{
  "best_metric": 0.3666408658027649,
  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_s8000/checkpoint-7500",
  "epoch": 12.565445026178011,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 6.784719467163086,
      "learning_rate": 2.4500000000000004e-07,
      "loss": 1.1115,
      "step": 50
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 12.37061595916748,
      "learning_rate": 4.95e-07,
      "loss": 1.1117,
      "step": 100
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 2.4181911945343018,
      "learning_rate": 7.450000000000001e-07,
      "loss": 0.9761,
      "step": 150
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 5.356767177581787,
      "learning_rate": 9.950000000000002e-07,
      "loss": 0.8691,
      "step": 200
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 1.991185188293457,
      "learning_rate": 1.2450000000000002e-06,
      "loss": 0.816,
      "step": 250
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 1.8864444494247437,
      "learning_rate": 1.495e-06,
      "loss": 0.7841,
      "step": 300
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 2.3990137577056885,
      "learning_rate": 1.745e-06,
      "loss": 0.7553,
      "step": 350
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 3.0834362506866455,
      "learning_rate": 1.9950000000000004e-06,
      "loss": 0.7231,
      "step": 400
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 1.6590628623962402,
      "learning_rate": 2.245e-06,
      "loss": 0.6887,
      "step": 450
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 2.2680208683013916,
      "learning_rate": 2.4950000000000003e-06,
      "loss": 0.6645,
      "step": 500
    },
    {
      "epoch": 0.837696335078534,
      "eval_loss": 0.5697648525238037,
      "eval_runtime": 253.6,
      "eval_samples_per_second": 33.474,
      "eval_steps_per_second": 4.188,
      "step": 500
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 1.2383062839508057,
      "learning_rate": 2.7450000000000004e-06,
      "loss": 0.6485,
      "step": 550
    },
    {
      "epoch": 1.0052356020942408,
      "grad_norm": 1.7865532636642456,
      "learning_rate": 2.995e-06,
      "loss": 0.6305,
      "step": 600
    },
    {
      "epoch": 1.0890052356020943,
      "grad_norm": 5.1390790939331055,
      "learning_rate": 3.2450000000000003e-06,
      "loss": 0.625,
      "step": 650
    },
    {
      "epoch": 1.1727748691099475,
      "grad_norm": 1.56719172000885,
      "learning_rate": 3.495e-06,
      "loss": 0.6164,
      "step": 700
    },
    {
      "epoch": 1.256544502617801,
      "grad_norm": 2.732139825820923,
      "learning_rate": 3.745e-06,
      "loss": 0.5949,
      "step": 750
    },
    {
      "epoch": 1.3403141361256545,
      "grad_norm": 2.0998775959014893,
      "learning_rate": 3.995000000000001e-06,
      "loss": 0.5885,
      "step": 800
    },
    {
      "epoch": 1.4240837696335078,
      "grad_norm": 9.37508487701416,
      "learning_rate": 4.245e-06,
      "loss": 0.5763,
      "step": 850
    },
    {
      "epoch": 1.5078534031413613,
      "grad_norm": 2.467970132827759,
      "learning_rate": 4.495e-06,
      "loss": 0.5671,
      "step": 900
    },
    {
      "epoch": 1.5916230366492146,
      "grad_norm": 1.5992038249969482,
      "learning_rate": 4.745e-06,
      "loss": 0.5569,
      "step": 950
    },
    {
      "epoch": 1.675392670157068,
      "grad_norm": 2.854398727416992,
      "learning_rate": 4.9950000000000005e-06,
      "loss": 0.5581,
      "step": 1000
    },
    {
      "epoch": 1.675392670157068,
      "eval_loss": 0.4793892800807953,
