{
  "best_metric": 0.3917270302772522,
  "best_model_checkpoint": "mikhail-panzo/ceb_b32_le5_s8000/checkpoint-6500",
  "epoch": 158.41584158415841,
  "eval_steps": 500,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9900990099009901,
      "grad_norm": 3.180250644683838,
      "learning_rate": 2.4500000000000004e-07,
      "loss": 0.8005,
      "step": 50
    },
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 3.2645788192749023,
      "learning_rate": 4.95e-07,
      "loss": 0.7901,
      "step": 100
    },
    {
      "epoch": 2.9702970297029703,
      "grad_norm": 2.118293523788452,
      "learning_rate": 7.450000000000001e-07,
      "loss": 0.7371,
      "step": 150
    },
    {
      "epoch": 3.9603960396039604,
      "grad_norm": 1.8010900020599365,
      "learning_rate": 9.950000000000002e-07,
      "loss": 0.7469,
      "step": 200
    },
    {
      "epoch": 4.9504950495049505,
      "grad_norm": 1.6043368577957153,
      "learning_rate": 1.2450000000000002e-06,
      "loss": 0.7138,
      "step": 250
    },
    {
      "epoch": 5.9405940594059405,
      "grad_norm": 2.1476693153381348,
      "learning_rate": 1.495e-06,
      "loss": 0.7011,
      "step": 300
    },
    {
      "epoch": 6.930693069306931,
      "grad_norm": 1.7792506217956543,
      "learning_rate": 1.745e-06,
      "loss": 0.6728,
      "step": 350
    },
    {
      "epoch": 7.920792079207921,
      "grad_norm": 2.7465362548828125,
      "learning_rate": 1.9950000000000004e-06,
      "loss": 0.6505,
      "step": 400
    },
    {
      "epoch": 8.910891089108912,
      "grad_norm": 1.8483725786209106,
      "learning_rate": 2.245e-06,
      "loss": 0.617,
      "step": 450
    },
    {
      "epoch": 9.900990099009901,
      "grad_norm": 2.8589906692504883,
      "learning_rate": 2.4950000000000003e-06,
      "loss": 0.5667,
      "step": 500
    },
    {
      "epoch": 9.900990099009901,
      "eval_loss": 0.47756776213645935,
      "eval_runtime": 8.0921,
      "eval_samples_per_second": 22.244,
      "eval_steps_per_second": 2.842,
      "step": 500
    },
    {
      "epoch": 10.891089108910892,
      "grad_norm": 2.1117241382598877,
      "learning_rate": 2.7450000000000004e-06,
      "loss": 0.5433,
      "step": 550
    },
    {
      "epoch": 11.881188118811881,
      "grad_norm": 1.311270833015442,
      "learning_rate": 2.995e-06,
      "loss": 0.5224,
      "step": 600
    },
    {
      "epoch": 12.871287128712872,
      "grad_norm": 2.1834664344787598,
      "learning_rate": 3.2450000000000003e-06,
      "loss": 0.5134,
      "step": 650
    },
    {
      "epoch": 13.861386138613861,
      "grad_norm": 1.2540502548217773,
      "learning_rate": 3.495e-06,
      "loss": 0.4977,
      "step": 700
    },
    {
      "epoch": 14.851485148514852,
      "grad_norm": 1.5720808506011963,
      "learning_rate": 3.745e-06,
      "loss": 0.4992,
      "step": 750
    },
    {
      "epoch": 15.841584158415841,
      "grad_norm": 1.3779664039611816,
      "learning_rate": 3.995000000000001e-06,
      "loss": 0.5041,
      "step": 800
    },
    {
      "epoch": 16.831683168316832,
      "grad_norm": 1.2948031425476074,
      "learning_rate": 4.245e-06,
      "loss": 0.4998,
      "step": 850
    },
    {
      "epoch": 17.821782178217823,
      "grad_norm": 1.5043249130249023,
      "learning_rate": 4.495e-06,
      "loss": 0.4908,
      "step": 900
    },
    {
      "epoch": 18.81188118811881,
      "grad_norm": 1.9952532052993774,
      "learning_rate": 4.745e-06,
      "loss": 0.4828,
      "step": 950
    },
    {
      "epoch": 19.801980198019802,
      "grad_norm": 1.1994812488555908,
      "learning_rate": 4.9950000000000005e-06,
      "loss": 0.4838,
      "step": 1000
    },
    {
      "epoch": 19.801980198019802,
      "eval_loss": 0.4320996403694153,
      "eval_runtime": 7.9832,
      "eval_samples_per_second": 22.547,
