{
  "best_metric": 0.4066200256347656,
  "best_model_checkpoint": "mikhail_panzo/fil_b32_le4_s8000/checkpoint-5500",
  "epoch": 175.82417582417582,
  "eval_steps": 500,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.098901098901099,
      "grad_norm": 1.5917891263961792,
      "learning_rate": 2.5e-06,
      "loss": 0.7987,
      "step": 50
    },
    {
      "epoch": 2.197802197802198,
      "grad_norm": 2.48260235786438,
      "learning_rate": 4.950000000000001e-06,
      "loss": 0.7263,
      "step": 100
    },
    {
      "epoch": 3.2967032967032965,
      "grad_norm": 4.328954219818115,
      "learning_rate": 7.45e-06,
      "loss": 0.6646,
      "step": 150
    },
    {
      "epoch": 4.395604395604396,
      "grad_norm": 2.710536241531372,
      "learning_rate": 9.950000000000001e-06,
      "loss": 0.6154,
      "step": 200
    },
    {
      "epoch": 5.4945054945054945,
      "grad_norm": 2.7451584339141846,
      "learning_rate": 1.2450000000000001e-05,
      "loss": 0.5546,
      "step": 250
    },
    {
      "epoch": 6.593406593406593,
      "grad_norm": 2.8580620288848877,
      "learning_rate": 1.4950000000000001e-05,
      "loss": 0.529,
      "step": 300
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 2.8751707077026367,
      "learning_rate": 1.745e-05,
      "loss": 0.5265,
      "step": 350
    },
    {
      "epoch": 8.791208791208792,
      "grad_norm": 1.5821752548217773,
      "learning_rate": 1.995e-05,
      "loss": 0.4996,
      "step": 400
    },
    {
      "epoch": 9.89010989010989,
      "grad_norm": 1.9110968112945557,
      "learning_rate": 2.245e-05,
      "loss": 0.503,
      "step": 450
    },
    {
      "epoch": 10.989010989010989,
      "grad_norm": 3.7658803462982178,
      "learning_rate": 2.495e-05,
      "loss": 0.498,
      "step": 500
    },
    {
      "epoch": 10.989010989010989,
      "eval_loss": 0.4393291473388672,
      "eval_runtime": 12.9638,
      "eval_samples_per_second": 12.419,
      "eval_steps_per_second": 1.62,
      "step": 500
    },
    {
      "epoch": 12.087912087912088,
      "grad_norm": 2.30264949798584,
      "learning_rate": 2.7450000000000003e-05,
      "loss": 0.4936,
      "step": 550
    },
    {
      "epoch": 13.186813186813186,
      "grad_norm": 3.3187615871429443,
      "learning_rate": 2.995e-05,
      "loss": 0.4753,
      "step": 600
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 3.1503515243530273,
      "learning_rate": 3.245e-05,
      "loss": 0.4722,
      "step": 650
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 2.1364340782165527,
      "learning_rate": 3.495e-05,
      "loss": 0.4666,
      "step": 700
    },
    {
      "epoch": 16.483516483516482,
      "grad_norm": 1.972012996673584,
      "learning_rate": 3.745e-05,
      "loss": 0.462,
      "step": 750
    },
    {
      "epoch": 17.582417582417584,
      "grad_norm": 1.9032899141311646,
      "learning_rate": 3.995e-05,
      "loss": 0.4681,
      "step": 800
    },
    {
      "epoch": 18.681318681318682,
      "grad_norm": 1.9877583980560303,
      "learning_rate": 4.245e-05,
      "loss": 0.4647,
      "step": 850
    },
    {
      "epoch": 19.78021978021978,
      "grad_norm": 1.8145549297332764,
      "learning_rate": 4.495e-05,
      "loss": 0.4613,
      "step": 900
    },
    {
      "epoch": 20.87912087912088,
      "grad_norm": 2.619379997253418,
      "learning_rate": 4.745e-05,
      "loss": 0.4557,
      "step": 950
    },
    {
      "epoch": 21.978021978021978,
      "grad_norm": 4.430926322937012,
      "learning_rate": 4.995e-05,
      "loss": 0.448,
      "step": 1000
    },
    {
      "epoch": 21.978021978021978,
      "eval_loss": 0.4195362627506256,
      "eval_runtime": 12.2513,
      "eval_samples_per_second": 13.141,
      "eval_steps_per_second": 1.714,
      "step": 1000
    },
