{
  "best_metric": 0.3926040828227997,
  "best_model_checkpoint": "mikhail-panzo/zlm-fil-ceb_b64_le5_s8000/checkpoint-6500",
  "epoch": 316.83168316831683,
  "eval_steps": 500,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 1.2683665752410889,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.4932,
      "step": 50
    },
    {
      "epoch": 3.9603960396039604,
      "grad_norm": 1.1266463994979858,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4898,
      "step": 100
    },
    {
      "epoch": 5.9405940594059405,
      "grad_norm": 1.1800968647003174,
      "learning_rate": 7.5e-07,
      "loss": 0.4839,
      "step": 150
    },
    {
      "epoch": 7.920792079207921,
      "grad_norm": 0.8235568404197693,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.4785,
      "step": 200
    },
    {
      "epoch": 9.900990099009901,
      "grad_norm": 1.313211441040039,
      "learning_rate": 1.25e-06,
      "loss": 0.4767,
      "step": 250
    },
    {
      "epoch": 11.881188118811881,
      "grad_norm": 0.7831560373306274,
      "learning_rate": 1.5e-06,
      "loss": 0.4681,
      "step": 300
    },
    {
      "epoch": 13.861386138613861,
      "grad_norm": 0.7987237572669983,
      "learning_rate": 1.75e-06,
      "loss": 0.4658,
      "step": 350
    },
    {
      "epoch": 15.841584158415841,
      "grad_norm": 0.7143813371658325,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.4627,
      "step": 400
    },
    {
      "epoch": 17.821782178217823,
      "grad_norm": 0.8037531971931458,
      "learning_rate": 2.25e-06,
      "loss": 0.461,
      "step": 450
    },
    {
      "epoch": 19.801980198019802,
      "grad_norm": 0.79031902551651,
      "learning_rate": 2.5e-06,
      "loss": 0.4592,
      "step": 500
    },
    {
      "epoch": 19.801980198019802,
      "eval_loss": 0.42528408765792847,
      "eval_runtime": 7.8982,
      "eval_samples_per_second": 22.79,
      "eval_steps_per_second": 2.912,
      "step": 500
    },
    {
      "epoch": 21.782178217821784,
      "grad_norm": 0.7966288328170776,
      "learning_rate": 2.7500000000000004e-06,
      "loss": 0.4551,
      "step": 550
    },
    {
      "epoch": 23.762376237623762,
      "grad_norm": 0.8126183152198792,
      "learning_rate": 3e-06,
      "loss": 0.4542,
      "step": 600
    },
    {
      "epoch": 25.742574257425744,
      "grad_norm": 0.6602805852890015,
      "learning_rate": 3.2500000000000002e-06,
      "loss": 0.4535,
      "step": 650
    },
    {
      "epoch": 27.722772277227723,
      "grad_norm": 0.7177000641822815,
      "learning_rate": 3.5e-06,
      "loss": 0.4484,
      "step": 700
    },
    {
      "epoch": 29.702970297029704,
      "grad_norm": 0.7912996411323547,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.4486,
      "step": 750
    },
    {
      "epoch": 31.683168316831683,
      "grad_norm": 0.6251747608184814,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.4463,
      "step": 800
    },
    {
      "epoch": 33.663366336633665,
      "grad_norm": 1.003780722618103,
      "learning_rate": 4.25e-06,
      "loss": 0.4474,
      "step": 850
    },
    {
      "epoch": 35.64356435643565,
      "grad_norm": 0.742171585559845,
      "learning_rate": 4.5e-06,
      "loss": 0.4434,
      "step": 900
    },
    {
      "epoch": 37.62376237623762,
      "grad_norm": 0.6034647822380066,
      "learning_rate": 4.75e-06,
      "loss": 0.4425,
      "step": 950
    },
    {
      "epoch": 39.603960396039604,
      "grad_norm": 0.6990635395050049,
      "learning_rate": 5e-06,
      "loss": 0.4381,
      "step": 1000
    },
    {
      "epoch": 39.603960396039604,
      "eval_loss": 0.41002073884010315,
      "eval_runtime": 6.7782,