      "eval_steps_per_second": 2.881,
      "step": 1000
    },
    {
      "epoch": 20.792079207920793,
      "grad_norm": 2.356138229370117,
      "learning_rate": 5.245e-06,
      "loss": 0.4784,
      "step": 1050
    },
    {
      "epoch": 21.782178217821784,
      "grad_norm": 1.6155955791473389,
      "learning_rate": 5.495000000000001e-06,
      "loss": 0.4696,
      "step": 1100
    },
    {
      "epoch": 22.77227722772277,
      "grad_norm": 1.3177858591079712,
      "learning_rate": 5.745000000000001e-06,
      "loss": 0.4694,
      "step": 1150
    },
    {
      "epoch": 23.762376237623762,
      "grad_norm": 1.5806872844696045,
      "learning_rate": 5.995000000000001e-06,
      "loss": 0.4746,
      "step": 1200
    },
    {
      "epoch": 24.752475247524753,
      "grad_norm": 2.288752794265747,
      "learning_rate": 6.245000000000001e-06,
      "loss": 0.4676,
      "step": 1250
    },
    {
      "epoch": 25.742574257425744,
      "grad_norm": 1.542547583580017,
      "learning_rate": 6.4950000000000005e-06,
      "loss": 0.4773,
      "step": 1300
    },
    {
      "epoch": 26.73267326732673,
      "grad_norm": 1.0209100246429443,
      "learning_rate": 6.745000000000001e-06,
      "loss": 0.4683,
      "step": 1350
    },
    {
      "epoch": 27.722772277227723,
      "grad_norm": 1.2771400213241577,
      "learning_rate": 6.995000000000001e-06,
      "loss": 0.4577,
      "step": 1400
    },
    {
      "epoch": 28.712871287128714,
      "grad_norm": 1.348565936088562,
      "learning_rate": 7.245000000000001e-06,
      "loss": 0.4563,
      "step": 1450
    },
    {
      "epoch": 29.702970297029704,
      "grad_norm": 1.5620756149291992,
      "learning_rate": 7.495000000000001e-06,
      "loss": 0.4604,
      "step": 1500
    },
    {
      "epoch": 29.702970297029704,
      "eval_loss": 0.41570472717285156,
      "eval_runtime": 7.3746,
      "eval_samples_per_second": 24.408,
      "eval_steps_per_second": 3.119,
      "step": 1500
    },
    {
      "epoch": 30.693069306930692,
      "grad_norm": 1.3051259517669678,
      "learning_rate": 7.745e-06,
      "loss": 0.4551,
      "step": 1550
    },
    {
      "epoch": 31.683168316831683,
      "grad_norm": 1.6832257509231567,
      "learning_rate": 7.995e-06,
      "loss": 0.4517,
      "step": 1600
    },
    {
      "epoch": 32.67326732673267,
      "grad_norm": 2.1702144145965576,
      "learning_rate": 8.245000000000002e-06,
      "loss": 0.4546,
      "step": 1650
    },
    {
      "epoch": 33.663366336633665,
      "grad_norm": 1.5217938423156738,
      "learning_rate": 8.495e-06,
      "loss": 0.4549,
      "step": 1700
    },
    {
      "epoch": 34.65346534653465,
      "grad_norm": 1.8007372617721558,
      "learning_rate": 8.745000000000002e-06,
      "loss": 0.4453,
      "step": 1750
    },
    {
      "epoch": 35.64356435643565,
      "grad_norm": 1.490868091583252,
      "learning_rate": 8.995000000000001e-06,
      "loss": 0.4522,
      "step": 1800
    },
    {
      "epoch": 36.633663366336634,
      "grad_norm": 1.3354467153549194,
      "learning_rate": 9.245e-06,
      "loss": 0.4533,
      "step": 1850
    },
    {
      "epoch": 37.62376237623762,
      "grad_norm": 1.661230206489563,
      "learning_rate": 9.495000000000001e-06,
      "loss": 0.4467,
      "step": 1900
    },
    {
      "epoch": 38.613861386138616,
      "grad_norm": 1.4943360090255737,
      "learning_rate": 9.745e-06,
      "loss": 0.4451,
      "step": 1950
    },
    {
      "epoch": 39.603960396039604,
      "grad_norm": 2.01084566116333,
      "learning_rate": 9.995000000000002e-06,
      "loss": 0.4373,
      "step": 2000
    },
    {
      "epoch": 39.603960396039604,
      "eval_loss": 0.40344154834747314,
      "eval_runtime": 8.1054,
      "eval_samples_per_second": 22.207,
      "eval_steps_per_second": 2.838,
      "step": 2000