    {
      "epoch": 23.076923076923077,
      "grad_norm": 4.793135166168213,
      "learning_rate": 5.245e-05,
      "loss": 0.4498,
      "step": 1050
    },
    {
      "epoch": 24.175824175824175,
      "grad_norm": 2.612823009490967,
      "learning_rate": 5.495e-05,
      "loss": 0.443,
      "step": 1100
    },
    {
      "epoch": 25.274725274725274,
      "grad_norm": 3.3249428272247314,
      "learning_rate": 5.745e-05,
      "loss": 0.4456,
      "step": 1150
    },
    {
      "epoch": 26.373626373626372,
      "grad_norm": 3.8821775913238525,
      "learning_rate": 5.995000000000001e-05,
      "loss": 0.4586,
      "step": 1200
    },
    {
      "epoch": 27.47252747252747,
      "grad_norm": 6.9926252365112305,
      "learning_rate": 6.245000000000001e-05,
      "loss": 0.4479,
      "step": 1250
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 7.615652084350586,
      "learning_rate": 6.494999999999999e-05,
      "loss": 0.4446,
      "step": 1300
    },
    {
      "epoch": 29.67032967032967,
      "grad_norm": 1.9948512315750122,
      "learning_rate": 6.745e-05,
      "loss": 0.4547,
      "step": 1350
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 1.4590964317321777,
      "learning_rate": 6.995e-05,
      "loss": 0.4391,
      "step": 1400
    },
    {
      "epoch": 31.86813186813187,
      "grad_norm": 2.5002665519714355,
      "learning_rate": 7.245000000000001e-05,
      "loss": 0.4418,
      "step": 1450
    },
    {
      "epoch": 32.967032967032964,
      "grad_norm": 3.9836597442626953,
      "learning_rate": 7.495e-05,
      "loss": 0.4411,
      "step": 1500
    },
    {
      "epoch": 32.967032967032964,
      "eval_loss": 0.4205290973186493,
      "eval_runtime": 12.102,
      "eval_samples_per_second": 13.304,
      "eval_steps_per_second": 1.735,
      "step": 1500
    },
    {
      "epoch": 34.065934065934066,
      "grad_norm": 4.376543045043945,
      "learning_rate": 7.745e-05,
      "loss": 0.44,
      "step": 1550
    },
    {
      "epoch": 35.16483516483517,
      "grad_norm": 5.079801559448242,
      "learning_rate": 7.995e-05,
      "loss": 0.438,
      "step": 1600
    },
    {
      "epoch": 36.26373626373626,
      "grad_norm": 1.6529037952423096,
      "learning_rate": 8.245e-05,
      "loss": 0.4308,
      "step": 1650
    },
    {
      "epoch": 37.362637362637365,
      "grad_norm": 3.664045572280884,
      "learning_rate": 8.495e-05,
      "loss": 0.4397,
      "step": 1700
    },
    {
      "epoch": 38.46153846153846,
      "grad_norm": 2.1333911418914795,
      "learning_rate": 8.745000000000001e-05,
      "loss": 0.4441,
      "step": 1750
    },
    {
      "epoch": 39.56043956043956,
      "grad_norm": 3.5315911769866943,
      "learning_rate": 8.995e-05,
      "loss": 0.4325,
      "step": 1800
    },
    {
      "epoch": 40.65934065934066,
      "grad_norm": 7.4896674156188965,
      "learning_rate": 9.245e-05,
      "loss": 0.4299,
      "step": 1850
    },
    {
      "epoch": 41.75824175824176,
      "grad_norm": 3.11539363861084,
      "learning_rate": 9.495e-05,
      "loss": 0.4384,
      "step": 1900
    },
    {
      "epoch": 42.857142857142854,
      "grad_norm": 1.5338855981826782,
      "learning_rate": 9.745000000000001e-05,
      "loss": 0.4259,
      "step": 1950
    },
    {
      "epoch": 43.956043956043956,
      "grad_norm": 4.518911361694336,
      "learning_rate": 9.995e-05,
      "loss": 0.4347,
      "step": 2000
    },
    {
      "epoch": 43.956043956043956,
      "eval_loss": 0.4253252148628235,
      "eval_runtime": 11.7613,
      "eval_samples_per_second": 13.689,
      "eval_steps_per_second": 1.786,
      "step": 2000
    },
    {
      "epoch": 45.05494505494506,
      "grad_norm": 3.7814793586730957,
      "learning_rate": 9.918333333333334e-05,
      "loss": 0.4287,
