{
  "best_metric": 0.4078540802001953,
  "best_model_checkpoint": "mikhail-panzo/fil_b128_le4_s8000/checkpoint-1000",
  "epoch": 618.7845303867404,
  "eval_steps": 500,
  "global_step": 7000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.419889502762431,
      "grad_norm": 2.1200406551361084,
      "learning_rate": 2.5e-06,
      "loss": 0.77,
      "step": 50
    },
    {
      "epoch": 8.839779005524862,
      "grad_norm": 0.8919622302055359,
      "learning_rate": 5e-06,
      "loss": 0.695,
      "step": 100
    },
    {
      "epoch": 13.259668508287293,
      "grad_norm": 0.9864558577537537,
      "learning_rate": 7.5e-06,
      "loss": 0.6278,
      "step": 150
    },
    {
      "epoch": 17.679558011049725,
      "grad_norm": 0.9195986986160278,
      "learning_rate": 1e-05,
      "loss": 0.5431,
      "step": 200
    },
    {
      "epoch": 22.099447513812155,
      "grad_norm": 1.0175281763076782,
      "learning_rate": 1.25e-05,
      "loss": 0.5155,
      "step": 250
    },
    {
      "epoch": 26.519337016574585,
      "grad_norm": 1.0303566455841064,
      "learning_rate": 1.5e-05,
      "loss": 0.502,
      "step": 300
    },
    {
      "epoch": 30.939226519337016,
      "grad_norm": 1.005897879600525,
      "learning_rate": 1.75e-05,
      "loss": 0.4889,
      "step": 350
    },
    {
      "epoch": 35.35911602209945,
      "grad_norm": 1.0381360054016113,
      "learning_rate": 2e-05,
      "loss": 0.477,
      "step": 400
    },
    {
      "epoch": 39.77900552486188,
      "grad_norm": 1.2148911952972412,
      "learning_rate": 2.25e-05,
      "loss": 0.4711,
      "step": 450
    },
    {
      "epoch": 44.19889502762431,
      "grad_norm": 1.5404785871505737,
      "learning_rate": 2.5e-05,
      "loss": 0.463,
      "step": 500
    },
    {
      "epoch": 44.19889502762431,
      "eval_loss": 0.42523157596588135,
      "eval_runtime": 15.6706,
      "eval_samples_per_second": 10.274,
      "eval_steps_per_second": 1.34,
      "step": 500
    },
    {
      "epoch": 48.61878453038674,
      "grad_norm": 1.1097612380981445,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.4601,
      "step": 550
    },
    {
      "epoch": 53.03867403314917,
      "grad_norm": 1.2796109914779663,
      "learning_rate": 3e-05,
      "loss": 0.454,
      "step": 600
    },
    {
      "epoch": 57.4585635359116,
      "grad_norm": 1.0656301975250244,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.4498,
      "step": 650
    },
    {
      "epoch": 61.87845303867403,
      "grad_norm": 1.3971511125564575,
      "learning_rate": 3.5e-05,
      "loss": 0.4439,
      "step": 700
    },
    {
      "epoch": 66.29834254143647,
      "grad_norm": 1.9344899654388428,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.4418,
      "step": 750
    },
    {
      "epoch": 70.7182320441989,
      "grad_norm": 1.080334186553955,
      "learning_rate": 4e-05,
      "loss": 0.4379,
      "step": 800
    },
    {
      "epoch": 75.13812154696133,
      "grad_norm": 1.9891581535339355,
      "learning_rate": 4.25e-05,
      "loss": 0.4334,
      "step": 850
    },
    {
      "epoch": 79.55801104972376,
      "grad_norm": 1.3854402303695679,
      "learning_rate": 4.5e-05,
      "loss": 0.4331,
      "step": 900
    },
    {
      "epoch": 83.97790055248619,
      "grad_norm": 1.5818183422088623,