    },
    {
      "epoch": 40.59405940594059,
      "grad_norm": 2.1528162956237793,
      "learning_rate": 9.918333333333335e-06,
      "loss": 0.4446,
      "step": 2050
    },
    {
      "epoch": 41.584158415841586,
      "grad_norm": 1.692417025566101,
      "learning_rate": 9.835000000000002e-06,
      "loss": 0.4363,
      "step": 2100
    },
    {
      "epoch": 42.57425742574257,
      "grad_norm": 1.190176248550415,
      "learning_rate": 9.751666666666667e-06,
      "loss": 0.4425,
      "step": 2150
    },
    {
      "epoch": 43.56435643564357,
      "grad_norm": 1.027122974395752,
      "learning_rate": 9.668333333333334e-06,
      "loss": 0.4318,
      "step": 2200
    },
    {
      "epoch": 44.554455445544555,
      "grad_norm": 1.3497254848480225,
      "learning_rate": 9.585e-06,
      "loss": 0.4377,
      "step": 2250
    },
    {
      "epoch": 45.54455445544554,
      "grad_norm": 1.483748197555542,
      "learning_rate": 9.501666666666667e-06,
      "loss": 0.4358,
      "step": 2300
    },
    {
      "epoch": 46.53465346534654,
      "grad_norm": 1.2555333375930786,
      "learning_rate": 9.418333333333334e-06,
      "loss": 0.4397,
      "step": 2350
    },
    {
      "epoch": 47.524752475247524,
      "grad_norm": 1.345694899559021,
      "learning_rate": 9.335000000000001e-06,
      "loss": 0.4341,
      "step": 2400
    },
    {
      "epoch": 48.51485148514851,
      "grad_norm": 1.3507237434387207,
      "learning_rate": 9.251666666666668e-06,
      "loss": 0.4363,
      "step": 2450
    },
    {
      "epoch": 49.504950495049506,
      "grad_norm": 2.3417470455169678,
      "learning_rate": 9.168333333333333e-06,
      "loss": 0.4359,
      "step": 2500
    },
    {
      "epoch": 49.504950495049506,
      "eval_loss": 0.40056198835372925,
      "eval_runtime": 7.9369,
      "eval_samples_per_second": 22.679,
      "eval_steps_per_second": 2.898,
      "step": 2500
    },
    {
      "epoch": 50.495049504950494,
      "grad_norm": 1.1603953838348389,
      "learning_rate": 9.085e-06,
      "loss": 0.4316,
      "step": 2550
    },
    {
      "epoch": 51.48514851485149,
      "grad_norm": 1.2483686208724976,
      "learning_rate": 9.001666666666667e-06,
      "loss": 0.4264,
      "step": 2600
    },
    {
      "epoch": 52.475247524752476,
      "grad_norm": 1.3609150648117065,
      "learning_rate": 8.918333333333334e-06,
      "loss": 0.4375,
      "step": 2650
    },
    {
      "epoch": 53.46534653465346,
      "grad_norm": 1.287948727607727,
      "learning_rate": 8.835000000000001e-06,
      "loss": 0.4296,
      "step": 2700
    },
    {
      "epoch": 54.45544554455446,
      "grad_norm": 1.1371320486068726,
      "learning_rate": 8.751666666666668e-06,
      "loss": 0.4293,
      "step": 2750
    },
    {
      "epoch": 55.445544554455445,
      "grad_norm": 1.2330107688903809,
      "learning_rate": 8.668333333333335e-06,
      "loss": 0.4211,
      "step": 2800
    },
    {
      "epoch": 56.43564356435643,
      "grad_norm": 1.494728446006775,
      "learning_rate": 8.585000000000002e-06,
      "loss": 0.4265,
      "step": 2850
    },
    {
      "epoch": 57.42574257425743,
      "grad_norm": 1.195456862449646,
      "learning_rate": 8.501666666666667e-06,
      "loss": 0.4255,
      "step": 2900
    },
    {
      "epoch": 58.415841584158414,
      "grad_norm": 2.2788732051849365,
      "learning_rate": 8.418333333333334e-06,
      "loss": 0.4262,
      "step": 2950
    },
    {
      "epoch": 59.40594059405941,
      "grad_norm": 1.3329004049301147,
      "learning_rate": 8.335e-06,
      "loss": 0.4236,
      "step": 3000
    },
    {
      "epoch": 59.40594059405941,
      "eval_loss": 0.39749667048454285,
      "eval_runtime": 7.652,
      "eval_samples_per_second": 23.523,
      "eval_steps_per_second": 3.006,
      "step": 3000
    },
    {
      "epoch": 60.396039603960396,
