoch": 0.31218494698418214,
      "grad_norm": 0.7808955990860935,
      "learning_rate": 0.00019809624901064038,
      "loss": 0.5775,
      "step": 898
    },
    {
      "epoch": 0.31288023639840085,
      "grad_norm": 0.5527203146534614,
      "learning_rate": 0.00019807260298377626,
      "loss": 0.5934,
      "step": 900
    },
    {
      "epoch": 0.31288023639840085,
      "eval_loss": 0.666339099407196,
      "eval_runtime": 759.5196,
      "eval_samples_per_second": 6.375,
      "eval_steps_per_second": 0.2,
      "step": 900
    },
    {
      "epoch": 0.3135755258126195,
      "grad_norm": 1.151650071753606,
      "learning_rate": 0.00019804881243994118,
      "loss": 0.6459,
      "step": 902
    },
    {
      "epoch": 0.3142708152268382,
      "grad_norm": 0.37537177441864283,
      "learning_rate": 0.00019802487741419218,
      "loss": 0.5537,
      "step": 904
    },
    {
      "epoch": 0.31496610464105684,
      "grad_norm": 0.39806583735978385,
      "learning_rate": 0.00019800079794179927,
      "loss": 0.5765,
      "step": 906
    },
    {
      "epoch": 0.3156613940552755,
      "grad_norm": 0.9252532303995283,
      "learning_rate": 0.00019797657405824524,
      "loss": 0.6581,
      "step": 908
    },
    {
      "epoch": 0.3163566834694942,
      "grad_norm": 0.4242008643262632,
      "learning_rate": 0.00019795220579922572,
      "loss": 0.663,
      "step": 910
    },
    {
      "epoch": 0.31705197288371284,
      "grad_norm": 0.5557863138791925,
      "learning_rate": 0.00019792769320064904,
      "loss": 0.6492,
      "step": 912
    },
    {
      "epoch": 0.3177472622979315,
      "grad_norm": 0.5743017982975046,
      "learning_rate": 0.0001979030362986363,
      "loss": 0.6425,
      "step": 914
    },
    {
      "epoch": 0.3184425517121502,
      "grad_norm": 0.39667228882787314,
      "learning_rate": 0.0001978782351295212,
      "loss": 0.5658,
      "step": 916
    },
    {
      "epoch": 0.31913784112636884,
      "grad_norm": 1.2742981139875873,
      "learning_rate": 0.00019785328972985,
      "loss": 0.6042,
      "step": 918
    },
    {
      "epoch": 0.31983313054058754,
      "grad_norm": 0.7520790754771111,
      "learning_rate": 0.00019782820013638158,
      "loss": 0.6248,
      "step": 920
    },
    {
      "epoch": 0.3205284199548062,
      "grad_norm": 1.1777266516894538,
      "learning_rate": 0.0001978029663860872,
      "loss": 0.6394,
      "step": 922
    },
    {
      "epoch": 0.32122370936902483,
      "grad_norm": 0.5383416828808074,
      "learning_rate": 0.00019777758851615058,
      "loss": 0.6357,
      "step": 924
    },
    {
      "epoch": 0.32191899878324354,
      "grad_norm": 0.5351088818608489,
      "learning_rate": 0.00019775206656396787,
      "loss": 0.6111,
      "step": 926
    },
    {
      "epoch": 0.3226142881974622,
      "grad_norm": 0.7776255734128178,
      "learning_rate": 0.00019772640056714744,
      "loss": 0.5778,
      "step": 928
    },
    {
      "epoch": 0.3233095776116809,
      "grad_norm": 0.5049904332607067,
      "learning_rate": 0.00019770059056351,
      "loss": 0.5978,
      "step": 930
    },
    {
      "epoch": 0.32400486702589953,
      "grad_norm": 0.6894813643690206,
      "learning_rate": 0.00019767463659108841,
      "loss": 0.6727,
      "step": 932
    },
    {
      "epoch": 0.3247001564401182,
      "grad_norm": 0.6230252249989028,
      "learning_rate": 0.00019764853868812772,
      "loss": 0.5911,
      "step": 934
    },
    {
      "epoch": 0.3253954458543369,
      "grad_norm": 0.6699617199619087,
      "learning_rate": 0.00019762229689308499,
      "loss": 0.6694,
      "step": 936
    },
    {
      "epoch": 0.32609073526855553,
      "grad_norm": 0.9762605521595761,
      "learning_rate": 0.00019759591124462943,
      "loss": 0.7053,
      "step": 938
    },
    {
      "epoch": 0.3267860246827742,
      "grad_norm": 0.5216728233794251,
      "learning_rate": 0.0001975693817816422,
      "loss": 0.6958,
      "step": 940
    },
    {
      "epoch": 0.3274813140969929,
      "grad_norm": 0.5943791708445256,
      "learning_rate": 0.00019754270854321625,
      "loss": 0.6342,
      "step": 942
    },
    {
      "epoch": 0.32817660351121153,
      "grad_norm": 0.5341014737913188,
      "learning_rate": 0.00019751589156865663,
      "loss": 0.6272,
      "step": 944
    },
    {
      "epoch": 0.32887189292543023,
      "grad_norm": 0.8411647140863245,
      "learning_rate": 0.00019748893089747995,
      "loss": 0.6041,
      "step": 946
    },
    {
      "epoch": 0.3295671823396489,
      "grad_norm": 1.072323043427063,
      "learning_rate": 0.00019746182656941473,
      "loss": 0.7152,
      "step": 948
    },
    {
      "epoch": 0.3302624717538675,
      "grad_norm": 0.6497829380326366,
      "learning_rate": 0.00019743457862440115,
      "loss": 0.6176,
      "step": 950
    },
    {
      "epoch": 0.33095776116808623,
      "grad_norm": 0.28736093186011447,
      "learning_rate": 0.00019740718710259096,
      "loss": 0.6453,
      "step": 952
    },
    {
      "epoch": 0.3316530505823049,
      "grad_norm": 0.27868233108109625,
      "learning_rate": 0.00019737965204434757,
      "loss": 0.6051,
      "step": 954
    },
    {
      "epoch": 0.3323483399965236,
      "grad_norm": 0.40709235855818693,
      "learning_rate": 0.00019735197349024576,
      "loss": 0.6255,
      "step": 956
    },
    {
      "epoch": 0.3330436294107422,
      "grad_norm": 0.8385677925045294,
      "learning_rate": 0.00019732415148107199,
      "loss": 0.6455,
      "step": 958
    },
    {
      "epoch": 0.3337389188249609,
      "grad_norm": 0.5642576200414804,
      "learning_rate": 0.00019729618605782384,
      "loss": 0.6971,
      "step": 960
    },
    {
      "epoch": 0.3344342082391796,
      "grad_norm": 0.7034648545079693,
      "learning_rate": 0.00019726807726171039,
      "loss": 0.6177,
      "step": 962
    },
    {
      "epoch": 0.3351294976533982,
      "grad_norm": 1.9840633930320113,
      "learning_rate": 0.000197239825134152,
      "loss": 0.6776,
      "step": 964
    },
    {
      "epoch": 0.33582478706761687,
      "grad_norm": 1.0091982574836484,
      "learning_rate": 0.00019721142971678015,
      "loss": 0.6893,
      "step": 966
    },
    {
      "epoch": 0.3365200764818356,
      "grad_norm": 0.9742560258590767,
      "learning_rate": 0.00019718289105143753,
      "loss": 0.744,
      "step": 968
    },
    {
      "epoch": 0.3372153658960542,
      "grad_norm": 0.6897018399345455,
      "learning_rate": 0.00019715420918017793,
      "loss": 0.678,
      "step": 970
    },
    {
      "epoch": 0.3379106553102729,
      "grad_norm": 0.29102959771453246,
      "learning_rate": 0.00019712538414526606,
      "loss": 0.6663,
      "step": 972
    },
    {
      "epoch": 0.33860594472449157,
      "grad_norm": 0.7337107483377766,
      "learning_rate": 0.0001970964159891777,
      "loss": 0.663,
      "step": 974
    },
    {
      "epoch": 0.3393012341387102,
      "grad_norm": 0.5817704647699353,
      "learning_rate": 0.00019706730475459953,
      "loss": 0.6398,
      "step": 976
    },
    {
      "epoch": 0.3399965235529289,
      "grad_norm": 0.28703428796704483,
      "learning_rate": 0.00019703805048442897,
      "loss": 0.5906,
      "step": 978
    },
    {
      "epoch": 0.34069181296714757,
      "grad_norm": 0.41383789019772477,
      "learning_rate": 0.0001970086532217743,
      "loss": 0.6709,
      "step": 980
    },
    {
      "epoch": 0.34138710238136627,
      "grad_norm": 0.812487649001141,
      "learning_rate": 0.00019697911300995443,
      "loss": 0.6191,
      "step": 982
    },
    {
      "epoch": 0.3420823917955849,
      "grad_norm": 0.798027200072012,
      "learning_rate": 0.00019694942989249907,
      "loss": 0.6608,
      "step": 984
    },
    {
      "epoch": 0.34277768120980356,
      "grad_norm": 0.44029385955900757,
      "learning_rate": 0.00019691960391314837,
      "loss": 0.647,
      "step": 986
    },
    {
      "epoch": 0.34347297062402227,
      "grad_norm": 0.3824484030698272,
      "learning_rate": 0.00019688963511585295,
      "loss": 0.6378,
      "step": 988
    },
    {
      "epoch": 0.3441682600382409,
      "grad_norm": 0.4121768227084979,
      "learning_rate": 0.0001968595235447741,
      "loss": 0.5908,
      "step": 990
    },
    {
      "epoch": 0.34486354945245956,
      "grad_norm": 0.5154693781246049,
      "learning_rate": 0.0001968292692442833,
      "loss": 0.5632,
      "step": 992
    },
    {
      "epoch": 0.34555883886667826,
      "grad_norm": 0.3895510966829308,
      "learning_rate": 0.0001967988722589624,
      "loss": 0.4674,
      "step": 994
    },
    {
      "epoch": 0.3462541282808969,
      "grad_norm": 0.507531875733667,
      "learning_rate": 0.00019676833263360352,
      "loss": 0.5581,
      "step": 996
    },
    {
      "epoch": 0.3469494176951156,
      "grad_norm": 0.5476523355263471,
      "learning_rate": 0.00019673765041320907,
      "loss": 0.7421,
      "step": 998
    },
    {
      "epoch": 0.34764470710933426,
      "grad_norm": 0.8417172933340035,
      "learning_rate": 0.00019670682564299136,
      "loss": 0.6774,
      "step": 1000
    },
    {
      "epoch": 0.3483399965235529,
      "grad_norm": 0.6717112677412562,
      "learning_rate": 0.00019667585836837299,
      "loss": 0.6515,
      "step": 1002
    },
    {
      "epoch": 0.3490352859377716,
      "grad_norm": 0.7599904388695796,
      "learning_rate": 0.0001966447486349864,
      "loss": 0.5679,
      "step": 1004
    },
    {
      "epoch": 0.34973057535199026,
      "grad_norm": 0.44186748583335306,
      "learning_rate": 0.000196613496488674,
      "loss": 0.6067,
      "step": 1006
    },
    {
      "epoch": 0.35042586476620896,
      "grad_norm": 0.4287165077907837,
      "learning_rate": 0.00019658210197548805,
      "loss": 0.5706,
      "step": 1008
    },
    {
      "epoch": 0.3511211541804276,
      "grad_norm": 0.6051489125727973,
      "learning_rate": 0.0001965505651416906,
      "loss": 0.6178,
      "step": 1010
    },
    {
      "epoch": 0.35181644359464626,
      "grad_norm": 0.5003034918118222,
      "learning_rate": 0.00019651888603375346,
      "loss": 0.6,
      "step": 1012
    },
    {
      "epoch": 0.35251173300886496,
      "grad_norm": 0.6774513066433009,
      "learning_rate": 0.00019648706469835804,
      "loss": 0.6248,
      "step": 1014
    },
    {
      "epoch": 0.3532070224230836,
      "grad_norm": 0.5666799893616385,
      "learning_rate": 0.0001964551011823953,
      "loss": 0.6256,
      "step": 1016
    },
    {
      "epoch": 0.35390231183730225,
      "grad_norm": 0.6185519925235329,
      "learning_rate": 0.00019642299553296582,
      "loss": 0.5836,
      "step": 1018
    },
    {
      "epoch": 0.35459760125152096,
      "grad_norm": 0.5020407783730059,
      "learning_rate": 0.0001963907477973795,
      "loss": 0.5148,
      "step": 1020
    },
    {
      "epoch": 0.3552928906657396,
      "grad_norm": 0.966855697270511,
      "learning_rate": 0.00019635835802315574,
      "loss": 0.6335,
      "step": 1022
    },
    {
      "epoch": 0.3559881800799583,
      "grad_norm": 0.8705091175729548,
      "learning_rate": 0.00019632582625802317,
      "loss": 0.6313,
      "step": 1024
    },
    {
      "epoch": 0.35668346949417695,
      "grad_norm": 0.6028556619755229,
      "learning_rate": 0.00019629315254991964,
      "loss": 0.6483,
      "step": 1026
    },
    {
      "epoch": 0.3573787589083956,
      "grad_norm": 0.6075038119620636,
      "learning_rate": 0.00019626033694699214,
      "loss": 0.6271,
      "step": 1028
    },
    {
      "epoch": 0.3580740483226143,
      "grad_norm": 0.7923956541669288,
      "learning_rate": 0.00019622737949759694,
      "loss": 0.6338,
      "step": 1030
    },
    {
      "epoch": 0.35876933773683295,
      "grad_norm": 1.2067992138100796,
      "learning_rate": 0.00019619428025029905,
      "loss": 0.6308,
      "step": 1032
    },
    {
      "epoch": 0.35946462715105165,
      "grad_norm": 0.5446351671940789,
      "learning_rate": 0.00019616103925387265,
      "loss": 0.6475,
      "step": 1034
    },
    {
      "epoch": 0.3601599165652703,
      "grad_norm": 0.8842474031361561,
      "learning_rate": 0.0001961276565573007,
      "loss": 0.6654,
      "step": 1036
    },
    {
      "epoch": 0.36085520597948895,
      "grad_norm": 1.294693850012853,
      "learning_rate": 0.00019609413220977496,
      "loss": 0.6929,
      "step": 1038
    },
    {
      "epoch": 0.36155049539370765,
      "grad_norm": 0.7435682846586636,
      "learning_rate": 0.00019606046626069595,
      "loss": 0.6791,
      "step": 1040
    },
    {
      "epoch": 0.3622457848079263,
      "grad_norm": 0.45762946459115417,
      "learning_rate": 0.0001960266587596729,
      "loss": 0.5769,
      "step": 1042
    },
    {
      "epoch": 0.36294107422214494,
      "grad_norm": 0.5614638042598611,
      "learning_rate": 0.00019599270975652352,
      "loss": 0.6047,
      "step": 1044
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.6171161059961894,
      "learning_rate": 0.0001959586193012741,
      "loss": 0.6932,
      "step": 1046
    },
    {
      "epoch": 0.3643316530505823,
      "grad_norm": 1.5847051714441287,
      "learning_rate": 0.00019592438744415932,
      "loss": 0.5908,
      "step": 1048
    },
    {
      "epoch": 0.365026942464801,
      "grad_norm": 0.4282438415059217,
      "learning_rate": 0.00019589001423562233,
      "loss": 0.6749,
      "step": 1050
    },
    {
      "epoch": 0.365026942464801,
      "eval_loss": 0.6080955266952515,
      "eval_runtime": 710.9824,
      "eval_samples_per_second": 6.81,
      "eval_steps_per_second": 0.214,
      "step": 1050
    },
    {
      "epoch": 0.36572223187901964,
      "grad_norm": 1.0612232842206784,
      "learning_rate": 0.00019585549972631446,
      "loss": 0.5669,
      "step": 1052
    },
    {
      "epoch": 0.3664175212932383,