      "eval_samples_per_second": 26.556,
      "eval_steps_per_second": 3.393,
      "step": 1000
    },
    {
      "epoch": 41.584158415841586,
      "grad_norm": 0.9420008659362793,
      "learning_rate": 5.2500000000000006e-06,
      "loss": 0.4421,
      "step": 1050
    },
    {
      "epoch": 43.56435643564357,
      "grad_norm": 0.7143107652664185,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.4389,
      "step": 1100
    },
    {
      "epoch": 45.54455445544554,
      "grad_norm": 0.7934616804122925,
      "learning_rate": 5.75e-06,
      "loss": 0.4369,
      "step": 1150
    },
    {
      "epoch": 47.524752475247524,
      "grad_norm": 0.9682563543319702,
      "learning_rate": 6e-06,
      "loss": 0.4355,
      "step": 1200
    },
    {
      "epoch": 49.504950495049506,
      "grad_norm": 0.7880070209503174,
      "learning_rate": 6.25e-06,
      "loss": 0.4359,
      "step": 1250
    },
    {
      "epoch": 51.48514851485149,
      "grad_norm": 0.8443828821182251,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.4324,
      "step": 1300
    },
    {
      "epoch": 53.46534653465346,
      "grad_norm": 0.724743127822876,
      "learning_rate": 6.750000000000001e-06,
      "loss": 0.4348,
      "step": 1350
    },
    {
      "epoch": 55.445544554455445,
      "grad_norm": 0.7557791471481323,
      "learning_rate": 7e-06,
      "loss": 0.433,
      "step": 1400
    },
    {
      "epoch": 57.42574257425743,
      "grad_norm": 0.7936097383499146,
      "learning_rate": 7.25e-06,
      "loss": 0.4273,
      "step": 1450
    },
    {
      "epoch": 59.40594059405941,
      "grad_norm": 0.7041624784469604,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.4281,
      "step": 1500
    },
    {
      "epoch": 59.40594059405941,
      "eval_loss": 0.40218469500541687,
      "eval_runtime": 7.4309,
      "eval_samples_per_second": 24.223,
      "eval_steps_per_second": 3.095,
      "step": 1500
    },
    {
      "epoch": 61.386138613861384,
      "grad_norm": 0.9194286465644836,
      "learning_rate": 7.75e-06,
      "loss": 0.4275,
      "step": 1550
    },
    {
      "epoch": 63.366336633663366,
      "grad_norm": 0.7239477634429932,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.4245,
      "step": 1600
    },
    {
      "epoch": 65.34653465346534,
      "grad_norm": 0.7235706448554993,
      "learning_rate": 8.25e-06,
      "loss": 0.4248,
      "step": 1650
    },
    {
      "epoch": 67.32673267326733,
      "grad_norm": 0.7281640768051147,
      "learning_rate": 8.5e-06,
      "loss": 0.425,
      "step": 1700
    },
    {
      "epoch": 69.3069306930693,
      "grad_norm": 1.011294960975647,
      "learning_rate": 8.750000000000001e-06,
      "loss": 0.4238,
      "step": 1750
    },
    {
      "epoch": 71.2871287128713,
      "grad_norm": 1.1249107122421265,
      "learning_rate": 9e-06,
      "loss": 0.4203,
      "step": 1800
    },
    {
      "epoch": 73.26732673267327,
      "grad_norm": 0.7369679808616638,
      "learning_rate": 9.250000000000001e-06,
      "loss": 0.422,
      "step": 1850
    },
    {
      "epoch": 75.24752475247524,
      "grad_norm": 0.704367995262146,
      "learning_rate": 9.5e-06,
      "loss": 0.4201,
      "step": 1900
    },
    {
      "epoch": 77.22772277227723,
      "grad_norm": 0.8307437896728516,
      "learning_rate": 9.75e-06,
      "loss": 0.4203,
      "step": 1950
    },
    {
      "epoch": 79.20792079207921,
      "grad_norm": 0.7827133536338806,
      "learning_rate": 1e-05,
      "loss": 0.4195,
      "step": 2000
    },
    {
      "epoch": 79.20792079207921,
      "eval_loss": 0.39878609776496887,
      "eval_runtime": 8.1579,
      "eval_samples_per_second": 22.064,
      "eval_steps_per_second": 2.819,