      "step": 2050
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 6.209993362426758,
      "learning_rate": 9.835e-05,
      "loss": 0.4285,
      "step": 2100
    },
    {
      "epoch": 47.252747252747255,
      "grad_norm": 3.44171142578125,
      "learning_rate": 9.751666666666666e-05,
      "loss": 0.4301,
      "step": 2150
    },
    {
      "epoch": 48.35164835164835,
      "grad_norm": 3.956023931503296,
      "learning_rate": 9.668333333333334e-05,
      "loss": 0.4257,
      "step": 2200
    },
    {
      "epoch": 49.45054945054945,
      "grad_norm": 2.9259955883026123,
      "learning_rate": 9.586666666666667e-05,
      "loss": 0.441,
      "step": 2250
    },
    {
      "epoch": 50.54945054945055,
      "grad_norm": 3.86029314994812,
      "learning_rate": 9.503333333333334e-05,
      "loss": 0.4308,
      "step": 2300
    },
    {
      "epoch": 51.64835164835165,
      "grad_norm": 3.556795358657837,
      "learning_rate": 9.42e-05,
      "loss": 0.4165,
      "step": 2350
    },
    {
      "epoch": 52.747252747252745,
      "grad_norm": 3.4808130264282227,
      "learning_rate": 9.336666666666667e-05,
      "loss": 0.4153,
      "step": 2400
    },
    {
      "epoch": 53.84615384615385,
      "grad_norm": 2.37046217918396,
      "learning_rate": 9.253333333333334e-05,
      "loss": 0.416,
      "step": 2450
    },
    {
      "epoch": 54.94505494505494,
      "grad_norm": 1.724293828010559,
      "learning_rate": 9.17e-05,
      "loss": 0.4173,
      "step": 2500
    },
    {
      "epoch": 54.94505494505494,
      "eval_loss": 0.41511598229408264,
      "eval_runtime": 12.1765,
      "eval_samples_per_second": 13.222,
      "eval_steps_per_second": 1.725,
      "step": 2500
    },
    {
      "epoch": 56.043956043956044,
      "grad_norm": 1.3938038349151611,
      "learning_rate": 9.086666666666666e-05,
      "loss": 0.4175,
      "step": 2550
    },
    {
      "epoch": 57.142857142857146,
      "grad_norm": 1.3193321228027344,
      "learning_rate": 9.003333333333333e-05,
      "loss": 0.4169,
      "step": 2600
    },
    {
      "epoch": 58.24175824175824,
      "grad_norm": 4.696841716766357,
      "learning_rate": 8.92e-05,
      "loss": 0.4091,
      "step": 2650
    },
    {
      "epoch": 59.34065934065934,
      "grad_norm": 2.6104297637939453,
      "learning_rate": 8.836666666666667e-05,
      "loss": 0.4115,
      "step": 2700
    },
    {
      "epoch": 60.43956043956044,
      "grad_norm": 2.989949941635132,
      "learning_rate": 8.753333333333334e-05,
      "loss": 0.4266,
      "step": 2750
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 8.316046714782715,
      "learning_rate": 8.67e-05,
      "loss": 0.4087,
      "step": 2800
    },
    {
      "epoch": 62.637362637362635,
      "grad_norm": 4.265425205230713,
      "learning_rate": 8.586666666666668e-05,
      "loss": 0.4063,
      "step": 2850
    },
    {
      "epoch": 63.73626373626374,
      "grad_norm": 3.9043214321136475,
      "learning_rate": 8.503333333333334e-05,
      "loss": 0.4018,
      "step": 2900
    },
    {
      "epoch": 64.83516483516483,
      "grad_norm": 2.512340545654297,
      "learning_rate": 8.42e-05,
      "loss": 0.4059,
      "step": 2950
    },
    {
      "epoch": 65.93406593406593,
      "grad_norm": 3.410762071609497,
      "learning_rate": 8.336666666666667e-05,
      "loss": 0.4012,
      "step": 3000
    },
    {
      "epoch": 65.93406593406593,
      "eval_loss": 0.411817729473114,
      "eval_runtime": 12.4908,
      "eval_samples_per_second": 12.889,
      "eval_steps_per_second": 1.681,
      "step": 3000
    },
    {
      "epoch": 67.03296703296704,
      "grad_norm": 1.7443562746047974,
      "learning_rate": 8.253333333333334e-05,
      "loss": 0.3996,
      "step": 3050
    },
    {
      "epoch": 68.13186813186813,