      "grad_norm": 2.6994673924740358,
      "learning_rate": 0.0001958208439670953,
      "loss": 0.6455,
      "step": 1054
    },
    {
      "epoch": 0.367112810707457,
      "grad_norm": 2.022628249772274,
      "learning_rate": 0.0001957860470090326,
      "loss": 0.6395,
      "step": 1056
    },
    {
      "epoch": 0.36780810012167564,
      "grad_norm": 0.607126211578616,
      "learning_rate": 0.00019575110890340214,
      "loss": 0.593,
      "step": 1058
    },
    {
      "epoch": 0.36850338953589434,
      "grad_norm": 1.026539890410463,
      "learning_rate": 0.00019571602970168775,
      "loss": 0.5939,
      "step": 1060
    },
    {
      "epoch": 0.369198678950113,
      "grad_norm": 0.6663599873173821,
      "learning_rate": 0.00019568080945558104,
      "loss": 0.6415,
      "step": 1062
    },
    {
      "epoch": 0.36989396836433164,
      "grad_norm": 0.5967439670789174,
      "learning_rate": 0.00019564544821698167,
      "loss": 0.6348,
      "step": 1064
    },
    {
      "epoch": 0.37058925777855034,
      "grad_norm": 0.6468802992284401,
      "learning_rate": 0.00019560994603799682,
      "loss": 0.5991,
      "step": 1066
    },
    {
      "epoch": 0.371284547192769,
      "grad_norm": 1.2251498168873143,
      "learning_rate": 0.00019557430297094158,
      "loss": 0.6001,
      "step": 1068
    },
    {
      "epoch": 0.37197983660698763,
      "grad_norm": 0.4681978143920913,
      "learning_rate": 0.00019553851906833853,
      "loss": 0.5664,
      "step": 1070
    },
    {
      "epoch": 0.37267512602120634,
      "grad_norm": 0.7538053079694034,
      "learning_rate": 0.00019550259438291782,
      "loss": 0.5531,
      "step": 1072
    },
    {
      "epoch": 0.373370415435425,
      "grad_norm": 0.4626868039226879,
      "learning_rate": 0.00019546652896761696,
      "loss": 0.6268,
      "step": 1074
    },
    {
      "epoch": 0.3740657048496437,
      "grad_norm": 0.8012577314135656,
      "learning_rate": 0.00019543032287558097,
      "loss": 0.7819,
      "step": 1076
    },
    {
      "epoch": 0.37476099426386233,
      "grad_norm": 0.7004564506452116,
      "learning_rate": 0.0001953939761601621,
      "loss": 0.6505,
      "step": 1078
    },
    {
      "epoch": 0.375456283678081,
      "grad_norm": 1.352602706017517,
      "learning_rate": 0.0001953574888749198,
      "loss": 0.558,
      "step": 1080
    },
    {
      "epoch": 0.3761515730922997,
      "grad_norm": 0.3012261239096098,
      "learning_rate": 0.0001953208610736207,
      "loss": 0.5746,
      "step": 1082
    },
    {
      "epoch": 0.37684686250651833,
      "grad_norm": 0.49798092264749827,
      "learning_rate": 0.0001952840928102385,
      "loss": 0.5845,
      "step": 1084
    },
    {
      "epoch": 0.377542151920737,
      "grad_norm": 0.4457559765569793,
      "learning_rate": 0.00019524718413895382,
      "loss": 0.5859,
      "step": 1086
    },
    {
      "epoch": 0.3782374413349557,
      "grad_norm": 0.5943406854432696,
      "learning_rate": 0.00019521013511415426,
      "loss": 0.6581,
      "step": 1088
    },
    {
      "epoch": 0.37893273074917433,
      "grad_norm": 1.8614189583072707,
      "learning_rate": 0.00019517294579043414,
      "loss": 0.7019,
      "step": 1090
    },
    {
      "epoch": 0.37962802016339303,
      "grad_norm": 0.4212993231373999,
      "learning_rate": 0.00019513561622259466,
      "loss": 0.6106,
      "step": 1092
    },
    {
      "epoch": 0.3803233095776117,
      "grad_norm": 0.5224177310348964,
      "learning_rate": 0.00019509814646564354,
      "loss": 0.6012,
      "step": 1094
    },
    {
      "epoch": 0.3810185989918303,
      "grad_norm": 1.3980998842985117,
      "learning_rate": 0.0001950605365747951,
      "loss": 0.5875,
      "step": 1096
    },
    {
      "epoch": 0.38171388840604903,
      "grad_norm": 0.7255638165052757,
      "learning_rate": 0.0001950227866054703,
      "loss": 0.6366,
      "step": 1098
    },
    {
      "epoch": 0.3824091778202677,
      "grad_norm": 0.39401204779232035,
      "learning_rate": 0.00019498489661329632,
      "loss": 0.59,
      "step": 1100
    },
    {
      "epoch": 0.3831044672344864,
      "grad_norm": 0.6249434294547875,
      "learning_rate": 0.00019494686665410684,
      "loss": 0.5555,
      "step": 1102
    },
    {
      "epoch": 0.383799756648705,
      "grad_norm": 0.4970609140814689,
      "learning_rate": 0.00019490869678394165,
      "loss": 0.5872,
      "step": 1104
    },
    {
      "epoch": 0.3844950460629237,
      "grad_norm": 0.5522933346523479,
      "learning_rate": 0.0001948703870590468,
      "loss": 0.6296,
      "step": 1106
    },
    {
      "epoch": 0.3851903354771424,
      "grad_norm": 0.5358444643818808,
      "learning_rate": 0.0001948319375358744,
      "loss": 0.5677,
      "step": 1108
    },
    {
      "epoch": 0.385885624891361,
      "grad_norm": 0.5839030994818908,
      "learning_rate": 0.00019479334827108256,
      "loss": 0.5443,
      "step": 1110
    },
    {
      "epoch": 0.38658091430557967,
      "grad_norm": 0.9808370348899028,
      "learning_rate": 0.00019475461932153533,
      "loss": 0.6703,
      "step": 1112
    },
    {
      "epoch": 0.3872762037197984,
      "grad_norm": 0.5567454433782846,
      "learning_rate": 0.00019471575074430256,
      "loss": 0.585,
      "step": 1114
    },
    {
      "epoch": 0.387971493134017,
      "grad_norm": 0.8706120804701404,
      "learning_rate": 0.00019467674259665985,
      "loss": 0.5399,
      "step": 1116
    },
    {
      "epoch": 0.3886667825482357,
      "grad_norm": 0.607175198823171,
      "learning_rate": 0.00019463759493608854,
      "loss": 0.6016,
      "step": 1118
    },
    {
      "epoch": 0.38936207196245437,
      "grad_norm": 0.848040345554999,
      "learning_rate": 0.0001945983078202754,
      "loss": 0.5604,
      "step": 1120
    },
    {
      "epoch": 0.390057361376673,
      "grad_norm": 0.39050505941263947,
      "learning_rate": 0.00019455888130711296,
      "loss": 0.5391,
      "step": 1122
    },
    {
      "epoch": 0.3907526507908917,
      "grad_norm": 0.43688262723639604,
      "learning_rate": 0.00019451931545469883,
      "loss": 0.5911,
      "step": 1124
    },
    {
      "epoch": 0.39144794020511037,
      "grad_norm": 0.5216486488482895,
      "learning_rate": 0.00019447961032133623,
      "loss": 0.7729,
      "step": 1126
    },
    {
      "epoch": 0.39214322961932907,
      "grad_norm": 0.5238744592921366,
      "learning_rate": 0.0001944397659655334,
      "loss": 0.6561,
      "step": 1128
    },
    {
      "epoch": 0.3928385190335477,
      "grad_norm": 0.5356265007770037,
      "learning_rate": 0.00019439978244600392,
      "loss": 0.6139,
      "step": 1130
    },
    {
      "epoch": 0.39353380844776636,
      "grad_norm": 0.7845961219955971,
      "learning_rate": 0.00019435965982166634,
      "loss": 0.5682,
      "step": 1132
    },
    {
      "epoch": 0.39422909786198507,
      "grad_norm": 0.5467699155830194,
      "learning_rate": 0.0001943193981516442,
      "loss": 0.5801,
      "step": 1134
    },
    {
      "epoch": 0.3949243872762037,
      "grad_norm": 0.8265539141839208,
      "learning_rate": 0.00019427899749526592,
      "loss": 0.6112,
      "step": 1136
    },
    {
      "epoch": 0.39561967669042236,
      "grad_norm": 0.6952363088057728,
      "learning_rate": 0.0001942384579120648,
      "loss": 0.5585,
      "step": 1138
    },
    {
      "epoch": 0.39631496610464106,
      "grad_norm": 0.5390583855264917,
      "learning_rate": 0.00019419777946177872,
      "loss": 0.5437,
      "step": 1140
    },
    {
      "epoch": 0.3970102555188597,
      "grad_norm": 0.7703174634754777,
      "learning_rate": 0.0001941569622043504,
      "loss": 0.6353,
      "step": 1142
    },
    {
      "epoch": 0.3977055449330784,
      "grad_norm": 0.6599770887173079,
      "learning_rate": 0.0001941160061999268,
      "loss": 0.6984,
      "step": 1144
    },
    {
      "epoch": 0.39840083434729706,
      "grad_norm": 0.6388540477012332,
      "learning_rate": 0.0001940749115088597,
      "loss": 0.6716,
      "step": 1146
    },
    {
      "epoch": 0.3990961237615157,
      "grad_norm": 0.556740144667956,
      "learning_rate": 0.00019403367819170495,
      "loss": 0.5403,
      "step": 1148
    },
    {
      "epoch": 0.3997914131757344,
      "grad_norm": 0.5811188003032377,
      "learning_rate": 0.00019399230630922281,
      "loss": 0.6038,
      "step": 1150
    },
    {
      "epoch": 0.40048670258995306,
      "grad_norm": 0.5247246331826143,
      "learning_rate": 0.00019395079592237767,
      "loss": 0.5802,
      "step": 1152
    },
    {
      "epoch": 0.40118199200417176,
      "grad_norm": 0.6218229787877086,
      "learning_rate": 0.00019390914709233812,
      "loss": 0.613,
      "step": 1154
    },
    {
      "epoch": 0.4018772814183904,
      "grad_norm": 0.3768886648264559,
      "learning_rate": 0.00019386735988047657,
      "loss": 0.546,
      "step": 1156
    },
    {
      "epoch": 0.40257257083260906,
      "grad_norm": 0.7192604288297088,
      "learning_rate": 0.00019382543434836956,
      "loss": 0.6414,
      "step": 1158
    },
    {
      "epoch": 0.40326786024682776,
      "grad_norm": 0.6502767283266627,
      "learning_rate": 0.00019378337055779725,
      "loss": 0.6913,
      "step": 1160
    },
    {
      "epoch": 0.4039631496610464,
      "grad_norm": 0.5821846511835889,
      "learning_rate": 0.00019374116857074372,
      "loss": 0.6135,
      "step": 1162
    },
    {
      "epoch": 0.40465843907526505,
      "grad_norm": 0.4214614762979627,
      "learning_rate": 0.00019369882844939656,
      "loss": 0.5844,
      "step": 1164
    },
    {
      "epoch": 0.40535372848948376,
      "grad_norm": 0.3995087581540831,
      "learning_rate": 0.00019365635025614698,
      "loss": 0.597,
      "step": 1166
    },
    {
      "epoch": 0.4060490179037024,
      "grad_norm": 0.6448855951735888,
      "learning_rate": 0.0001936137340535896,
      "loss": 0.5964,
      "step": 1168
    },
    {
      "epoch": 0.4067443073179211,
      "grad_norm": 0.4901354392109259,
      "learning_rate": 0.00019357097990452244,
      "loss": 0.6544,
      "step": 1170
    },
    {
      "epoch": 0.40743959673213975,
      "grad_norm": 1.453302582941976,
      "learning_rate": 0.0001935280878719468,
      "loss": 0.7041,
      "step": 1172
    },
    {
      "epoch": 0.4081348861463584,
      "grad_norm": 0.5029703758127726,
      "learning_rate": 0.00019348505801906717,
      "loss": 0.5963,
      "step": 1174
    },
    {
      "epoch": 0.4088301755605771,
      "grad_norm": 0.42609969312576673,
      "learning_rate": 0.00019344189040929104,
      "loss": 0.5741,
      "step": 1176
    },
    {
      "epoch": 0.40952546497479575,
      "grad_norm": 0.7354212284029352,
      "learning_rate": 0.000193398585106229,
      "loss": 0.5587,
      "step": 1178
    },
    {
      "epoch": 0.41022075438901445,
      "grad_norm": 0.4221453871118396,
      "learning_rate": 0.00019335514217369448,
      "loss": 0.6008,
      "step": 1180
    },
    {
      "epoch": 0.4109160438032331,
      "grad_norm": 0.4305199213179302,
      "learning_rate": 0.00019331156167570377,
      "loss": 0.6589,
      "step": 1182
    },
    {
      "epoch": 0.41161133321745175,
      "grad_norm": 0.5908546860050797,
      "learning_rate": 0.0001932678436764758,
      "loss": 0.6791,
      "step": 1184
    },
    {
      "epoch": 0.41230662263167045,
      "grad_norm": 0.592125562883329,
      "learning_rate": 0.0001932239882404322,
      "loss": 0.6163,
      "step": 1186
    },
    {
      "epoch": 0.4130019120458891,
      "grad_norm": 0.3760868982177292,
      "learning_rate": 0.00019317999543219707,
      "loss": 0.6057,
      "step": 1188
    },
    {
      "epoch": 0.41369720146010774,
      "grad_norm": 0.42271908071862413,
      "learning_rate": 0.00019313586531659693,
      "loss": 0.6055,
      "step": 1190
    },
    {
      "epoch": 0.41439249087432645,
      "grad_norm": 0.7637973810401912,
      "learning_rate": 0.00019309159795866067,
      "loss": 0.6113,
      "step": 1192
    },
    {
      "epoch": 0.4150877802885451,
      "grad_norm": 0.31348977991538335,
      "learning_rate": 0.00019304719342361942,
      "loss": 0.5906,
      "step": 1194
    },
    {
      "epoch": 0.4157830697027638,
      "grad_norm": 0.34482048624358563,
      "learning_rate": 0.00019300265177690635,
      "loss": 0.5722,
      "step": 1196
    },
    {
      "epoch": 0.41647835911698244,
      "grad_norm": 0.8513089082812038,
      "learning_rate": 0.0001929579730841568,
      "loss": 0.6293,
      "step": 1198
    },
    {
      "epoch": 0.4171736485312011,
      "grad_norm": 0.4489163724605792,
      "learning_rate": 0.00019291315741120802,
      "loss": 0.5779,
      "step": 1200
    },
    {
      "epoch": 0.4171736485312011,
      "eval_loss": 0.6036229133605957,
      "eval_runtime": 710.4375,
      "eval_samples_per_second": 6.816,
      "eval_steps_per_second": 0.214,
      "step": 1200
    },
    {
      "epoch": 0.4178689379454198,
      "grad_norm": 0.4187714520019529,
      "learning_rate": 0.00019286820482409907,
      "loss": 0.6429,
      "step": 1202
    },
    {
      "epoch": 0.41856422735963844,
      "grad_norm": 0.6777550653965533,
      "learning_rate": 0.0001928231153890708,
      "loss": 0.7349,
      "step": 1204
    },
    {
      "epoch": 0.41925951677385714,
      "grad_norm": 0.5320560870085244,
      "learning_rate": 0.00019277788917256575,
      "loss": 0.6171,
      "step": 1206
    },
    {
      "epoch": 0.4199548061880758,
      "grad_norm": 0.4598764394317783,
      "learning_rate": 0.0001927325262412279,
      "loss": 0.568,
      "step": 1208
    },
    {
      "epoch": 0.42065009560229444,
      "grad_norm": 0.6091029764488312,
      "learning_rate": 0.0001926870266619028,
      "loss": 0.6179,
      "step": 1210
    },
    {
      "epoch": 0.42134538501651314,
      "grad_norm": 0.3831881756520618,