      "eval_runtime": 254.3858,
      "eval_samples_per_second": 33.371,
      "eval_steps_per_second": 4.175,
      "step": 1000
    },
    {
      "epoch": 1.7591623036649215,
      "grad_norm": 2.9313721656799316,
      "learning_rate": 5.245e-06,
      "loss": 0.5573,
      "step": 1050
    },
    {
      "epoch": 1.8429319371727748,
      "grad_norm": 1.984640121459961,
      "learning_rate": 5.495000000000001e-06,
      "loss": 0.5455,
      "step": 1100
    },
    {
      "epoch": 1.9267015706806283,
      "grad_norm": 2.0967133045196533,
      "learning_rate": 5.745000000000001e-06,
      "loss": 0.5301,
      "step": 1150
    },
    {
      "epoch": 2.0104712041884816,
      "grad_norm": 2.864819049835205,
      "learning_rate": 5.995000000000001e-06,
      "loss": 0.5249,
      "step": 1200
    },
    {
      "epoch": 2.094240837696335,
      "grad_norm": 1.5830483436584473,
      "learning_rate": 6.245000000000001e-06,
      "loss": 0.5265,
      "step": 1250
    },
    {
      "epoch": 2.1780104712041886,
      "grad_norm": 1.706629753112793,
      "learning_rate": 6.4950000000000005e-06,
      "loss": 0.5226,
      "step": 1300
    },
    {
      "epoch": 2.261780104712042,
      "grad_norm": 2.3449299335479736,
      "learning_rate": 6.745000000000001e-06,
      "loss": 0.5159,
      "step": 1350
    },
    {
      "epoch": 2.345549738219895,
      "grad_norm": 2.4408295154571533,
      "learning_rate": 6.995000000000001e-06,
      "loss": 0.517,
      "step": 1400
    },
    {
      "epoch": 2.4293193717277486,
      "grad_norm": 1.8883627653121948,
      "learning_rate": 7.245000000000001e-06,
      "loss": 0.5139,
      "step": 1450
    },
    {
      "epoch": 2.513089005235602,
      "grad_norm": 1.5606210231781006,
      "learning_rate": 7.495000000000001e-06,
      "loss": 0.5045,
      "step": 1500
    },
    {
      "epoch": 2.513089005235602,
      "eval_loss": 0.44668668508529663,
      "eval_runtime": 252.363,
      "eval_samples_per_second": 33.638,
      "eval_steps_per_second": 4.208,
      "step": 1500
    },
    {
      "epoch": 2.5968586387434556,
      "grad_norm": 1.6376454830169678,
      "learning_rate": 7.745e-06,
      "loss": 0.503,
      "step": 1550
    },
    {
      "epoch": 2.680628272251309,
      "grad_norm": 1.7104960680007935,
      "learning_rate": 7.995e-06,
      "loss": 0.4988,
      "step": 1600
    },
    {
      "epoch": 2.7643979057591626,
      "grad_norm": 2.3372185230255127,
      "learning_rate": 8.245000000000002e-06,
      "loss": 0.4954,
      "step": 1650
    },
    {
      "epoch": 2.8481675392670156,
      "grad_norm": 2.170474052429199,
      "learning_rate": 8.495e-06,
      "loss": 0.4971,
      "step": 1700
    },
    {
      "epoch": 2.931937172774869,
      "grad_norm": 2.3041465282440186,
      "learning_rate": 8.745000000000002e-06,
      "loss": 0.4948,
      "step": 1750
    },
    {
      "epoch": 3.0157068062827226,
      "grad_norm": 2.1798529624938965,
      "learning_rate": 8.995000000000001e-06,
      "loss": 0.4849,
      "step": 1800
    },
    {
      "epoch": 3.099476439790576,
      "grad_norm": 1.8813133239746094,
      "learning_rate": 9.245e-06,
      "loss": 0.4839,
      "step": 1850
    },
    {
      "epoch": 3.183246073298429,
      "grad_norm": 1.8749632835388184,
      "learning_rate": 9.49e-06,
      "loss": 0.4871,
      "step": 1900
    },
    {
      "epoch": 3.2670157068062826,
      "grad_norm": 1.917149305343628,
      "learning_rate": 9.74e-06,
      "loss": 0.4724,
      "step": 1950
    },
    {
      "epoch": 3.350785340314136,