      "grad_norm": 7.796085357666016,
      "learning_rate": 8.17e-05,
      "loss": 0.4001,
      "step": 3100
    },
    {
      "epoch": 69.23076923076923,
      "grad_norm": 4.286561965942383,
      "learning_rate": 8.086666666666666e-05,
      "loss": 0.4028,
      "step": 3150
    },
    {
      "epoch": 70.32967032967034,
      "grad_norm": 1.7267292737960815,
      "learning_rate": 8.003333333333333e-05,
      "loss": 0.3981,
      "step": 3200
    },
    {
      "epoch": 71.42857142857143,
      "grad_norm": 1.720483422279358,
      "learning_rate": 7.920000000000001e-05,
      "loss": 0.4044,
      "step": 3250
    },
    {
      "epoch": 72.52747252747253,
      "grad_norm": 2.6275534629821777,
      "learning_rate": 7.836666666666667e-05,
      "loss": 0.3915,
      "step": 3300
    },
    {
      "epoch": 73.62637362637362,
      "grad_norm": 1.1752153635025024,
      "learning_rate": 7.753333333333334e-05,
      "loss": 0.4001,
      "step": 3350
    },
    {
      "epoch": 74.72527472527473,
      "grad_norm": 1.8559554815292358,
      "learning_rate": 7.670000000000001e-05,
      "loss": 0.3975,
      "step": 3400
    },
    {
      "epoch": 75.82417582417582,
      "grad_norm": 2.82486629486084,
      "learning_rate": 7.586666666666668e-05,
      "loss": 0.3986,
      "step": 3450
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 2.659985065460205,
      "learning_rate": 7.503333333333333e-05,
      "loss": 0.4023,
      "step": 3500
    },
    {
      "epoch": 76.92307692307692,
      "eval_loss": 0.40923717617988586,
      "eval_runtime": 12.2621,
      "eval_samples_per_second": 13.13,
      "eval_steps_per_second": 1.713,
      "step": 3500
    },
    {
      "epoch": 78.02197802197803,
      "grad_norm": 2.169241428375244,
      "learning_rate": 7.42e-05,
      "loss": 0.3924,
      "step": 3550
    },
    {
      "epoch": 79.12087912087912,
      "grad_norm": 1.7540392875671387,
      "learning_rate": 7.336666666666667e-05,
      "loss": 0.3892,
      "step": 3600
    },
    {
      "epoch": 80.21978021978022,
      "grad_norm": 1.3049631118774414,
      "learning_rate": 7.253333333333334e-05,
      "loss": 0.3969,
      "step": 3650
    },
    {
      "epoch": 81.31868131868131,
      "grad_norm": 1.0093408823013306,
      "learning_rate": 7.17e-05,
      "loss": 0.395,
      "step": 3700
    },
    {
      "epoch": 82.41758241758242,
      "grad_norm": 1.3580188751220703,
      "learning_rate": 7.086666666666666e-05,
      "loss": 0.3913,
      "step": 3750
    },
    {
      "epoch": 83.51648351648352,
      "grad_norm": 1.6554079055786133,
      "learning_rate": 7.003333333333335e-05,
      "loss": 0.3935,
      "step": 3800
    },
    {
      "epoch": 84.61538461538461,
      "grad_norm": 2.003830909729004,
      "learning_rate": 6.92e-05,
      "loss": 0.3981,
      "step": 3850
    },
    {
      "epoch": 85.71428571428571,
      "grad_norm": 3.338681697845459,
      "learning_rate": 6.836666666666667e-05,
      "loss": 0.3873,
      "step": 3900
    },
    {
      "epoch": 86.81318681318682,
      "grad_norm": 2.0752291679382324,
      "learning_rate": 6.753333333333334e-05,
      "loss": 0.388,
      "step": 3950
    },
    {
      "epoch": 87.91208791208791,
      "grad_norm": 1.809194564819336,
      "learning_rate": 6.670000000000001e-05,
      "loss": 0.3873,
      "step": 4000
    },
    {
      "epoch": 87.91208791208791,
      "eval_loss": 0.4116482436656952,
      "eval_runtime": 13.309,
      "eval_samples_per_second": 12.097,
      "eval_steps_per_second": 1.578,
      "step": 4000
    },
    {
      "epoch": 89.01098901098901,
      "grad_norm": 1.8667621612548828,
      "learning_rate": 6.586666666666666e-05,
      "loss": 0.3867,
      "step": 4050
    },
    {
      "epoch": 90.10989010989012,
      "grad_norm": 1.7244277000427246,