      "learning_rate": 4.75e-05,
      "loss": 0.428,
      "step": 950
    },
    {
      "epoch": 88.39779005524862,
      "grad_norm": 1.8474106788635254,
      "learning_rate": 5e-05,
      "loss": 0.4238,
      "step": 1000
    },
    {
      "epoch": 88.39779005524862,
      "eval_loss": 0.4078540802001953,
      "eval_runtime": 14.8436,
      "eval_samples_per_second": 10.846,
      "eval_steps_per_second": 1.415,
      "step": 1000
    },
    {
      "epoch": 92.81767955801105,
      "grad_norm": 3.1851751804351807,
      "learning_rate": 5.25e-05,
      "loss": 0.4234,
      "step": 1050
    },
    {
      "epoch": 97.23756906077348,
      "grad_norm": 1.3754140138626099,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.4209,
      "step": 1100
    },
    {
      "epoch": 101.65745856353591,
      "grad_norm": 1.6143475770950317,
      "learning_rate": 5.7499999999999995e-05,
      "loss": 0.4228,
      "step": 1150
    },
    {
      "epoch": 106.07734806629834,
      "grad_norm": 1.9961720705032349,
      "learning_rate": 6e-05,
      "loss": 0.4162,
      "step": 1200
    },
    {
      "epoch": 110.49723756906077,
      "grad_norm": 1.9125347137451172,
      "learning_rate": 6.25e-05,
      "loss": 0.4178,
      "step": 1250
    },
    {
      "epoch": 114.9171270718232,
      "grad_norm": 3.1913959980010986,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.4143,
      "step": 1300
    },
    {
      "epoch": 119.33701657458563,
      "grad_norm": 2.1010069847106934,
      "learning_rate": 6.750000000000001e-05,
      "loss": 0.4118,
      "step": 1350
    },
    {
      "epoch": 123.75690607734806,
      "grad_norm": 1.3079431056976318,
      "learning_rate": 7e-05,
      "loss": 0.4091,
      "step": 1400
    },
    {
      "epoch": 128.1767955801105,
      "grad_norm": 2.3663132190704346,
      "learning_rate": 7.25e-05,
      "loss": 0.4092,
      "step": 1450
    },
    {
      "epoch": 132.59668508287294,
      "grad_norm": 1.031698226928711,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.4083,
      "step": 1500
    },
    {
      "epoch": 132.59668508287294,
      "eval_loss": 0.4082767069339752,
      "eval_runtime": 15.1068,
      "eval_samples_per_second": 10.657,
      "eval_steps_per_second": 1.39,
      "step": 1500
    },
    {
      "epoch": 137.01657458563537,
      "grad_norm": 2.7781426906585693,
      "learning_rate": 7.75e-05,
      "loss": 0.4079,
      "step": 1550
    },
    {
      "epoch": 141.4364640883978,
      "grad_norm": 1.136169672012329,
      "learning_rate": 8e-05,
      "loss": 0.4028,
      "step": 1600
    },
    {
      "epoch": 145.85635359116023,
      "grad_norm": 1.7299158573150635,
      "learning_rate": 8.25e-05,
      "loss": 0.4034,
      "step": 1650
    },
    {
      "epoch": 150.27624309392266,
      "grad_norm": 2.4281983375549316,
      "learning_rate": 8.5e-05,
      "loss": 0.3992,
      "step": 1700
    },
    {
      "epoch": 154.6961325966851,
      "grad_norm": 3.6834373474121094,
      "learning_rate": 8.75e-05,
      "loss": 0.3997,
      "step": 1750
    },
    {
      "epoch": 159.11602209944752,
      "grad_norm": 3.1484293937683105,
      "learning_rate": 9e-05,
      "loss": 0.396,
      "step": 1800
    },
    {
      "epoch": 163.53591160220995,
      "grad_norm": 3.978259563446045,
      "learning_rate": 9.250000000000001e-05,