      "grad_norm": 1.0736905336380005,
      "learning_rate": 8.251666666666668e-06,
      "loss": 0.4324,
      "step": 3050
    },
    {
      "epoch": 61.386138613861384,
      "grad_norm": 1.4117296934127808,
      "learning_rate": 8.168333333333334e-06,
      "loss": 0.421,
      "step": 3100
    },
    {
      "epoch": 62.37623762376238,
      "grad_norm": 1.2299913167953491,
      "learning_rate": 8.085000000000001e-06,
      "loss": 0.426,
      "step": 3150
    },
    {
      "epoch": 63.366336633663366,
      "grad_norm": 1.780414342880249,
      "learning_rate": 8.001666666666668e-06,
      "loss": 0.4217,
      "step": 3200
    },
    {
      "epoch": 64.35643564356435,
      "grad_norm": 0.953036367893219,
      "learning_rate": 7.918333333333333e-06,
      "loss": 0.4275,
      "step": 3250
    },
    {
      "epoch": 65.34653465346534,
      "grad_norm": 1.4100078344345093,
      "learning_rate": 7.835e-06,
      "loss": 0.4207,
      "step": 3300
    },
    {
      "epoch": 66.33663366336634,
      "grad_norm": 1.2050296068191528,
      "learning_rate": 7.751666666666667e-06,
      "loss": 0.4194,
      "step": 3350
    },
    {
      "epoch": 67.32673267326733,
      "grad_norm": 1.2712194919586182,
      "learning_rate": 7.668333333333334e-06,
      "loss": 0.4185,
      "step": 3400
    },
    {
      "epoch": 68.31683168316832,
      "grad_norm": 1.4407145977020264,
      "learning_rate": 7.585e-06,
      "loss": 0.42,
      "step": 3450
    },
    {
      "epoch": 69.3069306930693,
      "grad_norm": 0.9915832281112671,
      "learning_rate": 7.501666666666667e-06,
      "loss": 0.4196,
      "step": 3500
    },
    {
      "epoch": 69.3069306930693,
      "eval_loss": 0.3955594301223755,
      "eval_runtime": 8.0095,
      "eval_samples_per_second": 22.473,
      "eval_steps_per_second": 2.872,
      "step": 3500
    },
    {
      "epoch": 70.29702970297029,
      "grad_norm": 1.399967074394226,
      "learning_rate": 7.418333333333334e-06,
      "loss": 0.413,
      "step": 3550
    },
    {
      "epoch": 71.2871287128713,
      "grad_norm": 1.4354674816131592,
      "learning_rate": 7.335000000000001e-06,
      "loss": 0.4166,
      "step": 3600
    },
    {
      "epoch": 72.27722772277228,
      "grad_norm": 1.5903120040893555,
      "learning_rate": 7.251666666666667e-06,
      "loss": 0.4206,
      "step": 3650
    },
    {
      "epoch": 73.26732673267327,
      "grad_norm": 1.184046983718872,
      "learning_rate": 7.168333333333334e-06,
      "loss": 0.4179,
      "step": 3700
    },
    {
      "epoch": 74.25742574257426,
      "grad_norm": 1.1453702449798584,
      "learning_rate": 7.085000000000001e-06,
      "loss": 0.4145,
      "step": 3750
    },
    {
      "epoch": 75.24752475247524,
      "grad_norm": 2.3300318717956543,
      "learning_rate": 7.001666666666668e-06,
      "loss": 0.4255,
      "step": 3800
    },
    {
      "epoch": 76.23762376237623,
      "grad_norm": 1.671020746231079,
      "learning_rate": 6.918333333333334e-06,
      "loss": 0.4213,
      "step": 3850
    },
    {
      "epoch": 77.22772277227723,
      "grad_norm": 1.1744346618652344,
      "learning_rate": 6.835000000000001e-06,
      "loss": 0.4181,
      "step": 3900
    },
    {
      "epoch": 78.21782178217822,
      "grad_norm": 1.3221676349639893,
      "learning_rate": 6.7516666666666675e-06,
      "loss": 0.4163,
      "step": 3950
    },
    {
      "epoch": 79.20792079207921,
      "grad_norm": 1.0609053373336792,
      "learning_rate": 6.668333333333334e-06,
      "loss": 0.4183,
      "step": 4000
    },
    {
      "epoch": 79.20792079207921,
      "eval_loss": 0.39377516508102417,
      "eval_runtime": 7.9303,
      "eval_samples_per_second": 22.698,
      "eval_steps_per_second": 2.9,
      "step": 4000
    },
    {
      "epoch": 80.1980198019802,