      "learning_rate": 0.00019264139050163733,
      "loss": 0.6277,
      "step": 1212
    },
    {
      "epoch": 0.4220406744307318,
      "grad_norm": 0.46975774391187974,
      "learning_rate": 0.00019259561782767964,
      "loss": 0.6365,
      "step": 1214
    },
    {
      "epoch": 0.42273596384495044,
      "grad_norm": 0.43958415828248065,
      "learning_rate": 0.00019254970870747896,
      "loss": 0.5599,
      "step": 1216
    },
    {
      "epoch": 0.42343125325916914,
      "grad_norm": 0.3718315331231464,
      "learning_rate": 0.00019250366320868573,
      "loss": 0.5466,
      "step": 1218
    },
    {
      "epoch": 0.4241265426733878,
      "grad_norm": 0.4132277343365207,
      "learning_rate": 0.00019245748139915122,
      "loss": 0.5505,
      "step": 1220
    },
    {
      "epoch": 0.4248218320876065,
      "grad_norm": 0.699259138789713,
      "learning_rate": 0.00019241116334692767,
      "loss": 0.6155,
      "step": 1222
    },
    {
      "epoch": 0.42551712150182514,
      "grad_norm": 0.8108211625089844,
      "learning_rate": 0.00019236470912026795,
      "loss": 0.6389,
      "step": 1224
    },
    {
      "epoch": 0.4262124109160438,
      "grad_norm": 0.7991219996691611,
      "learning_rate": 0.00019231811878762578,
      "loss": 0.6218,
      "step": 1226
    },
    {
      "epoch": 0.4269077003302625,
      "grad_norm": 0.6512869246829502,
      "learning_rate": 0.00019227139241765527,
      "loss": 0.6263,
      "step": 1228
    },
    {
      "epoch": 0.42760298974448113,
      "grad_norm": 0.6616729861929296,
      "learning_rate": 0.00019222453007921103,
      "loss": 0.633,
      "step": 1230
    },
    {
      "epoch": 0.42829827915869984,
      "grad_norm": 0.6441717114274472,
      "learning_rate": 0.0001921775318413481,
      "loss": 0.598,
      "step": 1232
    },
    {
      "epoch": 0.4289935685729185,
      "grad_norm": 0.5398693786298329,
      "learning_rate": 0.00019213039777332173,
      "loss": 0.6128,
      "step": 1234
    },
    {
      "epoch": 0.42968885798713713,
      "grad_norm": 0.7990572416423266,
      "learning_rate": 0.00019208312794458734,
      "loss": 0.6237,
      "step": 1236
    },
    {
      "epoch": 0.43038414740135583,
      "grad_norm": 0.6792096783122197,
      "learning_rate": 0.00019203572242480033,
      "loss": 0.6167,
      "step": 1238
    },
    {
      "epoch": 0.4310794368155745,
      "grad_norm": 0.569412739061848,
      "learning_rate": 0.00019198818128381622,
      "loss": 0.7147,
      "step": 1240
    },
    {
      "epoch": 0.4317747262297931,
      "grad_norm": 0.3343242957299063,
      "learning_rate": 0.00019194050459169016,
      "loss": 0.5585,
      "step": 1242
    },
    {
      "epoch": 0.43247001564401183,
      "grad_norm": 0.41389163470101814,
      "learning_rate": 0.00019189269241867726,
      "loss": 0.6206,
      "step": 1244
    },
    {
      "epoch": 0.4331653050582305,
      "grad_norm": 0.6725681446750024,
      "learning_rate": 0.00019184474483523208,
      "loss": 0.6592,
      "step": 1246
    },
    {
      "epoch": 0.4338605944724492,
      "grad_norm": 1.063023358388528,
      "learning_rate": 0.0001917966619120088,
      "loss": 0.6998,
      "step": 1248
    },
    {
      "epoch": 0.4345558838866678,
      "grad_norm": 1.0481626997531774,
      "learning_rate": 0.00019174844371986111,
      "loss": 0.5649,
      "step": 1250
    },
    {
      "epoch": 0.4352511733008865,
      "grad_norm": 0.6436744584316785,
      "learning_rate": 0.00019170009032984188,
      "loss": 0.6011,
      "step": 1252
    },
    {
      "epoch": 0.4359464627151052,
      "grad_norm": 0.6100699160131396,
      "learning_rate": 0.00019165160181320331,
      "loss": 0.6229,
      "step": 1254
    },
    {
      "epoch": 0.4366417521293238,
      "grad_norm": 0.43075456887586555,
      "learning_rate": 0.00019160297824139671,
      "loss": 0.5985,
      "step": 1256
    },
    {
      "epoch": 0.4373370415435425,
      "grad_norm": 1.1133754307509498,
      "learning_rate": 0.0001915542196860723,
      "loss": 0.582,
      "step": 1258
    },
    {
      "epoch": 0.4380323309577612,
      "grad_norm": 0.8115545368570724,
      "learning_rate": 0.00019150532621907935,
      "loss": 0.6281,
      "step": 1260
    },
    {
      "epoch": 0.4387276203719798,
      "grad_norm": 0.6501357510133978,
      "learning_rate": 0.00019145629791246586,
      "loss": 0.7366,
      "step": 1262
    },
    {
      "epoch": 0.4394229097861985,
      "grad_norm": 1.0301487119179336,
      "learning_rate": 0.00019140713483847854,
      "loss": 0.5878,
      "step": 1264
    },
    {
      "epoch": 0.44011819920041717,
      "grad_norm": 0.9720409793058,
      "learning_rate": 0.00019135783706956266,
      "loss": 0.5096,
      "step": 1266
    },
    {
      "epoch": 0.4408134886146358,
      "grad_norm": 0.5612642914060567,
      "learning_rate": 0.000191308404678362,
      "loss": 0.6624,
      "step": 1268
    },
    {
      "epoch": 0.4415087780288545,
      "grad_norm": 0.46162422465509245,
      "learning_rate": 0.00019125883773771874,
      "loss": 0.5714,
      "step": 1270
    },
    {
      "epoch": 0.44220406744307317,
      "grad_norm": 0.7669640725186674,
      "learning_rate": 0.00019120913632067325,
      "loss": 0.6786,
      "step": 1272
    },
    {
      "epoch": 0.44289935685729187,
      "grad_norm": 0.4747976521609601,
      "learning_rate": 0.00019115930050046416,
      "loss": 0.5964,
      "step": 1274
    },
    {
      "epoch": 0.4435946462715105,
      "grad_norm": 0.8528446125546961,
      "learning_rate": 0.0001911093303505281,
      "loss": 0.6584,
      "step": 1276
    },
    {
      "epoch": 0.44428993568572916,
      "grad_norm": 0.444423482826933,
      "learning_rate": 0.00019105922594449962,
      "loss": 0.5931,
      "step": 1278
    },
    {
      "epoch": 0.44498522509994787,
      "grad_norm": 0.4296567435369722,
      "learning_rate": 0.00019100898735621114,
      "loss": 0.5614,
      "step": 1280
    },
    {
      "epoch": 0.4456805145141665,
      "grad_norm": 0.3735319415225358,
      "learning_rate": 0.0001909586146596928,
      "loss": 0.5585,
      "step": 1282
    },
    {
      "epoch": 0.4463758039283852,
      "grad_norm": 0.3997395665103462,
      "learning_rate": 0.0001909081079291724,
      "loss": 0.4911,
      "step": 1284
    },
    {
      "epoch": 0.44707109334260386,
      "grad_norm": 0.5067764089739433,
      "learning_rate": 0.00019085746723907513,
      "loss": 0.5912,
      "step": 1286
    },
    {
      "epoch": 0.4477663827568225,
      "grad_norm": 1.5023247096958177,
      "learning_rate": 0.00019080669266402373,
      "loss": 0.7538,
      "step": 1288
    },
    {
      "epoch": 0.4484616721710412,
      "grad_norm": 1.0536335728512713,
      "learning_rate": 0.0001907557842788381,
      "loss": 0.7347,
      "step": 1290
    },
    {
      "epoch": 0.44915696158525986,
      "grad_norm": 1.3178937768983239,
      "learning_rate": 0.00019070474215853543,
      "loss": 0.6302,
      "step": 1292
    },
    {
      "epoch": 0.4498522509994785,
      "grad_norm": 0.5893948676721078,
      "learning_rate": 0.00019065356637832986,
      "loss": 0.626,
      "step": 1294
    },
    {
      "epoch": 0.4505475404136972,
      "grad_norm": 1.118265378089079,
      "learning_rate": 0.0001906022570136326,
      "loss": 0.5155,
      "step": 1296
    },
    {
      "epoch": 0.45124282982791586,
      "grad_norm": 0.3585975576830803,
      "learning_rate": 0.00019055081414005165,
      "loss": 0.5558,
      "step": 1298
    },
    {
      "epoch": 0.45193811924213456,
      "grad_norm": 1.3203699153765986,
      "learning_rate": 0.00019049923783339171,
      "loss": 0.6516,
      "step": 1300
    },
    {
      "epoch": 0.4526334086563532,
      "grad_norm": 0.8836139230641887,
      "learning_rate": 0.0001904475281696542,
      "loss": 0.6532,
      "step": 1302
    },
    {
      "epoch": 0.45332869807057186,
      "grad_norm": 0.7111007749248109,
      "learning_rate": 0.00019039568522503694,
      "loss": 0.6273,
      "step": 1304
    },
    {
      "epoch": 0.45402398748479056,
      "grad_norm": 0.4303152450432435,
      "learning_rate": 0.00019034370907593427,
      "loss": 0.5596,
      "step": 1306
    },
    {
      "epoch": 0.4547192768990092,
      "grad_norm": 0.46058879749054005,
      "learning_rate": 0.00019029159979893669,
      "loss": 0.6002,
      "step": 1308
    },
    {
      "epoch": 0.4554145663132279,
      "grad_norm": 1.1757657004565762,
      "learning_rate": 0.00019023935747083094,
      "loss": 0.5425,
      "step": 1310
    },
    {
      "epoch": 0.45610985572744656,
      "grad_norm": 0.6763815344574016,
      "learning_rate": 0.00019018698216859985,
      "loss": 0.7178,
      "step": 1312
    },
    {
      "epoch": 0.4568051451416652,
      "grad_norm": 0.6557528693425259,
      "learning_rate": 0.00019013447396942215,
      "loss": 0.5482,
      "step": 1314
    },
    {
      "epoch": 0.4575004345558839,
      "grad_norm": 0.40206624331116686,
      "learning_rate": 0.0001900818329506724,
      "loss": 0.5626,
      "step": 1316
    },
    {
      "epoch": 0.45819572397010255,
      "grad_norm": 0.39959833266619904,
      "learning_rate": 0.0001900290591899209,
      "loss": 0.5805,
      "step": 1318
    },
    {
      "epoch": 0.4588910133843212,
      "grad_norm": 1.157537080813341,
      "learning_rate": 0.00018997615276493353,
      "loss": 0.5895,
      "step": 1320
    },
    {
      "epoch": 0.4595863027985399,
      "grad_norm": 1.9818188598958286,
      "learning_rate": 0.0001899231137536717,
      "loss": 0.6688,
      "step": 1322
    },
    {
      "epoch": 0.46028159221275855,
      "grad_norm": 0.6448001142497345,
      "learning_rate": 0.00018986994223429217,
      "loss": 0.5878,
      "step": 1324
    },
    {
      "epoch": 0.46097688162697725,
      "grad_norm": 0.7035140294155504,
      "learning_rate": 0.000189816638285147,
      "loss": 0.59,
      "step": 1326
    },
    {
      "epoch": 0.4616721710411959,
      "grad_norm": 0.5547705850102905,
      "learning_rate": 0.00018976320198478327,
      "loss": 0.6194,
      "step": 1328
    },
    {
      "epoch": 0.46236746045541455,
      "grad_norm": 0.8218894600210293,
      "learning_rate": 0.00018970963341194327,
      "loss": 0.5957,
      "step": 1330
    },
    {
      "epoch": 0.46306274986963325,
      "grad_norm": 0.8582770909742338,
      "learning_rate": 0.00018965593264556405,
      "loss": 0.6027,
      "step": 1332
    },
    {
      "epoch": 0.4637580392838519,
      "grad_norm": 0.5859159431078994,
      "learning_rate": 0.00018960209976477755,
      "loss": 0.5701,
      "step": 1334
    },
    {
      "epoch": 0.46445332869807054,
      "grad_norm": 0.6055238281909472,
      "learning_rate": 0.00018954813484891033,
      "loss": 0.6069,
      "step": 1336
    },
    {
      "epoch": 0.46514861811228925,
      "grad_norm": 0.7336223128037048,
      "learning_rate": 0.00018949403797748356,
      "loss": 0.639,
      "step": 1338
    },
    {
      "epoch": 0.4658439075265079,
      "grad_norm": 0.4617499217742889,
      "learning_rate": 0.0001894398092302128,
      "loss": 0.5972,
      "step": 1340
    },
    {
      "epoch": 0.4665391969407266,
      "grad_norm": 0.6626369850916174,
      "learning_rate": 0.00018938544868700804,
      "loss": 0.6254,
      "step": 1342
    },
    {
      "epoch": 0.46723448635494524,
      "grad_norm": 0.6487571201649612,
      "learning_rate": 0.00018933095642797336,
      "loss": 0.6115,
      "step": 1344
    },
    {
      "epoch": 0.4679297757691639,
      "grad_norm": 0.721231100905907,
      "learning_rate": 0.00018927633253340703,
      "loss": 0.5056,
      "step": 1346
    },
    {
      "epoch": 0.4686250651833826,
      "grad_norm": 0.43640759465379947,
      "learning_rate": 0.0001892215770838012,
      "loss": 0.618,
      "step": 1348
    },
    {
      "epoch": 0.46932035459760124,
      "grad_norm": 0.7472697329585503,
      "learning_rate": 0.00018916669015984198,
      "loss": 0.5145,
      "step": 1350
    },
    {
      "epoch": 0.46932035459760124,
      "eval_loss": 0.6091039180755615,
      "eval_runtime": 712.5909,
      "eval_samples_per_second": 6.795,
      "eval_steps_per_second": 0.213,
      "step": 1350
    },
    {
      "epoch": 0.47001564401181994,
      "grad_norm": 0.8587847612083451,
      "learning_rate": 0.00018911167184240915,
      "loss": 0.6002,
      "step": 1352
    },
    {
      "epoch": 0.4707109334260386,
      "grad_norm": 0.570124970949044,
      "learning_rate": 0.0001890565222125761,
      "loss": 0.6562,
      "step": 1354
    },
    {
      "epoch": 0.47140622284025724,
      "grad_norm": 0.6984427176055181,
      "learning_rate": 0.00018900124135160976,
      "loss": 0.7262,
      "step": 1356
    },
    {
      "epoch": 0.47210151225447594,
      "grad_norm": 0.765890528385208,
      "learning_rate": 0.0001889458293409704,
      "loss": 0.5891,
      "step": 1358
    },
    {
      "epoch": 0.4727968016686946,
      "grad_norm": 0.609256703935707,
      "learning_rate": 0.0001888902862623116,
      "loss": 0.5919,
      "step": 1360
    },
    {
      "epoch": 0.47349209108291324,
      "grad_norm": 0.4622539193764851,
      "learning_rate": 0.00018883461219748,
      "loss": 0.6179,
      "step": 1362
    },
    {
      "epoch": 0.47418738049713194,
      "grad_norm": 0.46144204894965307,
      "learning_rate": 0.00018877880722851536,
      "loss": 0.6289,
      "step": 1364
    },
    {
      "epoch": 0.4748826699113506,
      "grad_norm": 0.8200013455891142,
      "learning_rate": 0.00018872287143765023,
      "loss": 0.5644,
      "step": 1366
    },
    {
      "epoch": 0.4755779593255693,
      "grad_norm": 0.5631892935159202,
      "learning_rate": 0.00018866680490730998,
      "loss": 0.5034,
      "step": 1368
    },
    {
      "epoch": 0.47627324873978794,
      "grad_norm": 0.5739777810808598,