      "loss": 0.3969,
      "step": 1850
    },
    {
      "epoch": 167.95580110497238,
      "grad_norm": 2.906266212463379,
      "learning_rate": 9.5e-05,
      "loss": 0.4017,
      "step": 1900
    },
    {
      "epoch": 172.3756906077348,
      "grad_norm": 1.1569880247116089,
      "learning_rate": 9.75e-05,
      "loss": 0.396,
      "step": 1950
    },
    {
      "epoch": 176.79558011049724,
      "grad_norm": 2.2872982025146484,
      "learning_rate": 0.0001,
      "loss": 0.3928,
      "step": 2000
    },
    {
      "epoch": 176.79558011049724,
      "eval_loss": 0.414584219455719,
      "eval_runtime": 15.2895,
      "eval_samples_per_second": 10.53,
      "eval_steps_per_second": 1.373,
      "step": 2000
    },
    {
      "epoch": 181.21546961325967,
      "grad_norm": 1.1768990755081177,
      "learning_rate": 9.916666666666667e-05,
      "loss": 0.3928,
      "step": 2050
    },
    {
      "epoch": 185.6353591160221,
      "grad_norm": 1.2087491750717163,
      "learning_rate": 9.833333333333333e-05,
      "loss": 0.3901,
      "step": 2100
    },
    {
      "epoch": 190.05524861878453,
      "grad_norm": 1.17588210105896,
      "learning_rate": 9.75e-05,
      "loss": 0.3877,
      "step": 2150
    },
    {
      "epoch": 194.47513812154696,
      "grad_norm": 1.5979018211364746,
      "learning_rate": 9.666666666666667e-05,
      "loss": 0.3879,
      "step": 2200
    },
    {
      "epoch": 198.8950276243094,
      "grad_norm": 1.0489567518234253,
      "learning_rate": 9.583333333333334e-05,
      "loss": 0.3869,
      "step": 2250
    },
    {
      "epoch": 203.31491712707182,
      "grad_norm": 1.833038091659546,
      "learning_rate": 9.501666666666668e-05,
      "loss": 0.3859,
      "step": 2300
    },
    {
      "epoch": 207.73480662983425,
      "grad_norm": 2.075849771499634,
      "learning_rate": 9.418333333333334e-05,
      "loss": 0.3843,
      "step": 2350
    },
    {
      "epoch": 212.15469613259668,
      "grad_norm": 2.8613317012786865,
      "learning_rate": 9.335e-05,
      "loss": 0.381,
      "step": 2400
    },
    {
      "epoch": 216.5745856353591,
      "grad_norm": 1.6291433572769165,
      "learning_rate": 9.251666666666667e-05,
      "loss": 0.3802,
      "step": 2450
    },
    {
      "epoch": 220.99447513812154,
      "grad_norm": 1.1548006534576416,
      "learning_rate": 9.168333333333333e-05,
      "loss": 0.381,
      "step": 2500
    },
    {
      "epoch": 220.99447513812154,
      "eval_loss": 0.4127160906791687,
      "eval_runtime": 15.1026,
      "eval_samples_per_second": 10.66,
      "eval_steps_per_second": 1.39,
      "step": 2500
    },
    {
      "epoch": 225.41436464088397,
      "grad_norm": 1.3956973552703857,
      "learning_rate": 9.085e-05,
      "loss": 0.3795,
      "step": 2550
    },
    {
      "epoch": 229.8342541436464,
      "grad_norm": 1.6613599061965942,
      "learning_rate": 9.001666666666667e-05,
      "loss": 0.3757,
      "step": 2600
    },
    {
      "epoch": 234.25414364640883,
      "grad_norm": 1.4951831102371216,
      "learning_rate": 8.918333333333334e-05,
      "loss": 0.3763,
      "step": 2650
    },
    {
      "epoch": 238.67403314917127,
      "grad_norm": 3.0197391510009766,
      "learning_rate": 8.834999999999999e-05,
      "loss": 0.3765,
      "step": 2700
    },
    {