      "grad_norm": 1.0227816104888916,
      "learning_rate": 6.5850000000000005e-06,
      "loss": 0.4164,
      "step": 4050
    },
    {
      "epoch": 81.18811881188118,
      "grad_norm": 1.0951721668243408,
      "learning_rate": 6.501666666666667e-06,
      "loss": 0.4093,
      "step": 4100
    },
    {
      "epoch": 82.17821782178218,
      "grad_norm": 1.0732530355453491,
      "learning_rate": 6.418333333333334e-06,
      "loss": 0.4161,
      "step": 4150
    },
    {
      "epoch": 83.16831683168317,
      "grad_norm": 1.13483464717865,
      "learning_rate": 6.336666666666667e-06,
      "loss": 0.4113,
      "step": 4200
    },
    {
      "epoch": 84.15841584158416,
      "grad_norm": 0.9911147356033325,
      "learning_rate": 6.253333333333333e-06,
      "loss": 0.4123,
      "step": 4250
    },
    {
      "epoch": 85.14851485148515,
      "grad_norm": 1.4925992488861084,
      "learning_rate": 6.17e-06,
      "loss": 0.4122,
      "step": 4300
    },
    {
      "epoch": 86.13861386138613,
      "grad_norm": 1.4414916038513184,
      "learning_rate": 6.086666666666667e-06,
      "loss": 0.4103,
      "step": 4350
    },
    {
      "epoch": 87.12871287128714,
      "grad_norm": 1.1858835220336914,
      "learning_rate": 6.003333333333334e-06,
      "loss": 0.4137,
      "step": 4400
    },
    {
      "epoch": 88.11881188118812,
      "grad_norm": 1.104232907295227,
      "learning_rate": 5.92e-06,
      "loss": 0.4096,
      "step": 4450
    },
    {
      "epoch": 89.10891089108911,
      "grad_norm": 1.7566862106323242,
      "learning_rate": 5.836666666666667e-06,
      "loss": 0.4148,
      "step": 4500
    },
    {
      "epoch": 89.10891089108911,
      "eval_loss": 0.3940875828266144,
      "eval_runtime": 7.934,
      "eval_samples_per_second": 22.687,
      "eval_steps_per_second": 2.899,
      "step": 4500
    },
    {
      "epoch": 90.0990099009901,
      "grad_norm": 1.0610746145248413,
      "learning_rate": 5.753333333333334e-06,
      "loss": 0.414,
      "step": 4550
    },
    {
      "epoch": 91.08910891089108,
      "grad_norm": 1.1602790355682373,
      "learning_rate": 5.67e-06,
      "loss": 0.415,
      "step": 4600
    },
    {
      "epoch": 92.07920792079207,
      "grad_norm": 1.1488662958145142,
      "learning_rate": 5.586666666666667e-06,
      "loss": 0.4162,
      "step": 4650
    },
    {
      "epoch": 93.06930693069307,
      "grad_norm": 1.279823899269104,
      "learning_rate": 5.503333333333334e-06,
      "loss": 0.4086,
      "step": 4700
    },
    {
      "epoch": 94.05940594059406,
      "grad_norm": 1.4046456813812256,
      "learning_rate": 5.420000000000001e-06,
      "loss": 0.4094,
      "step": 4750
    },
    {
      "epoch": 95.04950495049505,
      "grad_norm": 1.0302339792251587,
      "learning_rate": 5.336666666666667e-06,
      "loss": 0.4126,
      "step": 4800
    },
    {
      "epoch": 96.03960396039604,
      "grad_norm": 1.2423168420791626,
      "learning_rate": 5.2533333333333336e-06,
      "loss": 0.4115,
      "step": 4850
    },
    {
      "epoch": 97.02970297029702,
      "grad_norm": 1.0046170949935913,
      "learning_rate": 5.1700000000000005e-06,
      "loss": 0.4027,
      "step": 4900
    },
    {
      "epoch": 98.01980198019803,
      "grad_norm": 1.3599224090576172,
      "learning_rate": 5.086666666666667e-06,
      "loss": 0.4124,
      "step": 4950
    },
    {
      "epoch": 99.00990099009901,
      "grad_norm": 1.408510446548462,
      "learning_rate": 5.0033333333333334e-06,
      "loss": 0.4034,
      "step": 5000
    },
    {
      "epoch": 99.00990099009901,
      "eval_loss": 0.3930128216743469,
      "eval_runtime": 7.3645,
      "eval_samples_per_second": 24.442,
      "eval_steps_per_second": 3.123,
      "step": 5000
    },
    {
      "epoch": 100.0,
      "grad_norm": 1.2026501893997192,
      "learning_rate": 4.92e-06,