      "step": 2000
    },
    {
      "epoch": 81.18811881188118,
      "grad_norm": 0.8909313082695007,
      "learning_rate": 9.916666666666668e-06,
      "loss": 0.4203,
      "step": 2050
    },
    {
      "epoch": 83.16831683168317,
      "grad_norm": 0.7147483825683594,
      "learning_rate": 9.833333333333333e-06,
      "loss": 0.418,
      "step": 2100
    },
    {
      "epoch": 85.14851485148515,
      "grad_norm": 0.6844545006752014,
      "learning_rate": 9.75e-06,
      "loss": 0.4167,
      "step": 2150
    },
    {
      "epoch": 87.12871287128714,
      "grad_norm": 0.88512122631073,
      "learning_rate": 9.666666666666667e-06,
      "loss": 0.4108,
      "step": 2200
    },
    {
      "epoch": 89.10891089108911,
      "grad_norm": 0.7097252607345581,
      "learning_rate": 9.583333333333335e-06,
      "loss": 0.4135,
      "step": 2250
    },
    {
      "epoch": 91.08910891089108,
      "grad_norm": 0.8400272727012634,
      "learning_rate": 9.5e-06,
      "loss": 0.4147,
      "step": 2300
    },
    {
      "epoch": 93.06930693069307,
      "grad_norm": 0.7614939212799072,
      "learning_rate": 9.416666666666667e-06,
      "loss": 0.4119,
      "step": 2350
    },
    {
      "epoch": 95.04950495049505,
      "grad_norm": 0.8099732995033264,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.4108,
      "step": 2400
    },
    {
      "epoch": 97.02970297029702,
      "grad_norm": 1.0697815418243408,
      "learning_rate": 9.250000000000001e-06,
      "loss": 0.4095,
      "step": 2450
    },
    {
      "epoch": 99.00990099009901,
      "grad_norm": 0.7235235571861267,
      "learning_rate": 9.166666666666666e-06,
      "loss": 0.4134,
      "step": 2500
    },
    {
      "epoch": 99.00990099009901,
      "eval_loss": 0.39545077085494995,
      "eval_runtime": 7.724,
      "eval_samples_per_second": 23.304,
      "eval_steps_per_second": 2.978,
      "step": 2500
    },
    {
      "epoch": 100.99009900990099,
      "grad_norm": 0.6166990399360657,
      "learning_rate": 9.083333333333333e-06,
      "loss": 0.4097,
      "step": 2550
    },
    {
      "epoch": 102.97029702970298,
      "grad_norm": 0.6944295763969421,
      "learning_rate": 9e-06,
      "loss": 0.4107,
      "step": 2600
    },
    {
      "epoch": 104.95049504950495,
      "grad_norm": 0.8469858765602112,
      "learning_rate": 8.916666666666667e-06,
      "loss": 0.4056,
      "step": 2650
    },
    {
      "epoch": 106.93069306930693,
      "grad_norm": 0.6261239647865295,
      "learning_rate": 8.833333333333334e-06,
      "loss": 0.409,
      "step": 2700
    },
    {
      "epoch": 108.91089108910892,
      "grad_norm": 0.6304221749305725,
      "learning_rate": 8.750000000000001e-06,
      "loss": 0.4081,
      "step": 2750
    },
    {
      "epoch": 110.89108910891089,
      "grad_norm": 0.6513387560844421,
      "learning_rate": 8.666666666666668e-06,
      "loss": 0.4071,
      "step": 2800
    },
    {
      "epoch": 112.87128712871286,
      "grad_norm": 0.9311261177062988,
      "learning_rate": 8.583333333333333e-06,
      "loss": 0.4091,
      "step": 2850
    },
    {
      "epoch": 114.85148514851485,
      "grad_norm": 0.7296612858772278,
      "learning_rate": 8.5e-06,
      "loss": 0.4055,
      "step": 2900
    },
    {
      "epoch": 116.83168316831683,
      "grad_norm": 0.8149676322937012,
      "learning_rate": 8.416666666666667e-06,
      "loss": 0.4052,
      "step": 2950
    },
    {
      "epoch": 118.81188118811882,
      "grad_norm": 0.8266366124153137,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.4049,
      "step": 3000
    },
    {
      "epoch": 118.81188118811882,
      "eval_loss": 0.39345207810401917,