      "learning_rate": 6.503333333333333e-05,
      "loss": 0.3802,
      "step": 4100
    },
    {
      "epoch": 91.20879120879121,
      "grad_norm": 1.6144648790359497,
      "learning_rate": 6.42e-05,
      "loss": 0.3853,
      "step": 4150
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 1.299013614654541,
      "learning_rate": 6.336666666666667e-05,
      "loss": 0.3837,
      "step": 4200
    },
    {
      "epoch": 93.4065934065934,
      "grad_norm": 2.6695783138275146,
      "learning_rate": 6.253333333333333e-05,
      "loss": 0.3832,
      "step": 4250
    },
    {
      "epoch": 94.50549450549451,
      "grad_norm": 1.719193696975708,
      "learning_rate": 6.170000000000001e-05,
      "loss": 0.3832,
      "step": 4300
    },
    {
      "epoch": 95.6043956043956,
      "grad_norm": 2.300633668899536,
      "learning_rate": 6.086666666666667e-05,
      "loss": 0.381,
      "step": 4350
    },
    {
      "epoch": 96.7032967032967,
      "grad_norm": 1.0958266258239746,
      "learning_rate": 6.003333333333334e-05,
      "loss": 0.3834,
      "step": 4400
    },
    {
      "epoch": 97.8021978021978,
      "grad_norm": 1.2469346523284912,
      "learning_rate": 5.92e-05,
      "loss": 0.3746,
      "step": 4450
    },
    {
      "epoch": 98.9010989010989,
      "grad_norm": 2.032564878463745,
      "learning_rate": 5.836666666666667e-05,
      "loss": 0.381,
      "step": 4500
    },
    {
      "epoch": 98.9010989010989,
      "eval_loss": 0.4088541567325592,
      "eval_runtime": 12.5723,
      "eval_samples_per_second": 12.806,
      "eval_steps_per_second": 1.67,
      "step": 4500
    },
    {
      "epoch": 100.0,
      "grad_norm": 2.846341371536255,
      "learning_rate": 5.753333333333334e-05,
      "loss": 0.3763,
      "step": 4550
    },
    {
      "epoch": 101.0989010989011,
      "grad_norm": 1.3910062313079834,
      "learning_rate": 5.6699999999999996e-05,
      "loss": 0.3725,
      "step": 4600
    },
    {
      "epoch": 102.1978021978022,
      "grad_norm": 1.37867271900177,
      "learning_rate": 5.5866666666666665e-05,
      "loss": 0.3802,
      "step": 4650
    },
    {
      "epoch": 103.2967032967033,
      "grad_norm": 1.5553981065750122,
      "learning_rate": 5.5033333333333334e-05,
      "loss": 0.3762,
      "step": 4700
    },
    {
      "epoch": 104.3956043956044,
      "grad_norm": 1.2945665121078491,
      "learning_rate": 5.420000000000001e-05,
      "loss": 0.372,
      "step": 4750
    },
    {
      "epoch": 105.49450549450549,
      "grad_norm": 1.3214540481567383,
      "learning_rate": 5.3366666666666665e-05,
      "loss": 0.3738,
      "step": 4800
    },
    {
      "epoch": 106.5934065934066,
      "grad_norm": 1.4942032098770142,
      "learning_rate": 5.2533333333333334e-05,
      "loss": 0.3833,
      "step": 4850
    },
    {
      "epoch": 107.6923076923077,
      "grad_norm": 1.166876196861267,
      "learning_rate": 5.17e-05,
      "loss": 0.3703,
      "step": 4900
    },
    {
      "epoch": 108.79120879120879,
      "grad_norm": 1.1540136337280273,
      "learning_rate": 5.086666666666667e-05,
      "loss": 0.3757,
      "step": 4950
    },
    {
      "epoch": 109.89010989010988,
      "grad_norm": 2.1926376819610596,
      "learning_rate": 5.005e-05,
      "loss": 0.3804,
      "step": 5000
    },
    {
      "epoch": 109.89010989010988,
      "eval_loss": 0.4093373715877533,
      "eval_runtime": 12.5943,
      "eval_samples_per_second": 12.784,
      "eval_steps_per_second": 1.667,
      "step": 5000
    },
    {
      "epoch": 110.98901098901099,
      "grad_norm": 4.544153690338135,
      "learning_rate": 4.9216666666666666e-05,
      "loss": 0.3751,
      "step": 5050
    },
    {
      "epoch": 112.08791208791209,
      "grad_norm": 2.686535596847534,
      "learning_rate": 4.8383333333333335e-05,