      "grad_norm": 1.9401865005493164,
      "learning_rate": 9.990000000000001e-06,
      "loss": 0.4776,
      "step": 2000
    },
    {
      "epoch": 3.350785340314136,
      "eval_loss": 0.4236186444759369,
      "eval_runtime": 261.4044,
      "eval_samples_per_second": 32.475,
      "eval_steps_per_second": 4.063,
      "step": 2000
    },
    {
      "epoch": 3.4345549738219896,
      "grad_norm": 1.67532479763031,
      "learning_rate": 9.921666666666667e-06,
      "loss": 0.4797,
      "step": 2050
    },
    {
      "epoch": 3.518324607329843,
      "grad_norm": 2.044435977935791,
      "learning_rate": 9.838333333333334e-06,
      "loss": 0.475,
      "step": 2100
    },
    {
      "epoch": 3.6020942408376966,
      "grad_norm": 1.7103309631347656,
      "learning_rate": 9.755e-06,
      "loss": 0.4729,
      "step": 2150
    },
    {
      "epoch": 3.6858638743455496,
      "grad_norm": 2.557506799697876,
      "learning_rate": 9.671666666666668e-06,
      "loss": 0.4671,
      "step": 2200
    },
    {
      "epoch": 3.769633507853403,
      "grad_norm": 2.5080385208129883,
      "learning_rate": 9.588333333333334e-06,
      "loss": 0.4613,
      "step": 2250
    },
    {
      "epoch": 3.8534031413612566,
      "grad_norm": 1.6658800840377808,
      "learning_rate": 9.505000000000001e-06,
      "loss": 0.4614,
      "step": 2300
    },
    {
      "epoch": 3.93717277486911,
      "grad_norm": 2.167924404144287,
      "learning_rate": 9.421666666666668e-06,
      "loss": 0.4685,
      "step": 2350
    },
    {
      "epoch": 4.020942408376963,
      "grad_norm": 4.781851768493652,
      "learning_rate": 9.338333333333333e-06,
      "loss": 0.4578,
      "step": 2400
    },
    {
      "epoch": 4.104712041884817,
      "grad_norm": 2.1949734687805176,
      "learning_rate": 9.255e-06,
      "loss": 0.4613,
      "step": 2450
    },
    {
      "epoch": 4.18848167539267,
      "grad_norm": 1.9870787858963013,
      "learning_rate": 9.171666666666667e-06,
      "loss": 0.4553,
      "step": 2500
    },
    {
      "epoch": 4.18848167539267,
      "eval_loss": 0.40933898091316223,
      "eval_runtime": 252.812,
      "eval_samples_per_second": 33.578,
      "eval_steps_per_second": 4.201,
      "step": 2500
    },
    {
      "epoch": 4.272251308900524,
      "grad_norm": 5.594339370727539,
      "learning_rate": 9.088333333333334e-06,
      "loss": 0.4592,
      "step": 2550
    },
    {
      "epoch": 4.356020942408377,
      "grad_norm": 2.930290699005127,
      "learning_rate": 9.005000000000001e-06,
      "loss": 0.4549,
      "step": 2600
    },
    {
      "epoch": 4.439790575916231,
      "grad_norm": 2.199352741241455,
      "learning_rate": 8.921666666666668e-06,
      "loss": 0.4512,
      "step": 2650
    },
    {
      "epoch": 4.523560209424084,
      "grad_norm": 2.435898780822754,
      "learning_rate": 8.838333333333335e-06,
      "loss": 0.4557,
      "step": 2700
    },
    {
      "epoch": 4.607329842931938,
      "grad_norm": 3.2397773265838623,
      "learning_rate": 8.755e-06,
      "loss": 0.449,
      "step": 2750
    },
    {
      "epoch": 4.69109947643979,
      "grad_norm": 1.9361001253128052,
      "learning_rate": 8.671666666666667e-06,
      "loss": 0.4525,
      "step": 2800
    },
    {
      "epoch": 4.774869109947644,
      "grad_norm": 2.1852686405181885,
      "learning_rate": 8.588333333333334e-06,
      "loss": 0.4486,
      "step": 2850
    },
    {
      "epoch": 4.858638743455497,
      "grad_norm": 1.8136191368103027,
      "learning_rate": 8.505e-06,
      "loss": 0.4517,
      "step": 2900
    },
    {