      "eval_runtime": 7.2151,
      "eval_samples_per_second": 24.948,
      "eval_steps_per_second": 3.188,
      "step": 3000
    },
    {
      "epoch": 120.79207920792079,
      "grad_norm": 0.6681864261627197,
      "learning_rate": 8.25e-06,
      "loss": 0.4022,
      "step": 3050
    },
    {
      "epoch": 122.77227722772277,
      "grad_norm": 0.5798916220664978,
      "learning_rate": 8.166666666666668e-06,
      "loss": 0.4026,
      "step": 3100
    },
    {
      "epoch": 124.75247524752476,
      "grad_norm": 0.8516507744789124,
      "learning_rate": 8.083333333333334e-06,
      "loss": 0.4064,
      "step": 3150
    },
    {
      "epoch": 126.73267326732673,
      "grad_norm": 0.6938304901123047,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.4057,
      "step": 3200
    },
    {
      "epoch": 128.7128712871287,
      "grad_norm": 0.6889196038246155,
      "learning_rate": 7.916666666666667e-06,
      "loss": 0.4022,
      "step": 3250
    },
    {
      "epoch": 130.69306930693068,
      "grad_norm": 0.735680103302002,
      "learning_rate": 7.833333333333333e-06,
      "loss": 0.4019,
      "step": 3300
    },
    {
      "epoch": 132.67326732673268,
      "grad_norm": 1.0390517711639404,
      "learning_rate": 7.75e-06,
      "loss": 0.4014,
      "step": 3350
    },
    {
      "epoch": 134.65346534653466,
      "grad_norm": 0.807412326335907,
      "learning_rate": 7.666666666666667e-06,
      "loss": 0.3989,
      "step": 3400
    },
    {
      "epoch": 136.63366336633663,
      "grad_norm": 0.6928243637084961,
      "learning_rate": 7.583333333333333e-06,
      "loss": 0.4011,
      "step": 3450
    },
    {
      "epoch": 138.6138613861386,
      "grad_norm": 0.6894844770431519,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.4016,
      "step": 3500
    },
    {
      "epoch": 138.6138613861386,
      "eval_loss": 0.3931116759777069,
      "eval_runtime": 7.7493,
      "eval_samples_per_second": 23.228,
      "eval_steps_per_second": 2.968,
      "step": 3500
    },
    {
      "epoch": 140.59405940594058,
      "grad_norm": 0.9431400895118713,
      "learning_rate": 7.416666666666668e-06,
      "loss": 0.4028,
      "step": 3550
    },
    {
      "epoch": 142.5742574257426,
      "grad_norm": 0.6703277230262756,
      "learning_rate": 7.333333333333333e-06,
      "loss": 0.3992,
      "step": 3600
    },
    {
      "epoch": 144.55445544554456,
      "grad_norm": 0.8334361910820007,
      "learning_rate": 7.25e-06,
      "loss": 0.3979,
      "step": 3650
    },
    {
      "epoch": 146.53465346534654,
      "grad_norm": 0.8266313672065735,
      "learning_rate": 7.166666666666667e-06,
      "loss": 0.3964,
      "step": 3700
    },
    {
      "epoch": 148.5148514851485,
      "grad_norm": 0.7688102126121521,
      "learning_rate": 7.083333333333335e-06,
      "loss": 0.3968,
      "step": 3750
    },
    {
      "epoch": 150.4950495049505,
      "grad_norm": 0.7516873478889465,
      "learning_rate": 7e-06,
      "loss": 0.3971,
      "step": 3800
    },
    {
      "epoch": 152.47524752475246,
      "grad_norm": 0.8034970760345459,
      "learning_rate": 6.916666666666667e-06,
      "loss": 0.3926,
      "step": 3850
    },
    {
      "epoch": 154.45544554455446,
      "grad_norm": 0.681676983833313,
      "learning_rate": 6.833333333333334e-06,
      "loss": 0.3975,
      "step": 3900
    },
    {
      "epoch": 156.43564356435644,
      "grad_norm": 0.6939652562141418,
      "learning_rate": 6.750000000000001e-06,
      "loss": 0.4005,
      "step": 3950
    },
    {
      "epoch": 158.41584158415841,
      "grad_norm": 0.6093599200248718,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.3937,
      "step": 4000
    },
    {