      "loss": 0.3727,
      "step": 5100
    },
    {
      "epoch": 113.18681318681318,
      "grad_norm": 1.5733741521835327,
      "learning_rate": 4.755e-05,
      "loss": 0.3747,
      "step": 5150
    },
    {
      "epoch": 114.28571428571429,
      "grad_norm": 0.9618715643882751,
      "learning_rate": 4.671666666666667e-05,
      "loss": 0.3652,
      "step": 5200
    },
    {
      "epoch": 115.38461538461539,
      "grad_norm": 1.1063612699508667,
      "learning_rate": 4.5883333333333335e-05,
      "loss": 0.3731,
      "step": 5250
    },
    {
      "epoch": 116.48351648351648,
      "grad_norm": 2.2875099182128906,
      "learning_rate": 4.5050000000000004e-05,
      "loss": 0.3749,
      "step": 5300
    },
    {
      "epoch": 117.58241758241758,
      "grad_norm": 2.0566062927246094,
      "learning_rate": 4.4216666666666666e-05,
      "loss": 0.3648,
      "step": 5350
    },
    {
      "epoch": 118.68131868131869,
      "grad_norm": 1.7075659036636353,
      "learning_rate": 4.3383333333333335e-05,
      "loss": 0.3643,
      "step": 5400
    },
    {
      "epoch": 119.78021978021978,
      "grad_norm": 1.8719024658203125,
      "learning_rate": 4.2550000000000004e-05,
      "loss": 0.3664,
      "step": 5450
    },
    {
      "epoch": 120.87912087912088,
      "grad_norm": 1.7855234146118164,
      "learning_rate": 4.171666666666667e-05,
      "loss": 0.3724,
      "step": 5500
    },
    {
      "epoch": 120.87912087912088,
      "eval_loss": 0.4066200256347656,
      "eval_runtime": 12.1606,
      "eval_samples_per_second": 13.239,
      "eval_steps_per_second": 1.727,
      "step": 5500
    },
    {
      "epoch": 121.97802197802197,
      "grad_norm": 1.4458872079849243,
      "learning_rate": 4.0883333333333335e-05,
      "loss": 0.3685,
      "step": 5550
    },
    {
      "epoch": 123.07692307692308,
      "grad_norm": 1.9380217790603638,
      "learning_rate": 4.0050000000000004e-05,
      "loss": 0.3689,
      "step": 5600
    },
    {
      "epoch": 124.17582417582418,
      "grad_norm": 1.1235325336456299,
      "learning_rate": 3.921666666666667e-05,
      "loss": 0.3735,
      "step": 5650
    },
    {
      "epoch": 125.27472527472527,
      "grad_norm": 1.0528467893600464,
      "learning_rate": 3.8383333333333336e-05,
      "loss": 0.3645,
      "step": 5700
    },
    {
      "epoch": 126.37362637362638,
      "grad_norm": 0.8706468939781189,
      "learning_rate": 3.7550000000000005e-05,
      "loss": 0.3651,
      "step": 5750
    },
    {
      "epoch": 127.47252747252747,
      "grad_norm": 2.1038143634796143,
      "learning_rate": 3.671666666666667e-05,
      "loss": 0.3655,
      "step": 5800
    },
    {
      "epoch": 128.57142857142858,
      "grad_norm": 1.101017951965332,
      "learning_rate": 3.5883333333333336e-05,
      "loss": 0.3652,
      "step": 5850
    },
    {
      "epoch": 129.67032967032966,
      "grad_norm": 2.1845877170562744,
      "learning_rate": 3.505e-05,
      "loss": 0.3637,
      "step": 5900
    },
    {
      "epoch": 130.76923076923077,
      "grad_norm": 1.1006217002868652,
      "learning_rate": 3.421666666666667e-05,
      "loss": 0.362,
      "step": 5950
    },
    {
      "epoch": 131.86813186813185,
      "grad_norm": 1.1332905292510986,
      "learning_rate": 3.3383333333333336e-05,
      "loss": 0.3665,
      "step": 6000
    },
    {
      "epoch": 131.86813186813185,
      "eval_loss": 0.40923577547073364,
      "eval_runtime": 11.9318,
      "eval_samples_per_second": 13.493,
      "eval_steps_per_second": 1.76,
      "step": 6000
    },
    {
      "epoch": 132.96703296703296,
      "grad_norm": 1.4925904273986816,
      "learning_rate": 3.2550000000000005e-05,
      "loss": 0.3604,
      "step": 6050
    },
    {
      "epoch": 134.06593406593407,
      "grad_norm": 1.3827078342437744,