      "epoch": 4.942408376963351,
      "grad_norm": 2.694406032562256,
      "learning_rate": 8.421666666666668e-06,
      "loss": 0.4505,
      "step": 2950
    },
    {
      "epoch": 5.026178010471204,
      "grad_norm": 1.251187801361084,
      "learning_rate": 8.338333333333335e-06,
      "loss": 0.4489,
      "step": 3000
    },
    {
      "epoch": 5.026178010471204,
      "eval_loss": 0.39678552746772766,
      "eval_runtime": 256.1618,
      "eval_samples_per_second": 33.139,
      "eval_steps_per_second": 4.146,
      "step": 3000
    },
    {
      "epoch": 5.109947643979058,
      "grad_norm": 1.1454211473464966,
      "learning_rate": 8.255000000000001e-06,
      "loss": 0.4422,
      "step": 3050
    },
    {
      "epoch": 5.193717277486911,
      "grad_norm": 1.8685294389724731,
      "learning_rate": 8.171666666666668e-06,
      "loss": 0.4426,
      "step": 3100
    },
    {
      "epoch": 5.277486910994765,
      "grad_norm": 1.6863799095153809,
      "learning_rate": 8.088333333333334e-06,
      "loss": 0.4398,
      "step": 3150
    },
    {
      "epoch": 5.361256544502618,
      "grad_norm": 2.249805212020874,
      "learning_rate": 8.005e-06,
      "loss": 0.4384,
      "step": 3200
    },
    {
      "epoch": 5.445026178010472,
      "grad_norm": 2.1187326908111572,
      "learning_rate": 7.921666666666667e-06,
      "loss": 0.4431,
      "step": 3250
    },
    {
      "epoch": 5.528795811518324,
      "grad_norm": 1.8476357460021973,
      "learning_rate": 7.838333333333334e-06,
      "loss": 0.4434,
      "step": 3300
    },
    {
      "epoch": 5.612565445026178,
      "grad_norm": 1.6522760391235352,
      "learning_rate": 7.755000000000001e-06,
      "loss": 0.4343,
      "step": 3350
    },
    {
      "epoch": 5.696335078534031,
      "grad_norm": 1.3926664590835571,
      "learning_rate": 7.671666666666668e-06,
      "loss": 0.4365,
      "step": 3400
    },
    {
      "epoch": 5.780104712041885,
      "grad_norm": 2.1967947483062744,
      "learning_rate": 7.588333333333334e-06,
      "loss": 0.4362,
      "step": 3450
    },
    {
      "epoch": 5.863874345549738,
      "grad_norm": 1.5428054332733154,
      "learning_rate": 7.505e-06,
      "loss": 0.4337,
      "step": 3500
    },
    {
      "epoch": 5.863874345549738,
      "eval_loss": 0.39255669713020325,
      "eval_runtime": 251.4241,
      "eval_samples_per_second": 33.764,
      "eval_steps_per_second": 4.224,
      "step": 3500
    },
    {
      "epoch": 5.947643979057592,
      "grad_norm": 1.7545124292373657,
      "learning_rate": 7.421666666666667e-06,
      "loss": 0.4292,
      "step": 3550
    },
    {
      "epoch": 6.031413612565445,
      "grad_norm": 1.1912785768508911,
      "learning_rate": 7.338333333333334e-06,
      "loss": 0.4293,
      "step": 3600
    },
    {
      "epoch": 6.115183246073299,
      "grad_norm": 1.549297571182251,
      "learning_rate": 7.255000000000001e-06,
      "loss": 0.4333,
      "step": 3650
    },
    {
      "epoch": 6.198952879581152,
      "grad_norm": 1.8822449445724487,
      "learning_rate": 7.171666666666667e-06,
      "loss": 0.4308,
      "step": 3700
    },
    {
      "epoch": 6.282722513089006,
      "grad_norm": 1.6894242763519287,
      "learning_rate": 7.088333333333334e-06,
      "loss": 0.4336,
      "step": 3750
    },
    {
      "epoch": 6.366492146596858,
      "grad_norm": 1.63300621509552,
      "learning_rate": 7.005000000000001e-06,
      "loss": 0.432,
      "step": 3800
    },
    {
      "epoch": 6.450261780104712,
      "grad_norm": 1.4569647312164307,
      "learning_rate": 6.921666666666668e-06,