      "loss": 0.4097,
      "step": 5050
    },
    {
      "epoch": 100.99009900990099,
      "grad_norm": 1.1910372972488403,
      "learning_rate": 4.836666666666667e-06,
      "loss": 0.4126,
      "step": 5100
    },
    {
      "epoch": 101.98019801980197,
      "grad_norm": 1.084506630897522,
      "learning_rate": 4.753333333333333e-06,
      "loss": 0.4075,
      "step": 5150
    },
    {
      "epoch": 102.97029702970298,
      "grad_norm": 1.115693211555481,
      "learning_rate": 4.670000000000001e-06,
      "loss": 0.4069,
      "step": 5200
    },
    {
      "epoch": 103.96039603960396,
      "grad_norm": 1.1751248836517334,
      "learning_rate": 4.586666666666667e-06,
      "loss": 0.4098,
      "step": 5250
    },
    {
      "epoch": 104.95049504950495,
      "grad_norm": 1.0256706476211548,
      "learning_rate": 4.503333333333333e-06,
      "loss": 0.4044,
      "step": 5300
    },
    {
      "epoch": 105.94059405940594,
      "grad_norm": 1.200976848602295,
      "learning_rate": 4.42e-06,
      "loss": 0.4064,
      "step": 5350
    },
    {
      "epoch": 106.93069306930693,
      "grad_norm": 1.0933268070220947,
      "learning_rate": 4.336666666666667e-06,
      "loss": 0.4073,
      "step": 5400
    },
    {
      "epoch": 107.92079207920793,
      "grad_norm": 1.3021538257598877,
      "learning_rate": 4.253333333333334e-06,
      "loss": 0.4118,
      "step": 5450
    },
    {
      "epoch": 108.91089108910892,
      "grad_norm": 1.7753965854644775,
      "learning_rate": 4.17e-06,
      "loss": 0.4137,
      "step": 5500
    },
    {
      "epoch": 108.91089108910892,
      "eval_loss": 0.3955301344394684,
      "eval_runtime": 7.5702,
      "eval_samples_per_second": 23.777,
      "eval_steps_per_second": 3.038,
      "step": 5500
    },
    {
      "epoch": 109.9009900990099,
      "grad_norm": 1.4470231533050537,
      "learning_rate": 4.086666666666667e-06,
      "loss": 0.4053,
      "step": 5550
    },
    {
      "epoch": 110.89108910891089,
      "grad_norm": 1.7531636953353882,
      "learning_rate": 4.003333333333334e-06,
      "loss": 0.4061,
      "step": 5600
    },
    {
      "epoch": 111.88118811881188,
      "grad_norm": 0.9751603603363037,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.4086,
      "step": 5650
    },
    {
      "epoch": 112.87128712871286,
      "grad_norm": 1.1000280380249023,
      "learning_rate": 3.836666666666667e-06,
      "loss": 0.4031,
      "step": 5700
    },
    {
      "epoch": 113.86138613861387,
      "grad_norm": 1.2386143207550049,
      "learning_rate": 3.753333333333334e-06,
      "loss": 0.4059,
      "step": 5750
    },
    {
      "epoch": 114.85148514851485,
      "grad_norm": 1.4081382751464844,
      "learning_rate": 3.6700000000000004e-06,
      "loss": 0.4084,
      "step": 5800
    },
    {
      "epoch": 115.84158415841584,
      "grad_norm": 1.3724069595336914,
      "learning_rate": 3.5866666666666673e-06,
      "loss": 0.4046,
      "step": 5850
    },
    {
      "epoch": 116.83168316831683,
      "grad_norm": 1.188818097114563,
      "learning_rate": 3.5033333333333334e-06,
      "loss": 0.4109,
      "step": 5900
    },
    {
      "epoch": 117.82178217821782,
      "grad_norm": 1.3779815435409546,
      "learning_rate": 3.4200000000000007e-06,
      "loss": 0.4042,
      "step": 5950
    },
    {
      "epoch": 118.81188118811882,
      "grad_norm": 1.3800166845321655,
      "learning_rate": 3.3366666666666668e-06,
      "loss": 0.4094,
      "step": 6000
    },
    {
      "epoch": 118.81188118811882,
      "eval_loss": 0.39243730902671814,
      "eval_runtime": 7.911,
      "eval_samples_per_second": 22.753,
      "eval_steps_per_second": 2.907,
      "step": 6000
    },
    {
      "epoch": 119.8019801980198,
      "grad_norm": 1.2765828371047974,
      "learning_rate": 3.2533333333333332e-06,
      "loss": 0.4022,