      "learning_rate": 0.00018861060772011273,
      "loss": 0.651,
      "step": 1370
    },
    {
      "epoch": 0.4769685381540066,
      "grad_norm": 0.666694270081405,
      "learning_rate": 0.00018855427995886892,
      "loss": 0.58,
      "step": 1372
    },
    {
      "epoch": 0.4776638275682253,
      "grad_norm": 0.4844198549303021,
      "learning_rate": 0.00018849782170658158,
      "loss": 0.5842,
      "step": 1374
    },
    {
      "epoch": 0.47835911698244393,
      "grad_norm": 0.5289779719515055,
      "learning_rate": 0.00018844123304644596,
      "loss": 0.597,
      "step": 1376
    },
    {
      "epoch": 0.47905440639666264,
      "grad_norm": 1.0867526542265191,
      "learning_rate": 0.0001883845140618495,
      "loss": 0.6741,
      "step": 1378
    },
    {
      "epoch": 0.4797496958108813,
      "grad_norm": 0.6205435292114323,
      "learning_rate": 0.00018832766483637165,
      "loss": 0.6176,
      "step": 1380
    },
    {
      "epoch": 0.48044498522509993,
      "grad_norm": 0.49806740313966835,
      "learning_rate": 0.0001882706854537838,
      "loss": 0.6003,
      "step": 1382
    },
    {
      "epoch": 0.48114027463931863,
      "grad_norm": 0.713610401013068,
      "learning_rate": 0.00018821357599804915,
      "loss": 0.5746,
      "step": 1384
    },
    {
      "epoch": 0.4818355640535373,
      "grad_norm": 0.7286362166201054,
      "learning_rate": 0.00018815633655332252,
      "loss": 0.5824,
      "step": 1386
    },
    {
      "epoch": 0.4825308534677559,
      "grad_norm": 0.7609524076074863,
      "learning_rate": 0.00018809896720395033,
      "loss": 0.5422,
      "step": 1388
    },
    {
      "epoch": 0.48322614288197463,
      "grad_norm": 0.8489750160154781,
      "learning_rate": 0.0001880414680344704,
      "loss": 0.5797,
      "step": 1390
    },
    {
      "epoch": 0.4839214322961933,
      "grad_norm": 1.3137718080784093,
      "learning_rate": 0.00018798383912961187,
      "loss": 0.5397,
      "step": 1392
    },
    {
      "epoch": 0.484616721710412,
      "grad_norm": 0.7192606649476782,
      "learning_rate": 0.00018792608057429503,
      "loss": 0.6948,
      "step": 1394
    },
    {
      "epoch": 0.4853120111246306,
      "grad_norm": 0.8441204863561427,
      "learning_rate": 0.00018786819245363118,
      "loss": 0.5786,
      "step": 1396
    },
    {
      "epoch": 0.4860073005388493,
      "grad_norm": 0.6806475957554041,
      "learning_rate": 0.00018781017485292267,
      "loss": 0.6033,
      "step": 1398
    },
    {
      "epoch": 0.486702589953068,
      "grad_norm": 0.5168648519720452,
      "learning_rate": 0.00018775202785766256,
      "loss": 0.7098,
      "step": 1400
    },
    {
      "epoch": 0.4873978793672866,
      "grad_norm": 0.465013130005678,
      "learning_rate": 0.00018769375155353453,
      "loss": 0.7186,
      "step": 1402
    },
    {
      "epoch": 0.4880931687815053,
      "grad_norm": 0.453274392570444,
      "learning_rate": 0.00018763534602641291,
      "loss": 0.5782,
      "step": 1404
    },
    {
      "epoch": 0.488788458195724,
      "grad_norm": 0.5972503276564156,
      "learning_rate": 0.00018757681136236242,
      "loss": 0.6221,
      "step": 1406
    },
    {
      "epoch": 0.4894837476099426,
      "grad_norm": 0.36846650508391193,
      "learning_rate": 0.00018751814764763806,
      "loss": 0.5966,
      "step": 1408
    },
    {
      "epoch": 0.4901790370241613,
      "grad_norm": 0.3451947155848497,
      "learning_rate": 0.00018745935496868493,
      "loss": 0.5508,
      "step": 1410
    },
    {
      "epoch": 0.49087432643837997,
      "grad_norm": 0.5078133383734431,
      "learning_rate": 0.00018740043341213832,
      "loss": 0.5029,
      "step": 1412
    },
    {
      "epoch": 0.4915696158525986,
      "grad_norm": 0.767602706979167,
      "learning_rate": 0.00018734138306482332,
      "loss": 0.6083,
      "step": 1414
    },
    {
      "epoch": 0.4922649052668173,
      "grad_norm": 0.5648335118893248,
      "learning_rate": 0.00018728220401375477,
      "loss": 0.618,
      "step": 1416
    },
    {
      "epoch": 0.49296019468103597,
      "grad_norm": 0.7659928375858898,
      "learning_rate": 0.00018722289634613728,
      "loss": 0.5907,
      "step": 1418
    },
    {
      "epoch": 0.49365548409525467,
      "grad_norm": 0.9164796669061794,
      "learning_rate": 0.0001871634601493649,
      "loss": 0.4978,
      "step": 1420
    },
    {
      "epoch": 0.4943507735094733,
      "grad_norm": 0.49130889452749815,
      "learning_rate": 0.00018710389551102115,
      "loss": 0.5277,
      "step": 1422
    },
    {
      "epoch": 0.49504606292369197,
      "grad_norm": 0.5824962532822966,
      "learning_rate": 0.00018704420251887868,
      "loss": 0.5717,
      "step": 1424
    },
    {
      "epoch": 0.49574135233791067,
      "grad_norm": 0.5964322775094694,
      "learning_rate": 0.00018698438126089944,
      "loss": 0.6249,
      "step": 1426
    },
    {
      "epoch": 0.4964366417521293,
      "grad_norm": 0.42806592395765636,
      "learning_rate": 0.0001869244318252343,
      "loss": 0.5433,
      "step": 1428
    },
    {
      "epoch": 0.497131931166348,
      "grad_norm": 0.8340709008133664,
      "learning_rate": 0.000186864354300223,
      "loss": 0.5714,
      "step": 1430
    },
    {
      "epoch": 0.49782722058056667,
      "grad_norm": 0.6732191602218343,
      "learning_rate": 0.00018680414877439406,
      "loss": 0.6572,
      "step": 1432
    },
    {
      "epoch": 0.4985225099947853,
      "grad_norm": 0.5785424946243081,
      "learning_rate": 0.00018674381533646467,
      "loss": 0.5831,
      "step": 1434
    },
    {
      "epoch": 0.499217799409004,
      "grad_norm": 0.6435404053128269,
      "learning_rate": 0.0001866833540753404,
      "loss": 0.582,
      "step": 1436
    },
    {
      "epoch": 0.49991308882322266,
      "grad_norm": 0.5713158720544551,
      "learning_rate": 0.00018662276508011526,
      "loss": 0.5276,
      "step": 1438
    },
    {
      "epoch": 0.5006083782374413,
      "grad_norm": 0.43458150402178386,
      "learning_rate": 0.00018656204844007143,
      "loss": 0.5526,
      "step": 1440
    },
    {
      "epoch": 0.50130366765166,
      "grad_norm": 0.5522953777612155,
      "learning_rate": 0.00018650120424467922,
      "loss": 0.6259,
      "step": 1442
    },
    {
      "epoch": 0.5019989570658787,
      "grad_norm": 0.6700198262354904,
      "learning_rate": 0.00018644023258359684,
      "loss": 0.5976,
      "step": 1444
    },
    {
      "epoch": 0.5026942464800973,
      "grad_norm": 0.653642473920571,
      "learning_rate": 0.00018637913354667044,
      "loss": 0.6187,
      "step": 1446
    },
    {
      "epoch": 0.503389535894316,
      "grad_norm": 0.8572149566943323,
      "learning_rate": 0.00018631790722393378,
      "loss": 0.6685,
      "step": 1448
    },
    {
      "epoch": 0.5040848253085347,
      "grad_norm": 0.4144221514106495,
      "learning_rate": 0.00018625655370560823,
      "loss": 0.514,
      "step": 1450
    },
    {
      "epoch": 0.5047801147227533,
      "grad_norm": 0.8737004492907994,
      "learning_rate": 0.00018619507308210255,
      "loss": 0.6583,
      "step": 1452
    },
    {
      "epoch": 0.505475404136972,
      "grad_norm": 0.3987400457590611,
      "learning_rate": 0.00018613346544401281,
      "loss": 0.5663,
      "step": 1454
    },
    {
      "epoch": 0.5061706935511907,
      "grad_norm": 0.591401748694551,
      "learning_rate": 0.0001860717308821223,
      "loss": 0.6537,
      "step": 1456
    },
    {
      "epoch": 0.5068659829654093,
      "grad_norm": 0.555392995714629,
      "learning_rate": 0.00018600986948740124,
      "loss": 0.5692,
      "step": 1458
    },
    {
      "epoch": 0.507561272379628,
      "grad_norm": 0.39435725187187276,
      "learning_rate": 0.00018594788135100688,
      "loss": 0.6265,
      "step": 1460
    },
    {
      "epoch": 0.5082565617938467,
      "grad_norm": 0.4077334073620459,
      "learning_rate": 0.00018588576656428308,
      "loss": 0.5554,
      "step": 1462
    },
    {
      "epoch": 0.5089518512080654,
      "grad_norm": 0.38687853565198027,
      "learning_rate": 0.00018582352521876046,
      "loss": 0.6378,
      "step": 1464
    },
    {
      "epoch": 0.509647140622284,
      "grad_norm": 0.31280184968676994,
      "learning_rate": 0.00018576115740615606,
      "loss": 0.5416,
      "step": 1466
    },
    {
      "epoch": 0.5103424300365027,
      "grad_norm": 1.3853246114916373,
      "learning_rate": 0.00018569866321837327,
      "loss": 0.6763,
      "step": 1468
    },
    {
      "epoch": 0.5110377194507214,
      "grad_norm": 0.5024331777389988,
      "learning_rate": 0.00018563604274750178,
      "loss": 0.6372,
      "step": 1470
    },
    {
      "epoch": 0.51173300886494,
      "grad_norm": 0.32922333596717174,
      "learning_rate": 0.0001855732960858173,
      "loss": 0.5923,
      "step": 1472
    },
    {
      "epoch": 0.5124282982791587,
      "grad_norm": 0.9059198527449924,
      "learning_rate": 0.00018551042332578144,
      "loss": 0.659,
      "step": 1474
    },
    {
      "epoch": 0.5131235876933774,
      "grad_norm": 0.7045580743620987,
      "learning_rate": 0.00018544742456004181,
      "loss": 0.6416,
      "step": 1476
    },
    {
      "epoch": 0.513818877107596,
      "grad_norm": 1.6530740850637509,
      "learning_rate": 0.0001853842998814315,
      "loss": 0.5863,
      "step": 1478
    },
    {
      "epoch": 0.5145141665218147,
      "grad_norm": 0.742405889420684,
      "learning_rate": 0.00018532104938296927,
      "loss": 0.5942,
      "step": 1480
    },
    {
      "epoch": 0.5152094559360334,
      "grad_norm": 0.43475036345006896,
      "learning_rate": 0.0001852576731578592,
      "loss": 0.54,
      "step": 1482
    },
    {
      "epoch": 0.515904745350252,
      "grad_norm": 1.0939432717807218,
      "learning_rate": 0.00018519417129949072,
      "loss": 0.6034,
      "step": 1484
    },
    {
      "epoch": 0.5166000347644707,
      "grad_norm": 1.1841858811933605,
      "learning_rate": 0.0001851305439014383,
      "loss": 0.6119,
      "step": 1486
    },
    {
      "epoch": 0.5172953241786894,
      "grad_norm": 0.8950773429853176,
      "learning_rate": 0.00018506679105746149,
      "loss": 0.6224,
      "step": 1488
    },
    {
      "epoch": 0.5179906135929081,
      "grad_norm": 0.9558807568134642,
      "learning_rate": 0.0001850029128615046,
      "loss": 0.5986,
      "step": 1490
    },
    {
      "epoch": 0.5186859030071267,
      "grad_norm": 0.45382840521759343,
      "learning_rate": 0.00018493890940769671,
      "loss": 0.61,
      "step": 1492
    },
    {
      "epoch": 0.5193811924213454,
      "grad_norm": 0.4697574815571038,
      "learning_rate": 0.00018487478079035153,
      "loss": 0.5299,
      "step": 1494
    },
    {
      "epoch": 0.5200764818355641,
      "grad_norm": 0.699871643062895,
      "learning_rate": 0.00018481052710396707,
      "loss": 0.6021,
      "step": 1496
    },
    {
      "epoch": 0.5207717712497827,
      "grad_norm": 0.9389496797384421,
      "learning_rate": 0.0001847461484432258,
      "loss": 0.6119,
      "step": 1498
    },
    {
      "epoch": 0.5214670606640014,
      "grad_norm": 1.1160995208439726,
      "learning_rate": 0.00018468164490299417,
      "loss": 0.59,
      "step": 1500
    },
    {
      "epoch": 0.5214670606640014,
      "eval_loss": 0.6035115718841553,
      "eval_runtime": 374.7159,
      "eval_samples_per_second": 12.922,
      "eval_steps_per_second": 0.406,
      "step": 1500
    },
    {
      "epoch": 0.5221623500782201,
      "grad_norm": 0.8572137642677378,
      "learning_rate": 0.00018461701657832285,
      "loss": 0.5865,
      "step": 1502
    },
    {
      "epoch": 0.5228576394924387,
      "grad_norm": 0.9221217554185654,
      "learning_rate": 0.0001845522635644462,
      "loss": 0.5479,
      "step": 1504
    },
    {
      "epoch": 0.5235529289066574,
      "grad_norm": 2.0492350557846577,
      "learning_rate": 0.0001844873859567824,
      "loss": 0.7292,
      "step": 1506
    },
    {
      "epoch": 0.5242482183208761,
      "grad_norm": 1.1824788332559788,
      "learning_rate": 0.0001844223838509333,
      "loss": 0.6479,
      "step": 1508
    },
    {
      "epoch": 0.5249435077350947,
      "grad_norm": 0.7039837396797187,
      "learning_rate": 0.00018435725734268407,
      "loss": 0.6149,
      "step": 1510
    },
    {
      "epoch": 0.5256387971493134,
      "grad_norm": 0.4401391384170765,
      "learning_rate": 0.0001842920065280033,
      "loss": 0.5854,
      "step": 1512
    },
    {
      "epoch": 0.5263340865635321,
      "grad_norm": 0.4100923191497827,
      "learning_rate": 0.0001842266315030427,
      "loss": 0.6091,
      "step": 1514
    },
    {
      "epoch": 0.5270293759777508,
      "grad_norm": 0.5510661253393345,
      "learning_rate": 0.00018416113236413705,
      "loss": 0.5721,
      "step": 1516
    },
    {
      "epoch": 0.5277246653919694,
      "grad_norm": 0.4386368092109241,
      "learning_rate": 0.00018409550920780397,
      "loss": 0.6232,
      "step": 1518
    },
    {
      "epoch": 0.5284199548061881,
      "grad_norm": 0.5520600896880458,
      "learning_rate": 0.00018402976213074385,
      "loss": 0.6134,
      "step": 1520
    },
    {
      "epoch": 0.5291152442204068,
      "grad_norm": 0.5101647982715836,
      "learning_rate": 0.00018396389122983975,
      "loss": 0.5918,
      "step": 1522
    },
    {
      "epoch": 0.5298105336346254,
      "grad_norm": 0.7801958609328837,
      "learning_rate": 0.00018389789660215708,
      "loss": 0.6907,
      "step": 1524
    },
    {
      "epoch": 0.5305058230488441,
      "grad_norm": 0.5012202349580023,
      "learning_rate": 0.00018383177834494364,
      "loss": 0.5818,
      "step": 1526
    },
    {
      "epoch": 0.5312011124630628,
      "grad_norm": 0.9534096632353795,
      "learning_rate": 0.00018376553655562945,