      "loss": 0.4302,
      "step": 3850
    },
    {
      "epoch": 6.534031413612565,
      "grad_norm": 1.3649786710739136,
      "learning_rate": 6.838333333333334e-06,
      "loss": 0.4277,
      "step": 3900
    },
    {
      "epoch": 6.617801047120419,
      "grad_norm": 1.6974161863327026,
      "learning_rate": 6.7550000000000005e-06,
      "loss": 0.4329,
      "step": 3950
    },
    {
      "epoch": 6.701570680628272,
      "grad_norm": 8.575983047485352,
      "learning_rate": 6.6716666666666674e-06,
      "loss": 0.4282,
      "step": 4000
    },
    {
      "epoch": 6.701570680628272,
      "eval_loss": 0.3837336301803589,
      "eval_runtime": 252.6194,
      "eval_samples_per_second": 33.604,
      "eval_steps_per_second": 4.204,
      "step": 4000
    },
    {
      "epoch": 6.785340314136126,
      "grad_norm": 1.308820366859436,
      "learning_rate": 6.588333333333334e-06,
      "loss": 0.4233,
      "step": 4050
    },
    {
      "epoch": 6.869109947643979,
      "grad_norm": 1.6496186256408691,
      "learning_rate": 6.505e-06,
      "loss": 0.4235,
      "step": 4100
    },
    {
      "epoch": 6.952879581151833,
      "grad_norm": 2.9797613620758057,
      "learning_rate": 6.421666666666667e-06,
      "loss": 0.424,
      "step": 4150
    },
    {
      "epoch": 7.036649214659686,
      "grad_norm": 1.366092324256897,
      "learning_rate": 6.338333333333334e-06,
      "loss": 0.4241,
      "step": 4200
    },
    {
      "epoch": 7.12041884816754,
      "grad_norm": 1.2892742156982422,
      "learning_rate": 6.255e-06,
      "loss": 0.4306,
      "step": 4250
    },
    {
      "epoch": 7.204188481675392,
      "grad_norm": 1.769739031791687,
      "learning_rate": 6.171666666666667e-06,
      "loss": 0.4206,
      "step": 4300
    },
    {
      "epoch": 7.287958115183246,
      "grad_norm": 1.2120298147201538,
      "learning_rate": 6.088333333333334e-06,
      "loss": 0.4245,
      "step": 4350
    },
    {
      "epoch": 7.371727748691099,
      "grad_norm": 2.545499086380005,
      "learning_rate": 6.005000000000001e-06,
      "loss": 0.4234,
      "step": 4400
    },
    {
      "epoch": 7.455497382198953,
      "grad_norm": 1.57191801071167,
      "learning_rate": 5.921666666666667e-06,
      "loss": 0.426,
      "step": 4450
    },
    {
      "epoch": 7.539267015706806,
      "grad_norm": 1.3846454620361328,
      "learning_rate": 5.838333333333334e-06,
      "loss": 0.4188,
      "step": 4500
    },
    {
      "epoch": 7.539267015706806,
      "eval_loss": 0.37984684109687805,
      "eval_runtime": 250.9224,
      "eval_samples_per_second": 33.831,
      "eval_steps_per_second": 4.232,
      "step": 4500
    },
    {
      "epoch": 7.62303664921466,
      "grad_norm": 1.3911954164505005,
      "learning_rate": 5.755000000000001e-06,
      "loss": 0.42,
      "step": 4550
    },
    {
      "epoch": 7.706806282722513,
      "grad_norm": 1.3781930208206177,
      "learning_rate": 5.671666666666668e-06,
      "loss": 0.4234,
      "step": 4600
    },
    {
      "epoch": 7.790575916230367,
      "grad_norm": 1.5225090980529785,
      "learning_rate": 5.588333333333334e-06,
      "loss": 0.4242,
      "step": 4650
    },
    {
      "epoch": 7.87434554973822,
      "grad_norm": 1.4195870161056519,
      "learning_rate": 5.505000000000001e-06,
      "loss": 0.4243,
      "step": 4700
    },
    {
      "epoch": 7.958115183246074,
      "grad_norm": 1.9713116884231567,
      "learning_rate": 5.4216666666666676e-06,
      "loss": 0.4158,
      "step": 4750
    },
    {
      "epoch": 8.041884816753926,