      "epoch": 243.0939226519337,
      "grad_norm": 0.9016386866569519,
      "learning_rate": 8.751666666666668e-05,
      "loss": 0.3724,
      "step": 2750
    },
    {
      "epoch": 247.51381215469613,
      "grad_norm": 1.6974703073501587,
      "learning_rate": 8.668333333333334e-05,
      "loss": 0.3729,
      "step": 2800
    },
    {
      "epoch": 251.93370165745856,
      "grad_norm": 2.8571081161499023,
      "learning_rate": 8.585000000000001e-05,
      "loss": 0.3715,
      "step": 2850
    },
    {
      "epoch": 256.353591160221,
      "grad_norm": 2.8940412998199463,
      "learning_rate": 8.501666666666667e-05,
      "loss": 0.3716,
      "step": 2900
    },
    {
      "epoch": 260.77348066298345,
      "grad_norm": 1.1075533628463745,
      "learning_rate": 8.418333333333334e-05,
      "loss": 0.3691,
      "step": 2950
    },
    {
      "epoch": 265.1933701657459,
      "grad_norm": 1.145349383354187,
      "learning_rate": 8.335e-05,
      "loss": 0.3712,
      "step": 3000
    },
    {
      "epoch": 265.1933701657459,
      "eval_loss": 0.40815719962120056,
      "eval_runtime": 15.0404,
      "eval_samples_per_second": 10.705,
      "eval_steps_per_second": 1.396,
      "step": 3000
    },
    {
      "epoch": 269.6132596685083,
      "grad_norm": 0.788912832736969,
      "learning_rate": 8.251666666666668e-05,
      "loss": 0.3691,
      "step": 3050
    },
    {
      "epoch": 274.03314917127074,
      "grad_norm": 1.5521596670150757,
      "learning_rate": 8.168333333333333e-05,
      "loss": 0.3687,
      "step": 3100
    },
    {
      "epoch": 278.45303867403317,
      "grad_norm": 1.2638827562332153,
      "learning_rate": 8.085e-05,
      "loss": 0.3681,
      "step": 3150
    },
    {
      "epoch": 282.8729281767956,
      "grad_norm": 1.6849925518035889,
      "learning_rate": 8.001666666666667e-05,
      "loss": 0.3664,
      "step": 3200
    },
    {
      "epoch": 287.292817679558,
      "grad_norm": 1.9946844577789307,
      "learning_rate": 7.918333333333334e-05,
      "loss": 0.3671,
      "step": 3250
    },
    {
      "epoch": 291.71270718232046,
      "grad_norm": 1.3560407161712646,
      "learning_rate": 7.835000000000001e-05,
      "loss": 0.3631,
      "step": 3300
    },
    {
      "epoch": 296.1325966850829,
      "grad_norm": 1.4300700426101685,
      "learning_rate": 7.751666666666668e-05,
      "loss": 0.3616,
      "step": 3350
    },
    {
      "epoch": 300.5524861878453,
      "grad_norm": 1.0440477132797241,
      "learning_rate": 7.668333333333335e-05,
      "loss": 0.3627,
      "step": 3400
    },
    {
      "epoch": 304.97237569060775,
      "grad_norm": 0.78243088722229,
      "learning_rate": 7.585e-05,
      "loss": 0.3624,
      "step": 3450
    },
    {
      "epoch": 309.3922651933702,
      "grad_norm": 1.4992115497589111,
      "learning_rate": 7.501666666666667e-05,
      "loss": 0.3589,
      "step": 3500
    },
    {
      "epoch": 309.3922651933702,
      "eval_loss": 0.4111907482147217,
      "eval_runtime": 15.163,
      "eval_samples_per_second": 10.618,
      "eval_steps_per_second": 1.385,
      "step": 3500
    },
    {
      "epoch": 313.8121546961326,
      "grad_norm": 1.0604934692382812,
      "learning_rate": 7.418333333333334e-05,
      "loss": 0.3599,
      "step": 3550
    },
    {
      "epoch": 318.23204419889504,