      "loss": 0.661,
      "step": 1528
    },
    {
      "epoch": 0.5318964018772814,
      "grad_norm": 0.7514828327361283,
      "learning_rate": 0.00018369917133182644,
      "loss": 0.5764,
      "step": 1530
    },
    {
      "epoch": 0.5325916912915001,
      "grad_norm": 0.5336990715721825,
      "learning_rate": 0.0001836326827713285,
      "loss": 0.5652,
      "step": 1532
    },
    {
      "epoch": 0.5332869807057188,
      "grad_norm": 0.5458405171879034,
      "learning_rate": 0.0001835660709721113,
      "loss": 0.577,
      "step": 1534
    },
    {
      "epoch": 0.5339822701199374,
      "grad_norm": 0.649664481255766,
      "learning_rate": 0.00018349933603233206,
      "loss": 0.5599,
      "step": 1536
    },
    {
      "epoch": 0.5346775595341561,
      "grad_norm": 0.4148065039425594,
      "learning_rate": 0.0001834324780503295,
      "loss": 0.6437,
      "step": 1538
    },
    {
      "epoch": 0.5353728489483748,
      "grad_norm": 0.821550084995287,
      "learning_rate": 0.00018336549712462353,
      "loss": 0.6837,
      "step": 1540
    },
    {
      "epoch": 0.5360681383625935,
      "grad_norm": 0.5564568145383851,
      "learning_rate": 0.00018329839335391542,
      "loss": 0.5715,
      "step": 1542
    },
    {
      "epoch": 0.5367634277768121,
      "grad_norm": 0.6705822690313288,
      "learning_rate": 0.0001832311668370873,
      "loss": 0.5669,
      "step": 1544
    },
    {
      "epoch": 0.5374587171910308,
      "grad_norm": 0.7821014824731976,
      "learning_rate": 0.00018316381767320228,
      "loss": 0.6117,
      "step": 1546
    },
    {
      "epoch": 0.5381540066052495,
      "grad_norm": 0.416568249234258,
      "learning_rate": 0.00018309634596150412,
      "loss": 0.5604,
      "step": 1548
    },
    {
      "epoch": 0.5388492960194681,
      "grad_norm": 0.6063645649473942,
      "learning_rate": 0.0001830287518014172,
      "loss": 0.6944,
      "step": 1550
    },
    {
      "epoch": 0.5395445854336868,
      "grad_norm": 0.7965496657255345,
      "learning_rate": 0.0001829610352925463,
      "loss": 0.5101,
      "step": 1552
    },
    {
      "epoch": 0.5402398748479055,
      "grad_norm": 0.6352930080090265,
      "learning_rate": 0.0001828931965346766,
      "loss": 0.5392,
      "step": 1554
    },
    {
      "epoch": 0.5409351642621241,
      "grad_norm": 0.7018715777498272,
      "learning_rate": 0.00018282523562777327,
      "loss": 0.5115,
      "step": 1556
    },
    {
      "epoch": 0.5416304536763428,
      "grad_norm": 0.7055081199927248,
      "learning_rate": 0.0001827571526719816,
      "loss": 0.6342,
      "step": 1558
    },
    {
      "epoch": 0.5423257430905615,
      "grad_norm": 0.9176268923346018,
      "learning_rate": 0.0001826889477676267,
      "loss": 0.641,
      "step": 1560
    },
    {
      "epoch": 0.5430210325047801,
      "grad_norm": 0.9614728125926221,
      "learning_rate": 0.00018262062101521328,
      "loss": 0.662,
      "step": 1562
    },
    {
      "epoch": 0.5437163219189988,
      "grad_norm": 1.1496735145261052,
      "learning_rate": 0.00018255217251542574,
      "loss": 0.537,
      "step": 1564
    },
    {
      "epoch": 0.5444116113332175,
      "grad_norm": 0.6789192030318737,
      "learning_rate": 0.00018248360236912777,
      "loss": 0.5446,
      "step": 1566
    },
    {
      "epoch": 0.5451069007474362,
      "grad_norm": 0.8374235506929065,
      "learning_rate": 0.00018241491067736242,
      "loss": 0.6755,
      "step": 1568
    },
    {
      "epoch": 0.5458021901616548,
      "grad_norm": 1.0198490239995306,
      "learning_rate": 0.0001823460975413518,
      "loss": 0.5292,
      "step": 1570
    },
    {
      "epoch": 0.5464974795758735,
      "grad_norm": 0.5392136161037133,
      "learning_rate": 0.0001822771630624969,
      "loss": 0.6962,
      "step": 1572
    },
    {
      "epoch": 0.5471927689900922,
      "grad_norm": 0.5006386802591807,
      "learning_rate": 0.00018220810734237766,
      "loss": 0.5946,
      "step": 1574
    },
    {
      "epoch": 0.5478880584043108,
      "grad_norm": 0.5478881526480635,
      "learning_rate": 0.0001821389304827526,
      "loss": 0.5844,
      "step": 1576
    },
    {
      "epoch": 0.5485833478185295,
      "grad_norm": 0.3440683975593449,
      "learning_rate": 0.0001820696325855587,
      "loss": 0.5598,
      "step": 1578
    },
    {
      "epoch": 0.5492786372327482,
      "grad_norm": 0.36658750390429906,
      "learning_rate": 0.0001820002137529114,
      "loss": 0.6105,
      "step": 1580
    },
    {
      "epoch": 0.5499739266469668,
      "grad_norm": 0.8580953766157888,
      "learning_rate": 0.0001819306740871043,
      "loss": 0.6685,
      "step": 1582
    },
    {
      "epoch": 0.5506692160611855,
      "grad_norm": 0.7163438303185133,
      "learning_rate": 0.00018186101369060902,
      "loss": 0.593,
      "step": 1584
    },
    {
      "epoch": 0.5513645054754042,
      "grad_norm": 0.34431820289313025,
      "learning_rate": 0.00018179123266607515,
      "loss": 0.6024,
      "step": 1586
    },
    {
      "epoch": 0.5520597948896228,
      "grad_norm": 0.3504729261840606,
      "learning_rate": 0.00018172133111633,
      "loss": 0.6376,
      "step": 1588
    },
    {
      "epoch": 0.5527550843038415,
      "grad_norm": 0.3806243378944845,
      "learning_rate": 0.00018165130914437843,
      "loss": 0.4432,
      "step": 1590
    },
    {
      "epoch": 0.5534503737180602,
      "grad_norm": 0.44638264263646543,
      "learning_rate": 0.00018158116685340286,
      "loss": 0.584,
      "step": 1592
    },
    {
      "epoch": 0.5541456631322789,
      "grad_norm": 0.8596816723413095,
      "learning_rate": 0.00018151090434676296,
      "loss": 0.6232,
      "step": 1594
    },
    {
      "epoch": 0.5548409525464975,
      "grad_norm": 0.5102216453134625,
      "learning_rate": 0.00018144052172799555,
      "loss": 0.6171,
      "step": 1596
    },
    {
      "epoch": 0.5555362419607162,
      "grad_norm": 0.728624434959544,
      "learning_rate": 0.00018137001910081442,
      "loss": 0.5828,
      "step": 1598
    },
    {
      "epoch": 0.5562315313749349,
      "grad_norm": 0.6993841456831388,
      "learning_rate": 0.00018129939656911018,
      "loss": 0.6398,
      "step": 1600
    },
    {
      "epoch": 0.5569268207891535,
      "grad_norm": 0.5081736994898716,
      "learning_rate": 0.00018122865423695023,
      "loss": 0.5907,
      "step": 1602
    },
    {
      "epoch": 0.5576221102033722,
      "grad_norm": 0.9216105204641707,
      "learning_rate": 0.00018115779220857835,
      "loss": 0.475,
      "step": 1604
    },
    {
      "epoch": 0.5583173996175909,
      "grad_norm": 0.5887668473902828,
      "learning_rate": 0.0001810868105884149,
      "loss": 0.6423,
      "step": 1606
    },
    {
      "epoch": 0.5590126890318095,
      "grad_norm": 0.7052708958749251,
      "learning_rate": 0.0001810157094810563,
      "loss": 0.5979,
      "step": 1608
    },
    {
      "epoch": 0.5597079784460282,
      "grad_norm": 0.6684687038470192,
      "learning_rate": 0.00018094448899127506,
      "loss": 0.5597,
      "step": 1610
    },
    {
      "epoch": 0.5604032678602469,
      "grad_norm": 0.5472474373787231,
      "learning_rate": 0.00018087314922401974,
      "loss": 0.5604,
      "step": 1612
    },
    {
      "epoch": 0.5610985572744654,
      "grad_norm": 1.046907488888251,
      "learning_rate": 0.0001808016902844145,
      "loss": 0.5588,
      "step": 1614
    },
    {
      "epoch": 0.5617938466886842,
      "grad_norm": 0.599875325305421,
      "learning_rate": 0.0001807301122777592,
      "loss": 0.5333,
      "step": 1616
    },
    {
      "epoch": 0.5624891361029029,
      "grad_norm": 0.7162383538759113,
      "learning_rate": 0.0001806584153095291,
      "loss": 0.5305,
      "step": 1618
    },
    {
      "epoch": 0.5631844255171216,
      "grad_norm": 0.7631330239323099,
      "learning_rate": 0.0001805865994853749,
      "loss": 0.6774,
      "step": 1620
    },
    {
      "epoch": 0.5638797149313401,
      "grad_norm": 0.5524440961094408,
      "learning_rate": 0.00018051466491112223,
      "loss": 0.5632,
      "step": 1622
    },
    {
      "epoch": 0.5645750043455589,
      "grad_norm": 0.4987248734291855,
      "learning_rate": 0.00018044261169277185,
      "loss": 0.6215,
      "step": 1624
    },
    {
      "epoch": 0.5652702937597776,
      "grad_norm": 0.7807389251113444,
      "learning_rate": 0.00018037043993649935,
      "loss": 0.6827,
      "step": 1626
    },
    {
      "epoch": 0.5659655831739961,
      "grad_norm": 0.7200859056625541,
      "learning_rate": 0.0001802981497486549,
      "loss": 0.6462,
      "step": 1628
    },
    {
      "epoch": 0.5666608725882148,
      "grad_norm": 0.5427861037081861,
      "learning_rate": 0.0001802257412357632,
      "loss": 0.5992,
      "step": 1630
    },
    {
      "epoch": 0.5673561620024336,
      "grad_norm": 0.720385141439196,
      "learning_rate": 0.0001801532145045235,
      "loss": 0.5756,
      "step": 1632
    },
    {
      "epoch": 0.5680514514166521,
      "grad_norm": 0.9552724965328024,
      "learning_rate": 0.000180080569661809,
      "loss": 0.6362,
      "step": 1634
    },
    {
      "epoch": 0.5687467408308708,
      "grad_norm": 0.48995975515699663,
      "learning_rate": 0.00018000780681466706,
      "loss": 0.6285,
      "step": 1636
    },
    {
      "epoch": 0.5694420302450895,
      "grad_norm": 0.3350821235074382,
      "learning_rate": 0.00017993492607031896,
      "loss": 0.6077,
      "step": 1638
    },
    {
      "epoch": 0.5701373196593081,
      "grad_norm": 0.3998325347633491,
      "learning_rate": 0.00017986192753615966,
      "loss": 0.6461,
      "step": 1640
    },
    {
      "epoch": 0.5708326090735268,
      "grad_norm": 0.4435600494500259,
      "learning_rate": 0.0001797888113197577,
      "loss": 0.6354,
      "step": 1642
    },
    {
      "epoch": 0.5715278984877455,
      "grad_norm": 0.7021719076452477,
      "learning_rate": 0.000179715577528855,
      "loss": 0.609,
      "step": 1644
    },
    {
      "epoch": 0.5722231879019642,
      "grad_norm": 0.3254773922584778,
      "learning_rate": 0.00017964222627136684,
      "loss": 0.5671,
      "step": 1646
    },
    {
      "epoch": 0.5729184773161828,
      "grad_norm": 0.33864367997251066,
      "learning_rate": 0.00017956875765538148,
      "loss": 0.5037,
      "step": 1648
    },
    {
      "epoch": 0.5736137667304015,
      "grad_norm": 0.7020501782188753,
      "learning_rate": 0.00017949517178916018,
      "loss": 0.6464,
      "step": 1650
    },
    {
      "epoch": 0.5736137667304015,
      "eval_loss": 0.5962589979171753,
      "eval_runtime": 371.2694,
      "eval_samples_per_second": 13.042,
      "eval_steps_per_second": 0.409,
      "step": 1650
    },
    {
      "epoch": 0.5743090561446202,
      "grad_norm": 0.664359018460308,
      "learning_rate": 0.00017942146878113696,
      "loss": 0.54,
      "step": 1652
    },
    {
      "epoch": 0.5750043455588388,
      "grad_norm": 0.4547846259763968,
      "learning_rate": 0.00017934764873991842,
      "loss": 0.5265,
      "step": 1654
    },
    {
      "epoch": 0.5756996349730575,
      "grad_norm": 0.44499840681620506,
      "learning_rate": 0.0001792737117742837,
      "loss": 0.6455,
      "step": 1656
    },
    {
      "epoch": 0.5763949243872762,
      "grad_norm": 1.1836252545555186,
      "learning_rate": 0.00017919965799318414,
      "loss": 0.5493,
      "step": 1658
    },
    {
      "epoch": 0.5770902138014948,
      "grad_norm": 0.5250640289043013,
      "learning_rate": 0.00017912548750574332,
      "loss": 0.5912,
      "step": 1660
    },
    {
      "epoch": 0.5777855032157135,
      "grad_norm": 0.5921000904212905,
      "learning_rate": 0.00017905120042125674,
      "loss": 0.6786,
      "step": 1662
    },
    {
      "epoch": 0.5784807926299322,
      "grad_norm": 0.5069057332373537,
      "learning_rate": 0.00017897679684919165,
      "loss": 0.6578,
      "step": 1664
    },
    {
      "epoch": 0.5791760820441508,
      "grad_norm": 0.8149193785886022,
      "learning_rate": 0.00017890227689918708,
      "loss": 0.5759,
      "step": 1666
    },
    {
      "epoch": 0.5798713714583695,
      "grad_norm": 0.7674197048262045,
      "learning_rate": 0.00017882764068105343,
      "loss": 0.5992,
      "step": 1668
    },
    {
      "epoch": 0.5805666608725882,
      "grad_norm": 0.7221408349004688,
      "learning_rate": 0.00017875288830477255,
      "loss": 0.6254,
      "step": 1670
    },
    {
      "epoch": 0.5812619502868069,
      "grad_norm": 0.36822616889584014,
      "learning_rate": 0.00017867801988049735,
      "loss": 0.5479,
      "step": 1672
    },
    {
      "epoch": 0.5819572397010255,
      "grad_norm": 0.6285282750057019,
      "learning_rate": 0.00017860303551855183,
      "loss": 0.6414,
      "step": 1674
    },
    {
      "epoch": 0.5826525291152442,
      "grad_norm": 0.5000419603028949,
      "learning_rate": 0.00017852793532943074,
      "loss": 0.5929,
      "step": 1676
    },
    {
      "epoch": 0.5833478185294629,
      "grad_norm": 0.46115086426294316,
      "learning_rate": 0.0001784527194237996,
      "loss": 0.5626,
      "step": 1678
    },
    {
      "epoch": 0.5840431079436815,
      "grad_norm": 0.9767916828736493,
      "learning_rate": 0.00017837738791249438,
      "loss": 0.5637,
      "step": 1680
    },
    {
      "epoch": 0.5847383973579002,
      "grad_norm": 0.4454990407423214,
      "learning_rate": 0.00017830194090652146,
      "loss": 0.5709,
      "step": 1682
    },
    {
      "epoch": 0.5854336867721189,
      "grad_norm": 0.5297124460963825,
      "learning_rate": 0.0001782263785170574,
      "loss": 0.5378,
      "step": 1684
    },
    {
      "epoch": 0.5861289761863375,
      "grad_norm": 0.5577701746789038,
      "learning_rate": 0.0001781507008554487,
      "loss": 0.5511,
      "step": 1686
    },
    {