      "grad_norm": 1.3215075731277466,
      "learning_rate": 5.3383333333333345e-06,
      "loss": 0.4141,
      "step": 4800
    },
    {
      "epoch": 8.12565445026178,
      "grad_norm": 4.483395576477051,
      "learning_rate": 5.2550000000000005e-06,
      "loss": 0.4165,
      "step": 4850
    },
    {
      "epoch": 8.209424083769633,
      "grad_norm": 1.4759974479675293,
      "learning_rate": 5.171666666666667e-06,
      "loss": 0.4291,
      "step": 4900
    },
    {
      "epoch": 8.293193717277488,
      "grad_norm": 1.3336604833602905,
      "learning_rate": 5.088333333333334e-06,
      "loss": 0.4134,
      "step": 4950
    },
    {
      "epoch": 8.37696335078534,
      "grad_norm": 1.5618665218353271,
      "learning_rate": 5.0049999999999995e-06,
      "loss": 0.4222,
      "step": 5000
    },
    {
      "epoch": 8.37696335078534,
      "eval_loss": 0.37838345766067505,
      "eval_runtime": 256.1892,
      "eval_samples_per_second": 33.136,
      "eval_steps_per_second": 4.145,
      "step": 5000
    },
    {
      "epoch": 8.460732984293193,
      "grad_norm": 1.079275369644165,
      "learning_rate": 4.921666666666666e-06,
      "loss": 0.4143,
      "step": 5050
    },
    {
      "epoch": 8.544502617801047,
      "grad_norm": 1.0949628353118896,
      "learning_rate": 4.838333333333334e-06,
      "loss": 0.4123,
      "step": 5100
    },
    {
      "epoch": 8.6282722513089,
      "grad_norm": 1.5005624294281006,
      "learning_rate": 4.755e-06,
      "loss": 0.4218,
      "step": 5150
    },
    {
      "epoch": 8.712041884816754,
      "grad_norm": 5.408727169036865,
      "learning_rate": 4.671666666666667e-06,
      "loss": 0.4123,
      "step": 5200
    },
    {
      "epoch": 8.795811518324607,
      "grad_norm": 1.0426135063171387,
      "learning_rate": 4.588333333333333e-06,
      "loss": 0.4159,
      "step": 5250
    },
    {
      "epoch": 8.879581151832461,
      "grad_norm": 2.312485456466675,
      "learning_rate": 4.505e-06,
      "loss": 0.4149,
      "step": 5300
    },
    {
      "epoch": 8.963350785340314,
      "grad_norm": 1.6531749963760376,
      "learning_rate": 4.421666666666667e-06,
      "loss": 0.4147,
      "step": 5350
    },
    {
      "epoch": 9.047120418848168,
      "grad_norm": 1.2448443174362183,
      "learning_rate": 4.338333333333334e-06,
      "loss": 0.4191,
      "step": 5400
    },
    {
      "epoch": 9.13089005235602,
      "grad_norm": 1.1506507396697998,
      "learning_rate": 4.255e-06,
      "loss": 0.4094,
      "step": 5450
    },
    {
      "epoch": 9.214659685863875,
      "grad_norm": 1.2973967790603638,
      "learning_rate": 4.171666666666667e-06,
      "loss": 0.412,
      "step": 5500
    },
    {
      "epoch": 9.214659685863875,
      "eval_loss": 0.3729405403137207,
      "eval_runtime": 258.7159,
      "eval_samples_per_second": 32.812,
      "eval_steps_per_second": 4.105,
      "step": 5500
    },
    {
      "epoch": 9.298429319371728,
      "grad_norm": 1.0570714473724365,
      "learning_rate": 4.088333333333334e-06,
      "loss": 0.4144,
      "step": 5550
    },
    {
      "epoch": 9.38219895287958,
      "grad_norm": 1.1278653144836426,
      "learning_rate": 4.005000000000001e-06,
      "loss": 0.4154,
      "step": 5600
    },
    {
      "epoch": 9.465968586387435,
      "grad_norm": 5.811945915222168,
      "learning_rate": 3.921666666666667e-06,
      "loss": 0.4085,
      "step": 5650
    },
    {
      "epoch": 9.549738219895287,
      "grad_norm": 1.6973522901535034,
      "learning_rate": 3.8383333333333336e-06,
      "loss": 0.4133,
      "step": 5700