      "grad_norm": 0.8896504640579224,
      "learning_rate": 7.335000000000001e-05,
      "loss": 0.361,
      "step": 3600
    },
    {
      "epoch": 322.65193370165747,
      "grad_norm": 1.3247913122177124,
      "learning_rate": 7.251666666666666e-05,
      "loss": 0.3575,
      "step": 3650
    },
    {
      "epoch": 327.0718232044199,
      "grad_norm": 0.6475388407707214,
      "learning_rate": 7.168333333333333e-05,
      "loss": 0.3575,
      "step": 3700
    },
    {
      "epoch": 331.49171270718233,
      "grad_norm": 0.6724014282226562,
      "learning_rate": 7.085e-05,
      "loss": 0.3558,
      "step": 3750
    },
    {
      "epoch": 335.91160220994476,
      "grad_norm": 1.0705429315567017,
      "learning_rate": 7.001666666666667e-05,
      "loss": 0.3564,
      "step": 3800
    },
    {
      "epoch": 340.3314917127072,
      "grad_norm": 1.0692503452301025,
      "learning_rate": 6.918333333333334e-05,
      "loss": 0.3601,
      "step": 3850
    },
    {
      "epoch": 344.7513812154696,
      "grad_norm": 0.8828334212303162,
      "learning_rate": 6.835000000000001e-05,
      "loss": 0.3548,
      "step": 3900
    },
    {
      "epoch": 349.17127071823205,
      "grad_norm": 0.6390587091445923,
      "learning_rate": 6.751666666666668e-05,
      "loss": 0.3529,
      "step": 3950
    },
    {
      "epoch": 353.5911602209945,
      "grad_norm": 0.7502044439315796,
      "learning_rate": 6.668333333333333e-05,
      "loss": 0.3544,
      "step": 4000
    },
    {
      "epoch": 353.5911602209945,
      "eval_loss": 0.4128935635089874,
      "eval_runtime": 16.0241,
      "eval_samples_per_second": 10.047,
      "eval_steps_per_second": 1.311,
      "step": 4000
    },
    {
      "epoch": 358.0110497237569,
      "grad_norm": 0.8119373917579651,
      "learning_rate": 6.585e-05,
      "loss": 0.3527,
      "step": 4050
    },
    {
      "epoch": 362.43093922651934,
      "grad_norm": 0.7546930313110352,
      "learning_rate": 6.501666666666667e-05,
      "loss": 0.352,
      "step": 4100
    },
    {
      "epoch": 366.8508287292818,
      "grad_norm": 1.1060476303100586,
      "learning_rate": 6.418333333333334e-05,
      "loss": 0.3514,
      "step": 4150
    },
    {
      "epoch": 371.2707182320442,
      "grad_norm": 1.3087377548217773,
      "learning_rate": 6.335e-05,
      "loss": 0.3508,
      "step": 4200
    },
    {
      "epoch": 375.69060773480663,
      "grad_norm": 0.7953481078147888,
      "learning_rate": 6.251666666666666e-05,
      "loss": 0.3493,
      "step": 4250
    },
    {
      "epoch": 380.11049723756906,
      "grad_norm": 1.1858787536621094,
      "learning_rate": 6.168333333333333e-05,
      "loss": 0.3493,
      "step": 4300
    },
    {
      "epoch": 384.5303867403315,
      "grad_norm": 0.9929729700088501,
      "learning_rate": 6.085000000000001e-05,
      "loss": 0.3485,
      "step": 4350
    },
    {
      "epoch": 388.9502762430939,
      "grad_norm": 0.8156938552856445,
      "learning_rate": 6.0016666666666664e-05,
      "loss": 0.347,
      "step": 4400
    },
    {
      "epoch": 393.37016574585635,
      "grad_norm": 1.0693029165267944,
      "learning_rate": 5.918333333333333e-05,
      "loss": 0.3489,
      "step": 4450
    },
    {
      "epoch": 397.7900552486188,
      "grad_norm": 1.1495543718338013,
      "learning_rate": 5.835e-05,