      "step": 6050
    },
    {
      "epoch": 120.79207920792079,
      "grad_norm": 1.3544330596923828,
      "learning_rate": 3.17e-06,
      "loss": 0.4062,
      "step": 6100
    },
    {
      "epoch": 121.78217821782178,
      "grad_norm": 1.9719560146331787,
      "learning_rate": 3.0866666666666666e-06,
      "loss": 0.4002,
      "step": 6150
    },
    {
      "epoch": 122.77227722772277,
      "grad_norm": 1.2542638778686523,
      "learning_rate": 3.0033333333333335e-06,
      "loss": 0.4037,
      "step": 6200
    },
    {
      "epoch": 123.76237623762377,
      "grad_norm": 0.9950863122940063,
      "learning_rate": 2.92e-06,
      "loss": 0.4034,
      "step": 6250
    },
    {
      "epoch": 124.75247524752476,
      "grad_norm": 1.2220611572265625,
      "learning_rate": 2.836666666666667e-06,
      "loss": 0.405,
      "step": 6300
    },
    {
      "epoch": 125.74257425742574,
      "grad_norm": 1.3050748109817505,
      "learning_rate": 2.7533333333333334e-06,
      "loss": 0.4019,
      "step": 6350
    },
    {
      "epoch": 126.73267326732673,
      "grad_norm": 1.2218639850616455,
      "learning_rate": 2.6700000000000003e-06,
      "loss": 0.4077,
      "step": 6400
    },
    {
      "epoch": 127.72277227722772,
      "grad_norm": 1.0454591512680054,
      "learning_rate": 2.5866666666666667e-06,
      "loss": 0.4009,
      "step": 6450
    },
    {
      "epoch": 128.7128712871287,
      "grad_norm": 1.211366057395935,
      "learning_rate": 2.505e-06,
      "loss": 0.4112,
      "step": 6500
    },
    {
      "epoch": 128.7128712871287,
      "eval_loss": 0.3917270302772522,
      "eval_runtime": 7.8896,
      "eval_samples_per_second": 22.815,
      "eval_steps_per_second": 2.915,
      "step": 6500
    },
    {
      "epoch": 129.7029702970297,
      "grad_norm": 1.1961348056793213,
      "learning_rate": 2.421666666666667e-06,
      "loss": 0.4009,
      "step": 6550
    },
    {
      "epoch": 130.69306930693068,
      "grad_norm": 1.6463865041732788,
      "learning_rate": 2.3383333333333335e-06,
      "loss": 0.403,
      "step": 6600
    },
    {
      "epoch": 131.68316831683168,
      "grad_norm": 1.1030319929122925,
      "learning_rate": 2.2550000000000004e-06,
      "loss": 0.4009,
      "step": 6650
    },
    {
      "epoch": 132.67326732673268,
      "grad_norm": 1.0525782108306885,
      "learning_rate": 2.171666666666667e-06,
      "loss": 0.4038,
      "step": 6700
    },
    {
      "epoch": 133.66336633663366,
      "grad_norm": 1.0361270904541016,
      "learning_rate": 2.088333333333334e-06,
      "loss": 0.4046,
      "step": 6750
    },
    {
      "epoch": 134.65346534653466,
      "grad_norm": 1.3003863096237183,
      "learning_rate": 2.0050000000000003e-06,
      "loss": 0.404,
      "step": 6800
    },
    {
      "epoch": 135.64356435643563,
      "grad_norm": 1.6495325565338135,
      "learning_rate": 1.9216666666666668e-06,
      "loss": 0.401,
      "step": 6850
    },
    {
      "epoch": 136.63366336633663,
      "grad_norm": 1.070357084274292,
      "learning_rate": 1.8383333333333334e-06,
      "loss": 0.4033,
      "step": 6900
    },
    {
      "epoch": 137.62376237623764,
      "grad_norm": 0.9179391860961914,
      "learning_rate": 1.7550000000000001e-06,
      "loss": 0.4018,
      "step": 6950
    },
    {
      "epoch": 138.6138613861386,
      "grad_norm": 1.398165225982666,
      "learning_rate": 1.6716666666666666e-06,
      "loss": 0.4041,
      "step": 7000
    },
    {
      "epoch": 138.6138613861386,
      "eval_loss": 0.39232540130615234,
      "eval_runtime": 7.5562,
      "eval_samples_per_second": 23.822,
      "eval_steps_per_second": 3.044,
      "step": 7000
    },
    {
      "epoch": 139.6039603960396,
      "grad_norm": 1.1762559413909912,
      "learning_rate": 1.5900000000000002e-06,
      "loss": 0.4117,