    },
    {
      "epoch": 9.633507853403142,
      "grad_norm": 1.209333062171936,
      "learning_rate": 3.7550000000000005e-06,
      "loss": 0.4123,
      "step": 5750
    },
    {
      "epoch": 9.717277486910994,
      "grad_norm": 3.592991590499878,
      "learning_rate": 3.6716666666666665e-06,
      "loss": 0.4126,
      "step": 5800
    },
    {
      "epoch": 9.801047120418849,
      "grad_norm": 1.152239203453064,
      "learning_rate": 3.588333333333334e-06,
      "loss": 0.4115,
      "step": 5850
    },
    {
      "epoch": 9.884816753926701,
      "grad_norm": 1.6118751764297485,
      "learning_rate": 3.505e-06,
      "loss": 0.4036,
      "step": 5900
    },
    {
      "epoch": 9.968586387434556,
      "grad_norm": 1.4384329319000244,
      "learning_rate": 3.4216666666666672e-06,
      "loss": 0.4103,
      "step": 5950
    },
    {
      "epoch": 10.052356020942408,
      "grad_norm": 1.447549819946289,
      "learning_rate": 3.3383333333333333e-06,
      "loss": 0.4056,
      "step": 6000
    },
    {
      "epoch": 10.052356020942408,
      "eval_loss": 0.3696598410606384,
      "eval_runtime": 256.5857,
      "eval_samples_per_second": 33.084,
      "eval_steps_per_second": 4.139,
      "step": 6000
    },
    {
      "epoch": 10.136125654450261,
      "grad_norm": 1.2832478284835815,
      "learning_rate": 3.255e-06,
      "loss": 0.4086,
      "step": 6050
    },
    {
      "epoch": 10.219895287958115,
      "grad_norm": 1.0580389499664307,
      "learning_rate": 3.1716666666666667e-06,
      "loss": 0.4128,
      "step": 6100
    },
    {
      "epoch": 10.303664921465968,
      "grad_norm": 1.5535204410552979,
      "learning_rate": 3.0883333333333336e-06,
      "loss": 0.4088,
      "step": 6150
    },
    {
      "epoch": 10.387434554973822,
      "grad_norm": 0.8630754947662354,
      "learning_rate": 3.005e-06,
      "loss": 0.4073,
      "step": 6200
    },
    {
      "epoch": 10.471204188481675,
      "grad_norm": 1.248646855354309,
      "learning_rate": 2.921666666666667e-06,
      "loss": 0.4074,
      "step": 6250
    },
    {
      "epoch": 10.55497382198953,
      "grad_norm": 0.9990864396095276,
      "learning_rate": 2.8383333333333334e-06,
      "loss": 0.4076,
      "step": 6300
    },
    {
      "epoch": 10.638743455497382,
      "grad_norm": 1.100101351737976,
      "learning_rate": 2.7550000000000003e-06,
      "loss": 0.4081,
      "step": 6350
    },
    {
      "epoch": 10.722513089005236,
      "grad_norm": 1.1483354568481445,
      "learning_rate": 2.6716666666666668e-06,
      "loss": 0.4088,
      "step": 6400
    },
    {
      "epoch": 10.806282722513089,
      "grad_norm": 0.8169103860855103,
      "learning_rate": 2.5883333333333337e-06,
      "loss": 0.406,
      "step": 6450
    },
    {
      "epoch": 10.890052356020943,
      "grad_norm": 1.0567216873168945,
      "learning_rate": 2.505e-06,
      "loss": 0.4065,
      "step": 6500
    },
    {
      "epoch": 10.890052356020943,
      "eval_loss": 0.36852771043777466,
      "eval_runtime": 255.1911,
      "eval_samples_per_second": 33.265,
      "eval_steps_per_second": 4.162,
      "step": 6500
    },
    {
      "epoch": 10.973821989528796,
      "grad_norm": 1.0459740161895752,
      "learning_rate": 2.421666666666667e-06,
      "loss": 0.4074,
      "step": 6550
    },
    {
      "epoch": 11.057591623036648,
      "grad_norm": 0.9029247760772705,
      "learning_rate": 2.3383333333333335e-06,
      "loss": 0.4075,
      "step": 6600
    },
    {
      "epoch": 11.141361256544503,
      "grad_norm": 1.5372889041900635,