      "epoch": 0.5868242656005562,
      "grad_norm": 1.2633910061819977,
      "learning_rate": 0.00017807490803321182,
      "loss": 0.3968,
      "step": 1688
    },
    {
      "epoch": 0.5875195550147749,
      "grad_norm": 0.6839443146946156,
      "learning_rate": 0.00017799900016203293,
      "loss": 0.5388,
      "step": 1690
    },
    {
      "epoch": 0.5882148444289935,
      "grad_norm": 1.1726632523971559,
      "learning_rate": 0.00017792297735376761,
      "loss": 0.6796,
      "step": 1692
    },
    {
      "epoch": 0.5889101338432122,
      "grad_norm": 1.1509568442319629,
      "learning_rate": 0.00017784683972044095,
      "loss": 0.6486,
      "step": 1694
    },
    {
      "epoch": 0.5896054232574309,
      "grad_norm": 0.8030934293115218,
      "learning_rate": 0.00017777058737424714,
      "loss": 0.5735,
      "step": 1696
    },
    {
      "epoch": 0.5903007126716495,
      "grad_norm": 0.9804610100362895,
      "learning_rate": 0.00017769422042754947,
      "loss": 0.6533,
      "step": 1698
    },
    {
      "epoch": 0.5909960020858682,
      "grad_norm": 1.8668332123481661,
      "learning_rate": 0.00017761773899288008,
      "loss": 0.7231,
      "step": 1700
    },
    {
      "epoch": 0.5916912915000869,
      "grad_norm": 0.7350290447809635,
      "learning_rate": 0.00017754114318293982,
      "loss": 0.7144,
      "step": 1702
    },
    {
      "epoch": 0.5923865809143056,
      "grad_norm": 0.44898971630294976,
      "learning_rate": 0.00017746443311059804,
      "loss": 0.5719,
      "step": 1704
    },
    {
      "epoch": 0.5930818703285242,
      "grad_norm": 0.5076851350638184,
      "learning_rate": 0.00017738760888889253,
      "loss": 0.6627,
      "step": 1706
    },
    {
      "epoch": 0.5937771597427429,
      "grad_norm": 0.817639077876227,
      "learning_rate": 0.00017731067063102925,
      "loss": 0.5764,
      "step": 1708
    },
    {
      "epoch": 0.5944724491569616,
      "grad_norm": 0.9017538169950253,
      "learning_rate": 0.0001772336184503822,
      "loss": 0.5917,
      "step": 1710
    },
    {
      "epoch": 0.5951677385711802,
      "grad_norm": 0.40946041731193694,
      "learning_rate": 0.00017715645246049326,
      "loss": 0.6128,
      "step": 1712
    },
    {
      "epoch": 0.5958630279853989,
      "grad_norm": 0.5611668605116188,
      "learning_rate": 0.00017707917277507204,
      "loss": 0.6213,
      "step": 1714
    },
    {
      "epoch": 0.5965583173996176,
      "grad_norm": 0.530189165141287,
      "learning_rate": 0.00017700177950799562,
      "loss": 0.5389,
      "step": 1716
    },
    {
      "epoch": 0.5972536068138362,
      "grad_norm": 1.2382772702026799,
      "learning_rate": 0.00017692427277330857,
      "loss": 0.5313,
      "step": 1718
    },
    {
      "epoch": 0.5979488962280549,
      "grad_norm": 0.7009388588402383,
      "learning_rate": 0.00017684665268522253,
      "loss": 0.5763,
      "step": 1720
    },
    {
      "epoch": 0.5986441856422736,
      "grad_norm": 0.4981948622376413,
      "learning_rate": 0.00017676891935811622,
      "loss": 0.6326,
      "step": 1722
    },
    {
      "epoch": 0.5993394750564922,
      "grad_norm": 0.743480238762861,
      "learning_rate": 0.00017669107290653528,
      "loss": 0.5565,
      "step": 1724
    },
    {
      "epoch": 0.6000347644707109,
      "grad_norm": 0.5198052077722481,
      "learning_rate": 0.000176613113445192,
      "loss": 0.4672,
      "step": 1726
    },
    {
      "epoch": 0.6007300538849296,
      "grad_norm": 0.8298155463428107,
      "learning_rate": 0.00017653504108896517,
      "loss": 0.5494,
      "step": 1728
    },
    {
      "epoch": 0.6014253432991483,
      "grad_norm": 0.819830510986772,
      "learning_rate": 0.0001764568559529,
      "loss": 0.7491,
      "step": 1730
    },
    {
      "epoch": 0.6021206327133669,
      "grad_norm": 0.5441657313855216,
      "learning_rate": 0.00017637855815220784,
      "loss": 0.6218,
      "step": 1732
    },
    {
      "epoch": 0.6028159221275856,
      "grad_norm": 0.6531514104491785,
      "learning_rate": 0.00017630014780226612,
      "loss": 0.5577,
      "step": 1734
    },
    {
      "epoch": 0.6035112115418043,
      "grad_norm": 0.7295099022674164,
      "learning_rate": 0.00017622162501861803,
      "loss": 0.6255,
      "step": 1736
    },
    {
      "epoch": 0.6042065009560229,
      "grad_norm": 0.6744185040628968,
      "learning_rate": 0.0001761429899169725,
      "loss": 0.6667,
      "step": 1738
    },
    {
      "epoch": 0.6049017903702416,
      "grad_norm": 0.6781720439136543,
      "learning_rate": 0.00017606424261320402,
      "loss": 0.6276,
      "step": 1740
    },
    {
      "epoch": 0.6055970797844603,
      "grad_norm": 0.5375368781216472,
      "learning_rate": 0.00017598538322335228,
      "loss": 0.6052,
      "step": 1742
    },
    {
      "epoch": 0.6062923691986789,
      "grad_norm": 0.5740302297637102,
      "learning_rate": 0.00017590641186362223,
      "loss": 0.6069,
      "step": 1744
    },
    {
      "epoch": 0.6069876586128976,
      "grad_norm": 0.7486920619809492,
      "learning_rate": 0.0001758273286503838,
      "loss": 0.6501,
      "step": 1746
    },
    {
      "epoch": 0.6076829480271163,
      "grad_norm": 0.415243418739817,
      "learning_rate": 0.00017574813370017177,
      "loss": 0.6405,
      "step": 1748
    },
    {
      "epoch": 0.6083782374413349,
      "grad_norm": 0.33508147043872166,
      "learning_rate": 0.0001756688271296855,
      "loss": 0.6736,
      "step": 1750
    },
    {
      "epoch": 0.6090735268555536,
      "grad_norm": 0.6568565253926139,
      "learning_rate": 0.0001755894090557889,
      "loss": 0.5944,
      "step": 1752
    },
    {
      "epoch": 0.6097688162697723,
      "grad_norm": 0.3600602933243024,
      "learning_rate": 0.00017550987959551013,
      "loss": 0.5912,
      "step": 1754
    },
    {
      "epoch": 0.610464105683991,
      "grad_norm": 0.37379015662929643,
      "learning_rate": 0.00017543023886604158,
      "loss": 0.6873,
      "step": 1756
    },
    {
      "epoch": 0.6111593950982096,
      "grad_norm": 0.4236250679062018,
      "learning_rate": 0.00017535048698473951,
      "loss": 0.5878,
      "step": 1758
    },
    {
      "epoch": 0.6118546845124283,
      "grad_norm": 0.32077385067052405,
      "learning_rate": 0.00017527062406912398,
      "loss": 0.5699,
      "step": 1760
    },
    {
      "epoch": 0.612549973926647,
      "grad_norm": 0.7516616152359099,
      "learning_rate": 0.0001751906502368787,
      "loss": 0.5472,
      "step": 1762
    },
    {
      "epoch": 0.6132452633408656,
      "grad_norm": 0.40617593547014996,
      "learning_rate": 0.00017511056560585083,
      "loss": 0.6116,
      "step": 1764
    },
    {
      "epoch": 0.6139405527550843,
      "grad_norm": 0.6589240069077935,
      "learning_rate": 0.00017503037029405078,
      "loss": 0.5217,
      "step": 1766
    },
    {
      "epoch": 0.614635842169303,
      "grad_norm": 0.45796835520264856,
      "learning_rate": 0.00017495006441965204,
      "loss": 0.6046,
      "step": 1768
    },
    {
      "epoch": 0.6153311315835216,
      "grad_norm": 0.7238836440483356,
      "learning_rate": 0.00017486964810099106,
      "loss": 0.5221,
      "step": 1770
    },
    {
      "epoch": 0.6160264209977403,
      "grad_norm": 0.6979943995723972,
      "learning_rate": 0.00017478912145656698,
      "loss": 0.6296,
      "step": 1772
    },
    {
      "epoch": 0.616721710411959,
      "grad_norm": 0.8707905832977048,
      "learning_rate": 0.0001747084846050416,
      "loss": 0.5417,
      "step": 1774
    },
    {
      "epoch": 0.6174169998261776,
      "grad_norm": 0.7216519551841216,
      "learning_rate": 0.00017462773766523904,
      "loss": 0.641,
      "step": 1776
    },
    {
      "epoch": 0.6181122892403963,
      "grad_norm": 0.661681773522186,
      "learning_rate": 0.00017454688075614567,
      "loss": 0.6498,
      "step": 1778
    },
    {
      "epoch": 0.618807578654615,
      "grad_norm": 0.5732014964202184,
      "learning_rate": 0.00017446591399690995,
      "loss": 0.5302,
      "step": 1780
    },
    {
      "epoch": 0.6195028680688337,
      "grad_norm": 0.8272395967155222,
      "learning_rate": 0.00017438483750684212,
      "loss": 0.6592,
      "step": 1782
    },
    {
      "epoch": 0.6201981574830523,
      "grad_norm": 0.6210836312365995,
      "learning_rate": 0.00017430365140541415,
      "loss": 0.6877,
      "step": 1784
    },
    {
      "epoch": 0.620893446897271,
      "grad_norm": 0.46170985659939656,
      "learning_rate": 0.00017422235581225962,
      "loss": 0.6255,
      "step": 1786
    },
    {
      "epoch": 0.6215887363114897,
      "grad_norm": 0.3883099305940657,
      "learning_rate": 0.00017414095084717333,
      "loss": 0.5809,
      "step": 1788
    },
    {
      "epoch": 0.6222840257257083,
      "grad_norm": 0.43704267209324255,
      "learning_rate": 0.00017405943663011133,
      "loss": 0.5123,
      "step": 1790
    },
    {
      "epoch": 0.622979315139927,
      "grad_norm": 0.3930491987058269,
      "learning_rate": 0.0001739778132811906,
      "loss": 0.6012,
      "step": 1792
    },
    {
      "epoch": 0.6236746045541457,
      "grad_norm": 0.412852657314726,
      "learning_rate": 0.00017389608092068892,
      "loss": 0.6531,
      "step": 1794
    },
    {
      "epoch": 0.6243698939683643,
      "grad_norm": 0.36414563480361445,
      "learning_rate": 0.00017381423966904486,
      "loss": 0.5236,
      "step": 1796
    },
    {
      "epoch": 0.625065183382583,
      "grad_norm": 0.5001274802739109,
      "learning_rate": 0.00017373228964685726,
      "loss": 0.579,
      "step": 1798
    },
    {
      "epoch": 0.6257604727968017,
      "grad_norm": 0.4317988926228409,
      "learning_rate": 0.00017365023097488534,
      "loss": 0.5123,
      "step": 1800
    },
    {
      "epoch": 0.6257604727968017,
      "eval_loss": 0.6052089929580688,
      "eval_runtime": 370.9182,
      "eval_samples_per_second": 13.054,
      "eval_steps_per_second": 0.41,
      "step": 1800
    },
    {
      "epoch": 0.6264557622110203,
      "grad_norm": 0.5204270522892355,
      "learning_rate": 0.00017356806377404844,
      "loss": 0.653,
      "step": 1802
    },
    {
      "epoch": 0.627151051625239,
      "grad_norm": 0.7608529412962103,
      "learning_rate": 0.00017348578816542572,
      "loss": 0.6875,
      "step": 1804
    },
    {
      "epoch": 0.6278463410394577,
      "grad_norm": 0.6530765061280092,
      "learning_rate": 0.00017340340427025614,
      "loss": 0.573,
      "step": 1806
    },
    {
      "epoch": 0.6285416304536764,
      "grad_norm": 0.4895237421760521,
      "learning_rate": 0.0001733209122099383,
      "loss": 0.6624,
      "step": 1808
    },
    {
      "epoch": 0.629236919867895,
      "grad_norm": 0.4397644961630881,
      "learning_rate": 0.00017323831210603016,
      "loss": 0.5753,
      "step": 1810
    },
    {
      "epoch": 0.6299322092821137,
      "grad_norm": 0.8716885012986322,
      "learning_rate": 0.0001731556040802488,
      "loss": 0.634,
      "step": 1812
    },
    {
      "epoch": 0.6306274986963324,
      "grad_norm": 0.53660029520689,
      "learning_rate": 0.0001730727882544704,
      "loss": 0.547,
      "step": 1814
    },
    {
      "epoch": 0.631322788110551,
      "grad_norm": 0.6693656512741157,
      "learning_rate": 0.00017298986475072997,
      "loss": 0.542,
      "step": 1816
    },
    {
      "epoch": 0.6320180775247697,
      "grad_norm": 0.6427881239550128,
      "learning_rate": 0.00017290683369122122,
      "loss": 0.5778,
      "step": 1818
    },
    {
      "epoch": 0.6327133669389884,
      "grad_norm": 0.5700659225992518,
      "learning_rate": 0.00017282369519829635,
      "loss": 0.6321,
      "step": 1820
    },
    {
      "epoch": 0.633408656353207,
      "grad_norm": 0.5947857078301825,
      "learning_rate": 0.00017274044939446584,
      "loss": 0.6398,
      "step": 1822
    },
    {
      "epoch": 0.6341039457674257,
      "grad_norm": 0.8875363716928713,
      "learning_rate": 0.00017265709640239835,
      "loss": 0.6265,
      "step": 1824
    },
    {
      "epoch": 0.6347992351816444,
      "grad_norm": 0.5083457538417298,
      "learning_rate": 0.0001725736363449204,
      "loss": 0.5039,
      "step": 1826
    },
    {
      "epoch": 0.635494524595863,
      "grad_norm": 0.7572257750500727,
      "learning_rate": 0.00017249006934501635,
      "loss": 0.5638,
      "step": 1828
    },
    {
      "epoch": 0.6361898140100817,
      "grad_norm": 0.4381329875262636,
      "learning_rate": 0.0001724063955258281,
      "loss": 0.5685,
      "step": 1830
    },
    {
      "epoch": 0.6368851034243004,
      "grad_norm": 0.4769668851895218,
      "learning_rate": 0.00017232261501065507,
      "loss": 0.5773,
      "step": 1832
    },
    {
      "epoch": 0.6375803928385191,
      "grad_norm": 0.6333517381851363,
      "learning_rate": 0.00017223872792295378,
      "loss": 0.5593,
      "step": 1834
    },
    {
      "epoch": 0.6382756822527377,
      "grad_norm": 0.5103399981068355,
      "learning_rate": 0.0001721547343863378,
      "loss": 0.5575,
      "step": 1836
    },
    {
      "epoch": 0.6389709716669564,
      "grad_norm": 0.6125228317995195,
      "learning_rate": 0.00017207063452457763,
      "loss": 0.565,
      "step": 1838
    },
    {
      "epoch": 0.6396662610811751,
      "grad_norm": 0.7299577367692995,
      "learning_rate": 0.0001719864284616004,
      "loss": 0.7102,
      "step": 1840
    },
    {
      "epoch": 0.6403615504953937,
      "grad_norm": 0.8462870412150455,
      "learning_rate": 0.00017190211632148978,
      "loss": 0.5988,
      "step": 1842
    },
    {
      "epoch": 0.6410568399096124,
      "grad_norm": 0.6376476795268616,
      "learning_rate": 0.00017181769822848563,
      "loss": 0.6031,
      "step": 1844
    },
    {
      "epoch": 0.6417521293238311,
      "grad_norm": 0.5498259619130056,