      "loss": 0.347,
      "step": 4500
    },
    {
      "epoch": 397.7900552486188,
      "eval_loss": 0.41448456048965454,
      "eval_runtime": 15.2027,
      "eval_samples_per_second": 10.59,
      "eval_steps_per_second": 1.381,
      "step": 4500
    },
    {
      "epoch": 402.2099447513812,
      "grad_norm": 0.9324333071708679,
      "learning_rate": 5.751666666666667e-05,
      "loss": 0.3473,
      "step": 4550
    },
    {
      "epoch": 406.62983425414365,
      "grad_norm": 0.6897173523902893,
      "learning_rate": 5.668333333333333e-05,
      "loss": 0.3441,
      "step": 4600
    },
    {
      "epoch": 411.0497237569061,
      "grad_norm": 1.1474602222442627,
      "learning_rate": 5.585e-05,
      "loss": 0.346,
      "step": 4650
    },
    {
      "epoch": 415.4696132596685,
      "grad_norm": 1.4295645952224731,
      "learning_rate": 5.501666666666667e-05,
      "loss": 0.3442,
      "step": 4700
    },
    {
      "epoch": 419.88950276243094,
      "grad_norm": 0.8301011323928833,
      "learning_rate": 5.4183333333333334e-05,
      "loss": 0.344,
      "step": 4750
    },
    {
      "epoch": 424.30939226519337,
      "grad_norm": 1.2894037961959839,
      "learning_rate": 5.335e-05,
      "loss": 0.3431,
      "step": 4800
    },
    {
      "epoch": 428.7292817679558,
      "grad_norm": 0.6597224473953247,
      "learning_rate": 5.251666666666667e-05,
      "loss": 0.3443,
      "step": 4850
    },
    {
      "epoch": 433.1491712707182,
      "grad_norm": 0.8772273063659668,
      "learning_rate": 5.168333333333334e-05,
      "loss": 0.3447,
      "step": 4900
    },
    {
      "epoch": 437.56906077348066,
      "grad_norm": 0.6241065263748169,
      "learning_rate": 5.0849999999999996e-05,
      "loss": 0.3412,
      "step": 4950
    },
    {
      "epoch": 441.9889502762431,
      "grad_norm": 0.8275649547576904,
      "learning_rate": 5.0016666666666665e-05,
      "loss": 0.3417,
      "step": 5000
    },
    {
      "epoch": 441.9889502762431,
      "eval_loss": 0.42017292976379395,
      "eval_runtime": 15.281,
      "eval_samples_per_second": 10.536,
      "eval_steps_per_second": 1.374,
      "step": 5000
    },
    {
      "epoch": 446.4088397790055,
      "grad_norm": 0.9026231169700623,
      "learning_rate": 4.9183333333333334e-05,
      "loss": 0.3417,
      "step": 5050
    },
    {
      "epoch": 450.82872928176795,
      "grad_norm": 1.248936414718628,
      "learning_rate": 4.835e-05,
      "loss": 0.3396,
      "step": 5100
    },
    {
      "epoch": 455.2486187845304,
      "grad_norm": 1.257866621017456,
      "learning_rate": 4.751666666666667e-05,
      "loss": 0.3414,
      "step": 5150
    },
    {
      "epoch": 459.6685082872928,
      "grad_norm": 1.029221534729004,
      "learning_rate": 4.6683333333333334e-05,
      "loss": 0.3396,
      "step": 5200
    },
    {
      "epoch": 464.08839779005524,
      "grad_norm": 1.1000345945358276,
      "learning_rate": 4.585e-05,
      "loss": 0.3392,
      "step": 5250
    },
    {
      "epoch": 468.50828729281767,
      "grad_norm": 0.7120194435119629,
      "learning_rate": 4.5016666666666665e-05,
      "loss": 0.3412,
      "step": 5300
    },
    {
      "epoch": 472.9281767955801,
      "grad_norm": 0.5869295597076416,
      "learning_rate": 4.4183333333333334e-05,
      "loss": 0.3398,