      "learning_rate": 2.2550000000000004e-06,
      "loss": 0.4088,
      "step": 6650
    },
    {
      "epoch": 11.225130890052355,
      "grad_norm": 0.9959379434585571,
      "learning_rate": 2.171666666666667e-06,
      "loss": 0.4044,
      "step": 6700
    },
    {
      "epoch": 11.30890052356021,
      "grad_norm": 1.3793728351593018,
      "learning_rate": 2.088333333333334e-06,
      "loss": 0.4034,
      "step": 6750
    },
    {
      "epoch": 11.392670157068062,
      "grad_norm": 1.2086491584777832,
      "learning_rate": 2.0050000000000003e-06,
      "loss": 0.4073,
      "step": 6800
    },
    {
      "epoch": 11.476439790575917,
      "grad_norm": 1.07647705078125,
      "learning_rate": 1.9216666666666668e-06,
      "loss": 0.405,
      "step": 6850
    },
    {
      "epoch": 11.56020942408377,
      "grad_norm": 0.9849846363067627,
      "learning_rate": 1.8383333333333334e-06,
      "loss": 0.4037,
      "step": 6900
    },
    {
      "epoch": 11.643979057591624,
      "grad_norm": 1.2623456716537476,
      "learning_rate": 1.7550000000000001e-06,
      "loss": 0.4042,
      "step": 6950
    },
    {
      "epoch": 11.727748691099476,
      "grad_norm": 0.9488279819488525,
      "learning_rate": 1.6716666666666666e-06,
      "loss": 0.4069,
      "step": 7000
    },
    {
      "epoch": 11.727748691099476,
      "eval_loss": 0.3675082325935364,
      "eval_runtime": 257.7947,
      "eval_samples_per_second": 32.929,
      "eval_steps_per_second": 4.12,
      "step": 7000
    },
    {
      "epoch": 11.81151832460733,
      "grad_norm": 1.7772880792617798,
      "learning_rate": 1.5883333333333333e-06,
      "loss": 0.4062,
      "step": 7050
    },
    {
      "epoch": 11.895287958115183,
      "grad_norm": 0.9282962083816528,
      "learning_rate": 1.505e-06,
      "loss": 0.4097,
      "step": 7100
    },
    {
      "epoch": 11.979057591623036,
      "grad_norm": 0.9623945355415344,
      "learning_rate": 1.4216666666666667e-06,
      "loss": 0.4022,
      "step": 7150
    },
    {
      "epoch": 12.06282722513089,
      "grad_norm": 0.8984498381614685,
      "learning_rate": 1.34e-06,
      "loss": 0.4136,
      "step": 7200
    },
    {
      "epoch": 12.146596858638743,
      "grad_norm": 1.1274847984313965,
      "learning_rate": 1.2566666666666668e-06,
      "loss": 0.4074,
      "step": 7250
    },
    {
      "epoch": 12.230366492146597,
      "grad_norm": 0.9233041405677795,
      "learning_rate": 1.1733333333333335e-06,
      "loss": 0.4051,
      "step": 7300
    },
    {
      "epoch": 12.31413612565445,
      "grad_norm": 1.0804561376571655,
      "learning_rate": 1.0900000000000002e-06,
      "loss": 0.4045,
      "step": 7350
    },
    {
      "epoch": 12.397905759162304,
      "grad_norm": 0.9518505334854126,
      "learning_rate": 1.0066666666666668e-06,
      "loss": 0.4097,
      "step": 7400
    },
    {
      "epoch": 12.481675392670157,
      "grad_norm": 1.3323400020599365,
      "learning_rate": 9.233333333333334e-07,
      "loss": 0.4077,
      "step": 7450
    },
    {
      "epoch": 12.565445026178011,
      "grad_norm": 0.8265945911407471,
      "learning_rate": 8.400000000000001e-07,
      "loss": 0.4049,
      "step": 7500
    },
    {
      "epoch": 12.565445026178011,
      "eval_loss": 0.3666408658027649,
      "eval_runtime": 263.2918,
      "eval_samples_per_second": 32.242,
      "eval_steps_per_second": 4.034,
      "step": 7500
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 14,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3437364892618304e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}