      "epoch": 158.41584158415841,
      "eval_loss": 0.39361944794654846,
      "eval_runtime": 7.0786,
      "eval_samples_per_second": 25.429,
      "eval_steps_per_second": 3.249,
      "step": 4000
    },
    {
      "epoch": 160.3960396039604,
      "grad_norm": 0.6037696003913879,
      "learning_rate": 6.5833333333333335e-06,
      "loss": 0.3953,
      "step": 4050
    },
    {
      "epoch": 162.37623762376236,
      "grad_norm": 0.6949267387390137,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.398,
      "step": 4100
    },
    {
      "epoch": 164.35643564356437,
      "grad_norm": 0.7271093726158142,
      "learning_rate": 6.418333333333334e-06,
      "loss": 0.3945,
      "step": 4150
    },
    {
      "epoch": 166.33663366336634,
      "grad_norm": 0.6959238052368164,
      "learning_rate": 6.335e-06,
      "loss": 0.3957,
      "step": 4200
    },
    {
      "epoch": 168.31683168316832,
      "grad_norm": 0.6987124681472778,
      "learning_rate": 6.251666666666667e-06,
      "loss": 0.3973,
      "step": 4250
    },
    {
      "epoch": 170.2970297029703,
      "grad_norm": 0.6888285875320435,
      "learning_rate": 6.168333333333334e-06,
      "loss": 0.3971,
      "step": 4300
    },
    {
      "epoch": 172.27722772277227,
      "grad_norm": 0.6536576151847839,
      "learning_rate": 6.085000000000001e-06,
      "loss": 0.3939,
      "step": 4350
    },
    {
      "epoch": 174.25742574257427,
      "grad_norm": 0.6235485672950745,
      "learning_rate": 6.001666666666667e-06,
      "loss": 0.3949,
      "step": 4400
    },
    {
      "epoch": 176.23762376237624,
      "grad_norm": 0.7131723761558533,
      "learning_rate": 5.918333333333334e-06,
      "loss": 0.3944,
      "step": 4450
    },
    {
      "epoch": 178.21782178217822,
      "grad_norm": 0.5429028272628784,
      "learning_rate": 5.835000000000001e-06,
      "loss": 0.3953,
      "step": 4500
    },
    {
      "epoch": 178.21782178217822,
      "eval_loss": 0.3935137093067169,
      "eval_runtime": 6.8281,
      "eval_samples_per_second": 26.362,
      "eval_steps_per_second": 3.368,
      "step": 4500
    },
    {
      "epoch": 180.1980198019802,
      "grad_norm": 0.6293529272079468,
      "learning_rate": 5.751666666666668e-06,
      "loss": 0.3923,
      "step": 4550
    },
    {
      "epoch": 182.17821782178217,
      "grad_norm": 0.7379683256149292,
      "learning_rate": 5.668333333333334e-06,
      "loss": 0.393,
      "step": 4600
    },
    {
      "epoch": 184.15841584158414,
      "grad_norm": 0.6412554979324341,
      "learning_rate": 5.585000000000001e-06,
      "loss": 0.3915,
      "step": 4650
    },
    {
      "epoch": 186.13861386138615,
      "grad_norm": 0.7281537652015686,
      "learning_rate": 5.501666666666668e-06,
      "loss": 0.3928,
      "step": 4700
    },
    {
      "epoch": 188.11881188118812,
      "grad_norm": 0.640044093132019,
      "learning_rate": 5.418333333333333e-06,
      "loss": 0.3935,
      "step": 4750
    },
    {
      "epoch": 190.0990099009901,
      "grad_norm": 0.6560197472572327,
      "learning_rate": 5.335000000000001e-06,
      "loss": 0.3922,
      "step": 4800
    },
    {
      "epoch": 192.07920792079207,
      "grad_norm": 0.5895935893058777,
      "learning_rate": 5.2516666666666675e-06,
      "loss": 0.3939,
      "step": 4850
    },
    {
      "epoch": 194.05940594059405,
      "grad_norm": 0.7578818202018738,
      "learning_rate": 5.168333333333334e-06,
      "loss": 0.3939,
      "step": 4900
    },
    {
      "epoch": 196.03960396039605,
      "grad_norm": 0.6305426955223083,
      "learning_rate": 5.085e-06,
      "loss": 0.3917,
      "step": 4950
    },
    {
      "epoch": 198.01980198019803,