      "learning_rate": 0.00017173317430698413,
      "loss": 0.604,
      "step": 1846
    },
    {
      "epoch": 0.6424474187380497,
      "grad_norm": 0.6402197493873277,
      "learning_rate": 0.00017164854468153723,
      "loss": 0.5663,
      "step": 1848
    },
    {
      "epoch": 0.6431427081522684,
      "grad_norm": 0.49473627938254194,
      "learning_rate": 0.00017156380947685275,
      "loss": 0.4801,
      "step": 1850
    },
    {
      "epoch": 0.6438379975664871,
      "grad_norm": 0.5864701801475597,
      "learning_rate": 0.00017147896881779403,
      "loss": 0.5787,
      "step": 1852
    },
    {
      "epoch": 0.6445332869807057,
      "grad_norm": 0.9424086488351346,
      "learning_rate": 0.00017139402282937986,
      "loss": 0.6761,
      "step": 1854
    },
    {
      "epoch": 0.6452285763949244,
      "grad_norm": 0.5754580414862842,
      "learning_rate": 0.0001713089716367842,
      "loss": 0.5768,
      "step": 1856
    },
    {
      "epoch": 0.6459238658091431,
      "grad_norm": 0.6945657719528112,
      "learning_rate": 0.000171223815365336,
      "loss": 0.6376,
      "step": 1858
    },
    {
      "epoch": 0.6466191552233618,
      "grad_norm": 0.5299829269045951,
      "learning_rate": 0.00017113855414051918,
      "loss": 0.6682,
      "step": 1860
    },
    {
      "epoch": 0.6473144446375804,
      "grad_norm": 0.6178382762886556,
      "learning_rate": 0.00017105318808797215,
      "loss": 0.5745,
      "step": 1862
    },
    {
      "epoch": 0.6480097340517991,
      "grad_norm": 0.5777288231745069,
      "learning_rate": 0.00017096771733348794,
      "loss": 0.4917,
      "step": 1864
    },
    {
      "epoch": 0.6487050234660178,
      "grad_norm": 0.4060689941848452,
      "learning_rate": 0.00017088214200301373,
      "loss": 0.5554,
      "step": 1866
    },
    {
      "epoch": 0.6494003128802364,
      "grad_norm": 0.35541750930109944,
      "learning_rate": 0.0001707964622226509,
      "loss": 0.6036,
      "step": 1868
    },
    {
      "epoch": 0.6500956022944551,
      "grad_norm": 0.497757325255477,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.5215,
      "step": 1870
    },
    {
      "epoch": 0.6507908917086738,
      "grad_norm": 1.155534242094426,
      "learning_rate": 0.00017062478981743424,
      "loss": 0.6362,
      "step": 1872
    },
    {
      "epoch": 0.6514861811228924,
      "grad_norm": 0.5222241972069097,
      "learning_rate": 0.0001705387974455519,
      "loss": 0.593,
      "step": 1874
    },
    {
      "epoch": 0.6521814705371111,
      "grad_norm": 0.5041434419525581,
      "learning_rate": 0.00017045270112972362,
      "loss": 0.6426,
      "step": 1876
    },
    {
      "epoch": 0.6528767599513298,
      "grad_norm": 0.7687464251703052,
      "learning_rate": 0.00017036650099681847,
      "loss": 0.5629,
      "step": 1878
    },
    {
      "epoch": 0.6535720493655484,
      "grad_norm": 0.6523439997352839,
      "learning_rate": 0.0001702801971738585,
      "loss": 0.6066,
      "step": 1880
    },
    {
      "epoch": 0.6542673387797671,
      "grad_norm": 0.7148868461231821,
      "learning_rate": 0.0001701937897880185,
      "loss": 0.6964,
      "step": 1882
    },
    {
      "epoch": 0.6549626281939858,
      "grad_norm": 0.5711926468750704,
      "learning_rate": 0.00017010727896662594,
      "loss": 0.6143,
      "step": 1884
    },
    {
      "epoch": 0.6556579176082045,
      "grad_norm": 0.5433509637373257,
      "learning_rate": 0.00017002066483716067,
      "loss": 0.5491,
      "step": 1886
    },
    {
      "epoch": 0.6563532070224231,
      "grad_norm": 0.5034407658517631,
      "learning_rate": 0.00016993394752725483,
      "loss": 0.5197,
      "step": 1888
    },
    {
      "epoch": 0.6570484964366418,
      "grad_norm": 0.8373667811859606,
      "learning_rate": 0.0001698471271646925,
      "loss": 0.5873,
      "step": 1890
    },
    {
      "epoch": 0.6577437858508605,
      "grad_norm": 0.6227498724251839,
      "learning_rate": 0.0001697602038774097,
      "loss": 0.6962,
      "step": 1892
    },
    {
      "epoch": 0.658439075265079,
      "grad_norm": 1.2351405074494481,
      "learning_rate": 0.00016967317779349408,
      "loss": 0.5342,
      "step": 1894
    },
    {
      "epoch": 0.6591343646792978,
      "grad_norm": 0.5467600967193789,
      "learning_rate": 0.00016958604904118477,
      "loss": 0.5608,
      "step": 1896
    },
    {
      "epoch": 0.6598296540935165,
      "grad_norm": 0.5963111593389001,
      "learning_rate": 0.00016949881774887225,
      "loss": 0.517,
      "step": 1898
    },
    {
      "epoch": 0.660524943507735,
      "grad_norm": 0.7374594651213904,
      "learning_rate": 0.00016941148404509796,
      "loss": 0.4725,
      "step": 1900
    },
    {
      "epoch": 0.6612202329219538,
      "grad_norm": 0.9472414514119026,
      "learning_rate": 0.00016932404805855442,
      "loss": 0.5568,
      "step": 1902
    },
    {
      "epoch": 0.6619155223361725,
      "grad_norm": 1.0041060773211288,
      "learning_rate": 0.00016923650991808472,
      "loss": 0.603,
      "step": 1904
    },
    {
      "epoch": 0.662610811750391,
      "grad_norm": 1.0580932026507945,
      "learning_rate": 0.00016914886975268257,
      "loss": 0.6036,
      "step": 1906
    },
    {
      "epoch": 0.6633061011646098,
      "grad_norm": 0.8767482713317837,
      "learning_rate": 0.00016906112769149204,
      "loss": 0.5834,
      "step": 1908
    },
    {
      "epoch": 0.6640013905788285,
      "grad_norm": 1.012539952053466,
      "learning_rate": 0.00016897328386380732,
      "loss": 0.6149,
      "step": 1910
    },
    {
      "epoch": 0.6646966799930472,
      "grad_norm": 0.6627172772277559,
      "learning_rate": 0.00016888533839907246,
      "loss": 0.5652,
      "step": 1912
    },
    {
      "epoch": 0.6653919694072657,
      "grad_norm": 0.5784691725846585,
      "learning_rate": 0.00016879729142688142,
      "loss": 0.5892,
      "step": 1914
    },
    {
      "epoch": 0.6660872588214845,
      "grad_norm": 0.5351447888394464,
      "learning_rate": 0.00016870914307697774,
      "loss": 0.5569,
      "step": 1916
    },
    {
      "epoch": 0.6667825482357032,
      "grad_norm": 0.6075480060783698,
      "learning_rate": 0.00016862089347925423,
      "loss": 0.5714,
      "step": 1918
    },
    {
      "epoch": 0.6674778376499217,
      "grad_norm": 0.7061964040177015,
      "learning_rate": 0.00016853254276375296,
      "loss": 0.6222,
      "step": 1920
    },
    {
      "epoch": 0.6681731270641404,
      "grad_norm": 0.7493781087201877,
      "learning_rate": 0.00016844409106066505,
      "loss": 0.5536,
      "step": 1922
    },
    {
      "epoch": 0.6688684164783592,
      "grad_norm": 0.6889099042823535,
      "learning_rate": 0.0001683555385003304,
      "loss": 0.5381,
      "step": 1924
    },
    {
      "epoch": 0.6695637058925777,
      "grad_norm": 0.686174384092237,
      "learning_rate": 0.00016826688521323746,
      "loss": 0.5206,
      "step": 1926
    },
    {
      "epoch": 0.6702589953067964,
      "grad_norm": 0.6685983626439217,
      "learning_rate": 0.0001681781313300232,
      "loss": 0.619,
      "step": 1928
    },
    {
      "epoch": 0.6709542847210151,
      "grad_norm": 0.6684510056843382,
      "learning_rate": 0.0001680892769814728,
      "loss": 0.5298,
      "step": 1930
    },
    {
      "epoch": 0.6716495741352337,
      "grad_norm": 0.629367010591366,
      "learning_rate": 0.00016800032229851943,
      "loss": 0.556,
      "step": 1932
    },
    {
      "epoch": 0.6723448635494524,
      "grad_norm": 0.6511793907808336,
      "learning_rate": 0.0001679112674122442,
      "loss": 0.5667,
      "step": 1934
    },
    {
      "epoch": 0.6730401529636711,
      "grad_norm": 0.8544347958530318,
      "learning_rate": 0.00016782211245387586,
      "loss": 0.541,
      "step": 1936
    },
    {
      "epoch": 0.6737354423778898,
      "grad_norm": 0.6862974328397455,
      "learning_rate": 0.00016773285755479055,
      "loss": 0.5718,
      "step": 1938
    },
    {
      "epoch": 0.6744307317921084,
      "grad_norm": 0.9064411135544665,
      "learning_rate": 0.00016764350284651172,
      "loss": 0.5383,
      "step": 1940
    },
    {
      "epoch": 0.6751260212063271,
      "grad_norm": 1.2798081471760363,
      "learning_rate": 0.00016755404846070993,
      "loss": 0.6409,
      "step": 1942
    },
    {
      "epoch": 0.6758213106205458,
      "grad_norm": 0.7031747774944076,
      "learning_rate": 0.0001674644945292026,
      "loss": 0.6129,
      "step": 1944
    },
    {
      "epoch": 0.6765166000347644,
      "grad_norm": 0.8672546418037201,
      "learning_rate": 0.0001673748411839538,
      "loss": 0.5397,
      "step": 1946
    },
    {
      "epoch": 0.6772118894489831,
      "grad_norm": 1.3071689607309334,
      "learning_rate": 0.00016728508855707412,
      "loss": 0.6809,
      "step": 1948
    },
    {
      "epoch": 0.6779071788632018,
      "grad_norm": 0.43492321965313424,
      "learning_rate": 0.00016719523678082048,
      "loss": 0.5308,
      "step": 1950
    },
    {
      "epoch": 0.6779071788632018,
      "eval_loss": 0.6056139469146729,
      "eval_runtime": 371.18,
      "eval_samples_per_second": 13.045,
      "eval_steps_per_second": 0.41,
      "step": 1950
    },
    {
      "epoch": 0.6786024682774204,
      "grad_norm": 0.6569772769573995,
      "learning_rate": 0.00016710528598759588,
      "loss": 0.6151,
      "step": 1952
    },
    {
      "epoch": 0.6792977576916391,
      "grad_norm": 0.6233123597755471,
      "learning_rate": 0.00016701523630994921,
      "loss": 0.6345,
      "step": 1954
    },
    {
      "epoch": 0.6799930471058578,
      "grad_norm": 0.756483194755832,
      "learning_rate": 0.00016692508788057508,
      "loss": 0.6423,
      "step": 1956
    },
    {
      "epoch": 0.6806883365200764,
      "grad_norm": 0.5012190330749898,
      "learning_rate": 0.00016683484083231367,
      "loss": 0.5771,
      "step": 1958
    },
    {
      "epoch": 0.6813836259342951,
      "grad_norm": 0.45697084291469786,
      "learning_rate": 0.00016674449529815037,
      "loss": 0.6064,
      "step": 1960
    },
    {
      "epoch": 0.6820789153485138,
      "grad_norm": 0.6839659288985819,
      "learning_rate": 0.0001666540514112158,
      "loss": 0.6019,
      "step": 1962
    },
    {
      "epoch": 0.6827742047627325,
      "grad_norm": 0.6790048429974062,
      "learning_rate": 0.0001665635093047855,
      "loss": 0.5919,
      "step": 1964
    },
    {
      "epoch": 0.6834694941769511,
      "grad_norm": 0.7067958138632173,
      "learning_rate": 0.00016647286911227968,
      "loss": 0.5737,
      "step": 1966
    },
    {
      "epoch": 0.6841647835911698,
      "grad_norm": 0.6737064355710938,
      "learning_rate": 0.00016638213096726312,
      "loss": 0.6315,
      "step": 1968
    },
    {
      "epoch": 0.6848600730053885,
      "grad_norm": 0.39284397968929874,
      "learning_rate": 0.000166291295003445,
      "loss": 0.5519,
      "step": 1970
    },
    {
      "epoch": 0.6855553624196071,
      "grad_norm": 0.6110036966138074,
      "learning_rate": 0.00016620036135467855,
      "loss": 0.5699,
      "step": 1972
    },
    {
      "epoch": 0.6862506518338258,
      "grad_norm": 0.674185155315815,
      "learning_rate": 0.0001661093301549609,
      "loss": 0.6632,
      "step": 1974
    },
    {
      "epoch": 0.6869459412480445,
      "grad_norm": 1.34627131454609,
      "learning_rate": 0.0001660182015384332,
      "loss": 0.6096,
      "step": 1976
    },
    {
      "epoch": 0.6876412306622631,
      "grad_norm": 0.644484450399573,
      "learning_rate": 0.0001659269756393798,
      "loss": 0.5991,
      "step": 1978
    },
    {
      "epoch": 0.6883365200764818,
      "grad_norm": 0.4990172041361588,
      "learning_rate": 0.00016583565259222863,
      "loss": 0.5416,
      "step": 1980
    },
    {
      "epoch": 0.6890318094907005,
      "grad_norm": 0.4505095876483144,
      "learning_rate": 0.0001657442325315507,
      "loss": 0.5676,
      "step": 1982
    },
    {
      "epoch": 0.6897270989049191,
      "grad_norm": 0.46794666181349276,
      "learning_rate": 0.00016565271559205997,
      "loss": 0.5,
      "step": 1984
    },
    {
      "epoch": 0.6904223883191378,
      "grad_norm": 0.4544548690857851,
      "learning_rate": 0.0001655611019086132,
      "loss": 0.4838,
      "step": 1986
    },
    {
      "epoch": 0.6911176777333565,
      "grad_norm": 1.3379424680615957,
      "learning_rate": 0.00016546939161620966,
      "loss": 0.609,
      "step": 1988
    },
    {
      "epoch": 0.6918129671475752,
      "grad_norm": 0.717144004391169,
      "learning_rate": 0.00016537758484999105,
      "loss": 0.587,
      "step": 1990
    },
    {
      "epoch": 0.6925082565617938,
      "grad_norm": 0.8519131395212922,
      "learning_rate": 0.0001652856817452411,
      "loss": 0.6149,
      "step": 1992
    },
    {
      "epoch": 0.6932035459760125,
      "grad_norm": 0.8247198494798238,
      "learning_rate": 0.00016519368243738566,
      "loss": 0.5559,
      "step": 1994
    },
    {
      "epoch": 0.6938988353902312,
      "grad_norm": 0.9972243594284758,
      "learning_rate": 0.00016510158706199222,
      "loss": 0.7695,
      "step": 1996
    },
    {
      "epoch": 0.6945941248044498,
      "grad_norm": 0.6775709255460454,
      "learning_rate": 0.00016500939575476992,
      "loss": 0.5546,
      "step": 1998
    },
    {
      "epoch": 0.6952894142186685,
      "grad_norm": 0.6270536790838611,
      "learning_rate": 0.0001649171086515692,
      "loss": 0.6676,
      "step": 2000
    },
    {
      "epoch": 0.6959847036328872,
      "grad_norm": 0.5247284479563707,
      "learning_rate": 0.0001648247258883817,
      "loss": 0.6573,
      "step": 2002
    },
    {
      "epoch": 0.6966799930471058,
      "grad_norm": 1.3371640454637324,
      "learning_rate": 0.00016473224760134,
      "loss": 0.5688,