      "step": 5350
    },
    {
      "epoch": 477.34806629834253,
      "grad_norm": 0.8485609292984009,
      "learning_rate": 4.335e-05,
      "loss": 0.3385,
      "step": 5400
    },
    {
      "epoch": 481.76795580110496,
      "grad_norm": 0.520220160484314,
      "learning_rate": 4.251666666666667e-05,
      "loss": 0.3366,
      "step": 5450
    },
    {
      "epoch": 486.1878453038674,
      "grad_norm": 0.6742538809776306,
      "learning_rate": 4.1683333333333335e-05,
      "loss": 0.3372,
      "step": 5500
    },
    {
      "epoch": 486.1878453038674,
      "eval_loss": 0.41964003443717957,
      "eval_runtime": 16.3724,
      "eval_samples_per_second": 9.834,
      "eval_steps_per_second": 1.283,
      "step": 5500
    },
    {
      "epoch": 490.6077348066298,
      "grad_norm": 0.8028594255447388,
      "learning_rate": 4.085e-05,
      "loss": 0.3378,
      "step": 5550
    },
    {
      "epoch": 495.02762430939225,
      "grad_norm": 0.6300519704818726,
      "learning_rate": 4.0016666666666666e-05,
      "loss": 0.3373,
      "step": 5600
    },
    {
      "epoch": 499.4475138121547,
      "grad_norm": 0.6699241399765015,
      "learning_rate": 3.9183333333333335e-05,
      "loss": 0.3365,
      "step": 5650
    },
    {
      "epoch": 503.8674033149171,
      "grad_norm": 0.6375143527984619,
      "learning_rate": 3.8350000000000004e-05,
      "loss": 0.3358,
      "step": 5700
    },
    {
      "epoch": 508.28729281767954,
      "grad_norm": 0.6259489059448242,
      "learning_rate": 3.7516666666666666e-05,
      "loss": 0.3363,
      "step": 5750
    },
    {
      "epoch": 512.707182320442,
      "grad_norm": 0.5885080695152283,
      "learning_rate": 3.6683333333333335e-05,
      "loss": 0.3353,
      "step": 5800
    },
    {
      "epoch": 517.1270718232045,
      "grad_norm": 0.6459413766860962,
      "learning_rate": 3.585e-05,
      "loss": 0.3344,
      "step": 5850
    },
    {
      "epoch": 521.5469613259669,
      "grad_norm": 0.6893390417098999,
      "learning_rate": 3.501666666666667e-05,
      "loss": 0.3353,
      "step": 5900
    },
    {
      "epoch": 525.9668508287293,
      "grad_norm": 0.644479513168335,
      "learning_rate": 3.4183333333333335e-05,
      "loss": 0.3354,
      "step": 5950
    },
    {
      "epoch": 530.3867403314918,
      "grad_norm": 0.5544880628585815,
      "learning_rate": 3.3350000000000004e-05,
      "loss": 0.3335,
      "step": 6000
    },
    {
      "epoch": 530.3867403314918,
      "eval_loss": 0.4183831810951233,
      "eval_runtime": 15.2687,
      "eval_samples_per_second": 10.544,
      "eval_steps_per_second": 1.375,
      "step": 6000
    },
    {
      "epoch": 534.8066298342542,
      "grad_norm": 0.6328197717666626,
      "learning_rate": 3.2516666666666666e-05,
      "loss": 0.3333,
      "step": 6050
    },
    {
      "epoch": 539.2265193370166,
      "grad_norm": 0.6934103965759277,
      "learning_rate": 3.1683333333333335e-05,
      "loss": 0.3376,
      "step": 6100
    },
    {
      "epoch": 543.646408839779,
      "grad_norm": 0.659902036190033,
      "learning_rate": 3.0850000000000004e-05,
      "loss": 0.3343,
      "step": 6150
    },
    {
      "epoch": 548.0662983425415,
      "grad_norm": 0.6404502391815186,
      "learning_rate": 3.001666666666667e-05,
      "loss": 0.3343,
      "step": 6200
    },