      "learning_rate": 3.171666666666667e-05,
      "loss": 0.3639,
      "step": 6100
    },
    {
      "epoch": 135.16483516483515,
      "grad_norm": 1.6526421308517456,
      "learning_rate": 3.0883333333333336e-05,
      "loss": 0.3606,
      "step": 6150
    },
    {
      "epoch": 136.26373626373626,
      "grad_norm": 1.5088169574737549,
      "learning_rate": 3.0050000000000002e-05,
      "loss": 0.3645,
      "step": 6200
    },
    {
      "epoch": 137.36263736263737,
      "grad_norm": 1.899960994720459,
      "learning_rate": 2.921666666666667e-05,
      "loss": 0.3585,
      "step": 6250
    },
    {
      "epoch": 138.46153846153845,
      "grad_norm": 0.7441226243972778,
      "learning_rate": 2.8383333333333333e-05,
      "loss": 0.361,
      "step": 6300
    },
    {
      "epoch": 139.56043956043956,
      "grad_norm": 1.273861289024353,
      "learning_rate": 2.7550000000000002e-05,
      "loss": 0.3635,
      "step": 6350
    },
    {
      "epoch": 140.65934065934067,
      "grad_norm": 1.0249220132827759,
      "learning_rate": 2.6716666666666668e-05,
      "loss": 0.3637,
      "step": 6400
    },
    {
      "epoch": 141.75824175824175,
      "grad_norm": 1.155131220817566,
      "learning_rate": 2.5883333333333337e-05,
      "loss": 0.361,
      "step": 6450
    },
    {
      "epoch": 142.85714285714286,
      "grad_norm": 1.1497211456298828,
      "learning_rate": 2.5050000000000002e-05,
      "loss": 0.3635,
      "step": 6500
    },
    {
      "epoch": 142.85714285714286,
      "eval_loss": 0.4099365472793579,
      "eval_runtime": 12.3576,
      "eval_samples_per_second": 13.028,
      "eval_steps_per_second": 1.699,
      "step": 6500
    },
    {
      "epoch": 143.95604395604394,
      "grad_norm": 1.2265280485153198,
      "learning_rate": 2.4216666666666668e-05,
      "loss": 0.3604,
      "step": 6550
    },
    {
      "epoch": 145.05494505494505,
      "grad_norm": 1.600876808166504,
      "learning_rate": 2.3383333333333334e-05,
      "loss": 0.3592,
      "step": 6600
    },
    {
      "epoch": 146.15384615384616,
      "grad_norm": 1.137926697731018,
      "learning_rate": 2.2550000000000003e-05,
      "loss": 0.3608,
      "step": 6650
    },
    {
      "epoch": 147.25274725274724,
      "grad_norm": 1.297443151473999,
      "learning_rate": 2.1716666666666668e-05,
      "loss": 0.3576,
      "step": 6700
    },
    {
      "epoch": 148.35164835164835,
      "grad_norm": 0.8834235668182373,
      "learning_rate": 2.0883333333333334e-05,
      "loss": 0.3581,
      "step": 6750
    },
    {
      "epoch": 149.45054945054946,
      "grad_norm": 1.0530905723571777,
      "learning_rate": 2.0050000000000003e-05,
      "loss": 0.3569,
      "step": 6800
    },
    {
      "epoch": 150.54945054945054,
      "grad_norm": 1.1900426149368286,
      "learning_rate": 1.921666666666667e-05,
      "loss": 0.3642,
      "step": 6850
    },
    {
      "epoch": 151.64835164835165,
      "grad_norm": 0.733370304107666,
      "learning_rate": 1.8383333333333334e-05,
      "loss": 0.3556,
      "step": 6900
    },
    {
      "epoch": 152.74725274725276,
      "grad_norm": 1.85074782371521,
      "learning_rate": 1.755e-05,
      "loss": 0.3533,
      "step": 6950
    },
    {
      "epoch": 153.84615384615384,
      "grad_norm": 1.4953721761703491,
      "learning_rate": 1.6716666666666665e-05,
      "loss": 0.3562,
      "step": 7000
    },
    {
      "epoch": 153.84615384615384,
      "eval_loss": 0.4075365662574768,
      "eval_runtime": 13.1335,
      "eval_samples_per_second": 12.259,
      "eval_steps_per_second": 1.599,
      "step": 7000
    },
    {
      "epoch": 154.94505494505495,
      "grad_norm": 0.9426752328872681,
      "learning_rate": 1.5883333333333334e-05,
      "loss": 0.3567,
      "step": 7050
    },
    {