      "step": 2004
    },
    {
      "epoch": 0.6973752824613245,
      "grad_norm": 0.537239749385301,
      "learning_rate": 0.0001646396739267174,
      "loss": 0.5238,
      "step": 2006
    },
    {
      "epoch": 0.6980705718755432,
      "grad_norm": 0.683351825211432,
      "learning_rate": 0.00016454700500092786,
      "loss": 0.576,
      "step": 2008
    },
    {
      "epoch": 0.6987658612897618,
      "grad_norm": 1.266816676228039,
      "learning_rate": 0.00016445424096052567,
      "loss": 0.592,
      "step": 2010
    },
    {
      "epoch": 0.6994611507039805,
      "grad_norm": 0.8922765874170949,
      "learning_rate": 0.00016436138194220518,
      "loss": 0.5547,
      "step": 2012
    },
    {
      "epoch": 0.7001564401181992,
      "grad_norm": 0.602053244835111,
      "learning_rate": 0.00016426842808280086,
      "loss": 0.6122,
      "step": 2014
    },
    {
      "epoch": 0.7008517295324179,
      "grad_norm": 0.7130195810195582,
      "learning_rate": 0.00016417537951928676,
      "loss": 0.5297,
      "step": 2016
    },
    {
      "epoch": 0.7015470189466365,
      "grad_norm": 0.8636165886047156,
      "learning_rate": 0.00016408223638877665,
      "loss": 0.5752,
      "step": 2018
    },
    {
      "epoch": 0.7022423083608552,
      "grad_norm": 0.7876973687914621,
      "learning_rate": 0.00016398899882852353,
      "loss": 0.6652,
      "step": 2020
    },
    {
      "epoch": 0.7029375977750739,
      "grad_norm": 0.6531287665955688,
      "learning_rate": 0.0001638956669759196,
      "loss": 0.6521,
      "step": 2022
    },
    {
      "epoch": 0.7036328871892925,
      "grad_norm": 0.8086172007984764,
      "learning_rate": 0.00016380224096849597,
      "loss": 0.5881,
      "step": 2024
    },
    {
      "epoch": 0.7043281766035112,
      "grad_norm": 0.47789458830184245,
      "learning_rate": 0.00016370872094392255,
      "loss": 0.5258,
      "step": 2026
    },
    {
      "epoch": 0.7050234660177299,
      "grad_norm": 0.5867593662603546,
      "learning_rate": 0.00016361510704000778,
      "loss": 0.5487,
      "step": 2028
    },
    {
      "epoch": 0.7057187554319485,
      "grad_norm": 0.743493145379613,
      "learning_rate": 0.00016352139939469838,
      "loss": 0.6646,
      "step": 2030
    },
    {
      "epoch": 0.7064140448461672,
      "grad_norm": 0.5420751844584678,
      "learning_rate": 0.00016342759814607928,
      "loss": 0.6296,
      "step": 2032
    },
    {
      "epoch": 0.7071093342603859,
      "grad_norm": 0.4767527091363965,
      "learning_rate": 0.00016333370343237324,
      "loss": 0.6135,
      "step": 2034
    },
    {
      "epoch": 0.7078046236746045,
      "grad_norm": 0.7559831165564748,
      "learning_rate": 0.0001632397153919409,
      "loss": 0.6138,
      "step": 2036
    },
    {
      "epoch": 0.7084999130888232,
      "grad_norm": 0.442317218000476,
      "learning_rate": 0.0001631456341632803,
      "loss": 0.5502,
      "step": 2038
    },
    {
      "epoch": 0.7091952025030419,
      "grad_norm": 0.9012866177348007,
      "learning_rate": 0.00016305145988502684,
      "loss": 0.5459,
      "step": 2040
    },
    {
      "epoch": 0.7098904919172606,
      "grad_norm": 0.7045775124035536,
      "learning_rate": 0.000162957192695953,
      "loss": 0.4986,
      "step": 2042
    },
    {
      "epoch": 0.7105857813314792,
      "grad_norm": 0.5788828574335951,
      "learning_rate": 0.00016286283273496825,
      "loss": 0.5706,
      "step": 2044
    },
    {
      "epoch": 0.7112810707456979,
      "grad_norm": 0.7709837646633183,
      "learning_rate": 0.00016276838014111868,
      "loss": 0.6708,
      "step": 2046
    },
    {
      "epoch": 0.7119763601599166,
      "grad_norm": 0.9894566046216766,
      "learning_rate": 0.00016267383505358694,
      "loss": 0.4963,
      "step": 2048
    },
    {
      "epoch": 0.7126716495741352,
      "grad_norm": 0.9904706283602799,
      "learning_rate": 0.00016257919761169195,
      "loss": 0.6312,
      "step": 2050
    },
    {
      "epoch": 0.7133669389883539,
      "grad_norm": 1.2851758328250367,
      "learning_rate": 0.00016248446795488874,
      "loss": 0.55,
      "step": 2052
    },
    {
      "epoch": 0.7140622284025726,
      "grad_norm": 1.280763519444461,
      "learning_rate": 0.00016238964622276817,
      "loss": 0.6931,
      "step": 2054
    },
    {
      "epoch": 0.7147575178167912,
      "grad_norm": 1.4273656641407675,
      "learning_rate": 0.00016229473255505693,
      "loss": 0.5302,
      "step": 2056
    },
    {
      "epoch": 0.7154528072310099,
      "grad_norm": 0.6347338138187268,
      "learning_rate": 0.0001621997270916169,
      "loss": 0.5815,
      "step": 2058
    },
    {
      "epoch": 0.7161480966452286,
      "grad_norm": 0.6533886960783342,
      "learning_rate": 0.00016210462997244554,
      "loss": 0.6818,
      "step": 2060
    },
    {
      "epoch": 0.7168433860594472,
      "grad_norm": 0.5939404538830146,
      "learning_rate": 0.0001620094413376752,
      "loss": 0.5961,
      "step": 2062
    },
    {
      "epoch": 0.7175386754736659,
      "grad_norm": 1.0489562805261028,
      "learning_rate": 0.00016191416132757305,
      "loss": 0.5698,
      "step": 2064
    },
    {
      "epoch": 0.7182339648878846,
      "grad_norm": 0.45667736793871694,
      "learning_rate": 0.00016181879008254113,
      "loss": 0.5632,
      "step": 2066
    },
    {
      "epoch": 0.7189292543021033,
      "grad_norm": 0.6278641019713552,
      "learning_rate": 0.00016172332774311562,
      "loss": 0.6647,
      "step": 2068
    },
    {
      "epoch": 0.7196245437163219,
      "grad_norm": 0.47629081990627226,
      "learning_rate": 0.00016162777444996712,
      "loss": 0.6373,
      "step": 2070
    },
    {
      "epoch": 0.7203198331305406,
      "grad_norm": 0.6149498576605736,
      "learning_rate": 0.00016153213034390026,
      "loss": 0.5118,
      "step": 2072
    },
    {
      "epoch": 0.7210151225447593,
      "grad_norm": 0.56362872937635,
      "learning_rate": 0.00016143639556585342,
      "loss": 0.5284,
      "step": 2074
    },
    {
      "epoch": 0.7217104119589779,
      "grad_norm": 0.45928718600771984,
      "learning_rate": 0.00016134057025689858,
      "loss": 0.5591,
      "step": 2076
    },
    {
      "epoch": 0.7224057013731966,
      "grad_norm": 0.9311134574972124,
      "learning_rate": 0.00016124465455824127,
      "loss": 0.6373,
      "step": 2078
    },
    {
      "epoch": 0.7231009907874153,
      "grad_norm": 0.8117606338423109,
      "learning_rate": 0.00016114864861122,
      "loss": 0.6172,
      "step": 2080
    },
    {
      "epoch": 0.7237962802016339,
      "grad_norm": 0.9415924380727096,
      "learning_rate": 0.00016105255255730643,
      "loss": 0.651,
      "step": 2082
    },
    {
      "epoch": 0.7244915696158526,
      "grad_norm": 1.1201310609340696,
      "learning_rate": 0.00016095636653810493,
      "loss": 0.501,
      "step": 2084
    },
    {
      "epoch": 0.7251868590300713,
      "grad_norm": 0.7532957479811233,
      "learning_rate": 0.00016086009069535243,
      "loss": 0.6211,
      "step": 2086
    },
    {
      "epoch": 0.7258821484442899,
      "grad_norm": 0.4550840645762033,
      "learning_rate": 0.00016076372517091831,
      "loss": 0.5765,
      "step": 2088
    },
    {
      "epoch": 0.7265774378585086,
      "grad_norm": 0.5721008820271746,
      "learning_rate": 0.000160667270106804,
      "loss": 0.5022,
      "step": 2090
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.8683128387128771,
      "learning_rate": 0.00016057072564514292,
      "loss": 0.6232,
      "step": 2092
    },
    {
      "epoch": 0.727968016686946,
      "grad_norm": 0.6494718738069951,
      "learning_rate": 0.00016047409192820018,
      "loss": 0.5538,
      "step": 2094
    },
    {
      "epoch": 0.7286633061011646,
      "grad_norm": 1.1287879234892022,
      "learning_rate": 0.00016037736909837254,
      "loss": 0.6134,
      "step": 2096
    },
    {
      "epoch": 0.7293585955153833,
      "grad_norm": 0.852132328455506,
      "learning_rate": 0.00016028055729818785,
      "loss": 0.599,
      "step": 2098
    },
    {
      "epoch": 0.730053884929602,
      "grad_norm": 0.8301319013187483,
      "learning_rate": 0.00016018365667030534,
      "loss": 0.5511,
      "step": 2100
    },
    {
      "epoch": 0.730053884929602,
      "eval_loss": 0.6051976680755615,
      "eval_runtime": 370.8053,
      "eval_samples_per_second": 13.058,
      "eval_steps_per_second": 0.41,
      "step": 2100
    },
    {
      "epoch": 0.7307491743438206,
      "grad_norm": 0.835223857619091,
      "learning_rate": 0.00016008666735751488,
      "loss": 0.5362,
      "step": 2102
    },
    {
      "epoch": 0.7314444637580393,
      "grad_norm": 0.5859220724175553,
      "learning_rate": 0.00015998958950273718,
      "loss": 0.4207,
      "step": 2104
    },
    {
      "epoch": 0.732139753172258,
      "grad_norm": 0.8195876868296976,
      "learning_rate": 0.00015989242324902341,
      "loss": 0.6456,
      "step": 2106
    },
    {
      "epoch": 0.7328350425864766,
      "grad_norm": 0.5963426019757224,
      "learning_rate": 0.00015979516873955495,
      "loss": 0.5347,
      "step": 2108
    },
    {
      "epoch": 0.7335303320006953,
      "grad_norm": 0.5598552464839238,
      "learning_rate": 0.00015969782611764322,
      "loss": 0.5609,
      "step": 2110
    },
    {
      "epoch": 0.734225621414914,
      "grad_norm": 0.9190818791386364,
      "learning_rate": 0.0001596003955267295,
      "loss": 0.6444,
      "step": 2112
    },
    {
      "epoch": 0.7349209108291326,
      "grad_norm": 0.7074763887519752,
      "learning_rate": 0.00015950287711038482,
      "loss": 0.5513,
      "step": 2114
    },
    {
      "epoch": 0.7356162002433513,
      "grad_norm": 0.5303381980872908,
      "learning_rate": 0.00015940527101230938,
      "loss": 0.5649,
      "step": 2116
    },
    {
      "epoch": 0.73631148965757,
      "grad_norm": 0.7961849958708207,
      "learning_rate": 0.0001593075773763328,
      "loss": 0.5032,
      "step": 2118
    },
    {
      "epoch": 0.7370067790717887,
      "grad_norm": 0.6321864563119038,
      "learning_rate": 0.00015920979634641364,
      "loss": 0.5128,
      "step": 2120
    },
    {
      "epoch": 0.7377020684860073,
      "grad_norm": 0.6402923888426749,
      "learning_rate": 0.00015911192806663914,
      "loss": 0.6083,
      "step": 2122
    },
    {
      "epoch": 0.738397357900226,
      "grad_norm": 1.0036392568071666,
      "learning_rate": 0.00015901397268122525,
      "loss": 0.5654,
      "step": 2124
    },
    {
      "epoch": 0.7390926473144447,
      "grad_norm": 0.639210671172982,
      "learning_rate": 0.00015891593033451616,
      "loss": 0.5165,
      "step": 2126
    },
    {
      "epoch": 0.7397879367286633,
      "grad_norm": 0.5362341486809139,
      "learning_rate": 0.00015881780117098427,
      "loss": 0.6226,
      "step": 2128
    },
    {
      "epoch": 0.740483226142882,
      "grad_norm": 0.5732167979609127,
      "learning_rate": 0.00015871958533522995,
      "loss": 0.5074,
      "step": 2130
    },
    {
      "epoch": 0.7411785155571007,
      "grad_norm": 0.5175334271404235,
      "learning_rate": 0.00015862128297198117,
      "loss": 0.6992,
      "step": 2132
    },
    {
      "epoch": 0.7418738049713193,
      "grad_norm": 0.6574502026701543,
      "learning_rate": 0.0001585228942260935,
      "loss": 0.5352,
      "step": 2134
    },
    {
      "epoch": 0.742569094385538,
      "grad_norm": 0.6282329227137664,
      "learning_rate": 0.00015842441924254975,
      "loss": 0.5883,
      "step": 2136
    },
    {
      "epoch": 0.7432643837997567,
      "grad_norm": 0.612347648551905,
      "learning_rate": 0.00015832585816645986,
      "loss": 0.6135,
      "step": 2138
    },
    {
      "epoch": 0.7439596732139753,
      "grad_norm": 0.4562291436078377,
      "learning_rate": 0.00015822721114306057,
      "loss": 0.5953,
      "step": 2140
    },
    {
      "epoch": 0.744654962628194,
      "grad_norm": 0.7370092367487316,
      "learning_rate": 0.00015812847831771533,
      "loss": 0.5933,
      "step": 2142
    },
    {
      "epoch": 0.7453502520424127,
      "grad_norm": 0.734465754561315,
      "learning_rate": 0.00015802965983591398,
      "loss": 0.5499,
      "step": 2144
    },
    {
      "epoch": 0.7460455414566314,
      "grad_norm": 0.5173035061957143,
      "learning_rate": 0.0001579307558432726,
      "loss": 0.5704,
      "step": 2146
    },
    {
      "epoch": 0.74674083087085,
      "grad_norm": 0.42722584908793315,
      "learning_rate": 0.00015783176648553323,
      "loss": 0.5696,
      "step": 2148
    },
    {
      "epoch": 0.7474361202850687,
      "grad_norm": 0.5700444049601131,
      "learning_rate": 0.00015773269190856378,
      "loss": 0.5371,
      "step": 2150
    },
    {
      "epoch": 0.7481314096992874,
      "grad_norm": 0.5135665000835338,
      "learning_rate": 0.00015763353225835776,
      "loss": 0.5928,
      "step": 2152
    },
    {
      "epoch": 0.748826699113506,
      "grad_norm": 0.9427562655572416,
      "learning_rate": 0.00015753428768103386,
      "loss": 0.5577,
      "step": 2154
    },
    {
      "epoch": 0.7495219885277247,
      "grad_norm": 1.1255169215867615,
      "learning_rate": 0.00015743495832283615,
      "loss": 0.5936,
      "step": 2156
    },
    {
      "epoch": 0.7502172779419434,
      "grad_norm": 0.9053073535451409,
      "learning_rate": 0.00015733554433013347,
      "loss": 0.5747,
      "step": 2158
    },
    {
      "epoch": 0.750912567356162,
      "grad_norm": 0.5862209644479233,
      "learning_rate": 0.0001572360458494194,
      "loss": 0.5121,
      "step": 2160
    }
  ],
  "logging_steps": 2,
  "max_steps": 5752,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 90,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8197093900943360.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}