    {
      "epoch": 552.4861878453039,
      "grad_norm": 0.6743395924568176,
      "learning_rate": 2.9183333333333336e-05,
      "loss": 0.3347,
      "step": 6250
    },
    {
      "epoch": 556.9060773480663,
      "grad_norm": 0.9876683950424194,
      "learning_rate": 2.8349999999999998e-05,
      "loss": 0.3324,
      "step": 6300
    },
    {
      "epoch": 561.3259668508288,
      "grad_norm": 0.7507224678993225,
      "learning_rate": 2.7516666666666667e-05,
      "loss": 0.3318,
      "step": 6350
    },
    {
      "epoch": 565.7458563535912,
      "grad_norm": 0.7246477603912354,
      "learning_rate": 2.6683333333333333e-05,
      "loss": 0.3299,
      "step": 6400
    },
    {
      "epoch": 570.1657458563536,
      "grad_norm": 1.1144418716430664,
      "learning_rate": 2.585e-05,
      "loss": 0.3309,
      "step": 6450
    },
    {
      "epoch": 574.585635359116,
      "grad_norm": 0.5841721892356873,
      "learning_rate": 2.5016666666666667e-05,
      "loss": 0.3316,
      "step": 6500
    },
    {
      "epoch": 574.585635359116,
      "eval_loss": 0.42165789008140564,
      "eval_runtime": 16.1022,
      "eval_samples_per_second": 9.999,
      "eval_steps_per_second": 1.304,
      "step": 6500
    },
    {
      "epoch": 579.0055248618785,
      "grad_norm": 0.5832745432853699,
      "learning_rate": 2.4183333333333336e-05,
      "loss": 0.3304,
      "step": 6550
    },
    {
      "epoch": 583.4254143646409,
      "grad_norm": 0.5167184472084045,
      "learning_rate": 2.3350000000000002e-05,
      "loss": 0.3325,
      "step": 6600
    },
    {
      "epoch": 587.8453038674033,
      "grad_norm": 0.5355780124664307,
      "learning_rate": 2.2516666666666667e-05,
      "loss": 0.3302,
      "step": 6650
    },
    {
      "epoch": 592.2651933701658,
      "grad_norm": 0.5522999167442322,
      "learning_rate": 2.1683333333333333e-05,
      "loss": 0.3296,
      "step": 6700
    },
    {
      "epoch": 596.6850828729282,
      "grad_norm": 0.7138186097145081,
      "learning_rate": 2.085e-05,
      "loss": 0.3301,
      "step": 6750
    },
    {
      "epoch": 601.1049723756906,
      "grad_norm": 1.6618551015853882,
      "learning_rate": 2.0016666666666668e-05,
      "loss": 0.3324,
      "step": 6800
    },
    {
      "epoch": 605.5248618784531,
      "grad_norm": 0.5868383646011353,
      "learning_rate": 1.9183333333333333e-05,
      "loss": 0.327,
      "step": 6850
    },
    {
      "epoch": 609.9447513812155,
      "grad_norm": 0.6649237871170044,
      "learning_rate": 1.8350000000000002e-05,
      "loss": 0.3309,
      "step": 6900
    },
    {
      "epoch": 614.3646408839779,
      "grad_norm": 0.8205301761627197,
      "learning_rate": 1.7516666666666668e-05,
      "loss": 0.3307,
      "step": 6950
    },
    {
      "epoch": 618.7845303867404,
      "grad_norm": 0.48336291313171387,
      "learning_rate": 1.6683333333333333e-05,
      "loss": 0.3291,
      "step": 7000
    },
    {
      "epoch": 618.7845303867404,
      "eval_loss": 0.42143845558166504,
      "eval_runtime": 15.392,
      "eval_samples_per_second": 10.46,
      "eval_steps_per_second": 1.364,
      "step": 7000
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 728,
  "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": 1.585209327244028e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}