      "step": 7050
    },
    {
      "epoch": 140.59405940594058,
      "grad_norm": 1.515325665473938,
      "learning_rate": 1.506666666666667e-06,
      "loss": 0.4014,
      "step": 7100
    },
    {
      "epoch": 141.58415841584159,
      "grad_norm": 1.0724811553955078,
      "learning_rate": 1.4233333333333336e-06,
      "loss": 0.4028,
      "step": 7150
    },
    {
      "epoch": 142.5742574257426,
      "grad_norm": 1.0034635066986084,
      "learning_rate": 1.34e-06,
      "loss": 0.3977,
      "step": 7200
    },
    {
      "epoch": 143.56435643564356,
      "grad_norm": 1.1974977254867554,
      "learning_rate": 1.2566666666666668e-06,
      "loss": 0.4013,
      "step": 7250
    },
    {
      "epoch": 144.55445544554456,
      "grad_norm": 1.167351484298706,
      "learning_rate": 1.1733333333333335e-06,
      "loss": 0.3992,
      "step": 7300
    },
    {
      "epoch": 145.54455445544554,
      "grad_norm": 1.1566020250320435,
      "learning_rate": 1.0900000000000002e-06,
      "loss": 0.4,
      "step": 7350
    },
    {
      "epoch": 146.53465346534654,
      "grad_norm": 0.8720223307609558,
      "learning_rate": 1.0066666666666668e-06,
      "loss": 0.4006,
      "step": 7400
    },
    {
      "epoch": 147.52475247524754,
      "grad_norm": 1.1982648372650146,
      "learning_rate": 9.233333333333334e-07,
      "loss": 0.3999,
      "step": 7450
    },
    {
      "epoch": 148.5148514851485,
      "grad_norm": 1.1742349863052368,
      "learning_rate": 8.400000000000001e-07,
      "loss": 0.3989,
      "step": 7500
    },
    {
      "epoch": 148.5148514851485,
      "eval_loss": 0.3927270770072937,
      "eval_runtime": 8.9179,
      "eval_samples_per_second": 20.184,
      "eval_steps_per_second": 2.579,
      "step": 7500
    },
    {
      "epoch": 149.5049504950495,
      "grad_norm": 1.1068042516708374,
      "learning_rate": 7.566666666666667e-07,
      "loss": 0.4029,
      "step": 7550
    },
    {
      "epoch": 150.4950495049505,
      "grad_norm": 1.2785730361938477,
      "learning_rate": 6.733333333333334e-07,
      "loss": 0.4001,
      "step": 7600
    },
    {
      "epoch": 151.4851485148515,
      "grad_norm": 1.0063281059265137,
      "learning_rate": 5.900000000000001e-07,
      "loss": 0.3977,
      "step": 7650
    },
    {
      "epoch": 152.47524752475246,
      "grad_norm": 1.1545789241790771,
      "learning_rate": 5.066666666666667e-07,
      "loss": 0.4015,
      "step": 7700
    },
    {
      "epoch": 153.46534653465346,
      "grad_norm": 1.0624966621398926,
      "learning_rate": 4.233333333333334e-07,
      "loss": 0.4044,
      "step": 7750
    },
    {
      "epoch": 154.45544554455446,
      "grad_norm": 1.3898247480392456,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 0.4005,
      "step": 7800
    },
    {
      "epoch": 155.44554455445544,
      "grad_norm": 1.2618342638015747,
      "learning_rate": 2.566666666666667e-07,
      "loss": 0.4025,
      "step": 7850
    },
    {
      "epoch": 156.43564356435644,
      "grad_norm": 1.0798155069351196,
      "learning_rate": 1.7333333333333335e-07,
      "loss": 0.4015,
      "step": 7900
    },
    {
      "epoch": 157.4257425742574,
      "grad_norm": 1.0550168752670288,
      "learning_rate": 9e-08,
      "loss": 0.3952,
      "step": 7950
    },
    {
      "epoch": 158.41584158415841,
      "grad_norm": 1.0714497566223145,
      "learning_rate": 6.666666666666667e-09,
      "loss": 0.3989,
      "step": 8000
    },
    {
      "epoch": 158.41584158415841,
      "eval_loss": 0.39282524585723877,
      "eval_runtime": 8.2483,
      "eval_samples_per_second": 21.823,
      "eval_steps_per_second": 2.788,
      "step": 8000
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 160,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.322114324975938e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}