      "epoch": 156.04395604395606,
      "grad_norm": 1.0722122192382812,
      "learning_rate": 1.505e-05,
      "loss": 0.3559,
      "step": 7100
    },
    {
      "epoch": 157.14285714285714,
      "grad_norm": 1.1290276050567627,
      "learning_rate": 1.4216666666666667e-05,
      "loss": 0.3654,
      "step": 7150
    },
    {
      "epoch": 158.24175824175825,
      "grad_norm": 1.0128904581069946,
      "learning_rate": 1.3383333333333335e-05,
      "loss": 0.357,
      "step": 7200
    },
    {
      "epoch": 159.34065934065933,
      "grad_norm": 1.2913401126861572,
      "learning_rate": 1.255e-05,
      "loss": 0.3556,
      "step": 7250
    },
    {
      "epoch": 160.43956043956044,
      "grad_norm": 1.0247690677642822,
      "learning_rate": 1.1716666666666667e-05,
      "loss": 0.3512,
      "step": 7300
    },
    {
      "epoch": 161.53846153846155,
      "grad_norm": 1.1389926671981812,
      "learning_rate": 1.0883333333333335e-05,
      "loss": 0.3553,
      "step": 7350
    },
    {
      "epoch": 162.63736263736263,
      "grad_norm": 0.853523850440979,
      "learning_rate": 1.005e-05,
      "loss": 0.357,
      "step": 7400
    },
    {
      "epoch": 163.73626373626374,
      "grad_norm": 0.8105210065841675,
      "learning_rate": 9.216666666666666e-06,
      "loss": 0.3512,
      "step": 7450
    },
    {
      "epoch": 164.83516483516485,
      "grad_norm": 0.891890287399292,
      "learning_rate": 8.383333333333333e-06,
      "loss": 0.3581,
      "step": 7500
    },
    {
      "epoch": 164.83516483516485,
      "eval_loss": 0.4096794128417969,
      "eval_runtime": 12.3085,
      "eval_samples_per_second": 13.08,
      "eval_steps_per_second": 1.706,
      "step": 7500
    },
    {
      "epoch": 165.93406593406593,
      "grad_norm": 1.1622841358184814,
      "learning_rate": 7.55e-06,
      "loss": 0.3542,
      "step": 7550
    },
    {
      "epoch": 167.03296703296704,
      "grad_norm": 0.767804741859436,
      "learning_rate": 6.716666666666667e-06,
      "loss": 0.3516,
      "step": 7600
    },
    {
      "epoch": 168.13186813186815,
      "grad_norm": 0.7939496636390686,
      "learning_rate": 5.8833333333333335e-06,
      "loss": 0.3516,
      "step": 7650
    },
    {
      "epoch": 169.23076923076923,
      "grad_norm": 0.826722264289856,
      "learning_rate": 5.050000000000001e-06,
      "loss": 0.3536,
      "step": 7700
    },
    {
      "epoch": 170.32967032967034,
      "grad_norm": 1.0055506229400635,
      "learning_rate": 4.216666666666666e-06,
      "loss": 0.3515,
      "step": 7750
    },
    {
      "epoch": 171.42857142857142,
      "grad_norm": 0.9588921070098877,
      "learning_rate": 3.3833333333333337e-06,
      "loss": 0.3501,
      "step": 7800
    },
    {
      "epoch": 172.52747252747253,
      "grad_norm": 0.7313802242279053,
      "learning_rate": 2.55e-06,
      "loss": 0.353,
      "step": 7850
    },
    {
      "epoch": 173.62637362637363,
      "grad_norm": 0.7923821210861206,
      "learning_rate": 1.7166666666666668e-06,
      "loss": 0.3526,
      "step": 7900
    },
    {
      "epoch": 174.72527472527472,
      "grad_norm": 0.7615464329719543,
      "learning_rate": 8.833333333333334e-07,
      "loss": 0.3473,
      "step": 7950
    },
    {
      "epoch": 175.82417582417582,
      "grad_norm": 0.6758411526679993,
      "learning_rate": 5.0000000000000004e-08,
      "loss": 0.3461,
      "step": 8000
    },
    {
      "epoch": 175.82417582417582,
      "eval_loss": 0.40872836112976074,
      "eval_runtime": 13.3243,
      "eval_samples_per_second": 12.083,
      "eval_steps_per_second": 1.576,
      "step": 8000
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 178,
  "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": 4.743072546477725e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}