      "grad_norm": 0.8592019081115723,
      "learning_rate": 5.0016666666666665e-06,
      "loss": 0.3933,
      "step": 5000
    },
    {
      "epoch": 198.01980198019803,
      "eval_loss": 0.39355847239494324,
      "eval_runtime": 7.4596,
      "eval_samples_per_second": 24.13,
      "eval_steps_per_second": 3.083,
      "step": 5000
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.9197565317153931,
      "learning_rate": 4.918333333333334e-06,
      "loss": 0.3947,
      "step": 5050
    },
    {
      "epoch": 201.98019801980197,
      "grad_norm": 0.675254225730896,
      "learning_rate": 4.835e-06,
      "loss": 0.3921,
      "step": 5100
    },
    {
      "epoch": 203.96039603960395,
      "grad_norm": 0.7896837592124939,
      "learning_rate": 4.751666666666667e-06,
      "loss": 0.3937,
      "step": 5150
    },
    {
      "epoch": 205.94059405940595,
      "grad_norm": 0.6978718042373657,
      "learning_rate": 4.668333333333333e-06,
      "loss": 0.3912,
      "step": 5200
    },
    {
      "epoch": 207.92079207920793,
      "grad_norm": 0.6968542337417603,
      "learning_rate": 4.585e-06,
      "loss": 0.3902,
      "step": 5250
    },
    {
      "epoch": 209.9009900990099,
      "grad_norm": 0.6489793062210083,
      "learning_rate": 4.501666666666667e-06,
      "loss": 0.3911,
      "step": 5300
    },
    {
      "epoch": 211.88118811881188,
      "grad_norm": 0.6600732803344727,
      "learning_rate": 4.418333333333334e-06,
      "loss": 0.3912,
      "step": 5350
    },
    {
      "epoch": 213.86138613861385,
      "grad_norm": 0.6637029051780701,
      "learning_rate": 4.335e-06,
      "loss": 0.3903,
      "step": 5400
    },
    {
      "epoch": 215.84158415841586,
      "grad_norm": 0.9407865405082703,
      "learning_rate": 4.251666666666667e-06,
      "loss": 0.3913,
      "step": 5450
    },
    {
      "epoch": 217.82178217821783,
      "grad_norm": 0.674074649810791,
      "learning_rate": 4.168333333333334e-06,
      "loss": 0.3932,
      "step": 5500
    },
    {
      "epoch": 217.82178217821783,
      "eval_loss": 0.3936135172843933,
      "eval_runtime": 7.411,
      "eval_samples_per_second": 24.288,
      "eval_steps_per_second": 3.103,
      "step": 5500
    },
    {
      "epoch": 219.8019801980198,
      "grad_norm": 0.5686890482902527,
      "learning_rate": 4.085e-06,
      "loss": 0.391,
      "step": 5550
    },
    {
      "epoch": 221.78217821782178,
      "grad_norm": 0.5418526530265808,
      "learning_rate": 4.001666666666667e-06,
      "loss": 0.3863,
      "step": 5600
    },
    {
      "epoch": 223.76237623762376,
      "grad_norm": 0.7057704925537109,
      "learning_rate": 3.918333333333334e-06,
      "loss": 0.394,
      "step": 5650
    },
    {
      "epoch": 225.74257425742573,
      "grad_norm": 0.6317301392555237,
      "learning_rate": 3.8350000000000006e-06,
      "loss": 0.3906,
      "step": 5700
    },
    {
      "epoch": 227.72277227722773,
      "grad_norm": 0.6327551007270813,
      "learning_rate": 3.7516666666666666e-06,
      "loss": 0.3902,
      "step": 5750
    },
    {
      "epoch": 229.7029702970297,
      "grad_norm": 0.6199086904525757,
      "learning_rate": 3.668333333333334e-06,
      "loss": 0.3912,
      "step": 5800
    },
    {
      "epoch": 231.68316831683168,
      "grad_norm": 0.6341702938079834,
      "learning_rate": 3.585e-06,
      "loss": 0.391,
      "step": 5850
    },
    {
      "epoch": 233.66336633663366,
      "grad_norm": 0.7226659655570984,
      "learning_rate": 3.5016666666666673e-06,
      "loss": 0.3908,
      "step": 5900
    },
    {
      "epoch": 235.64356435643563,
      "grad_norm": 0.6628900170326233,
      "learning_rate": 3.4183333333333334e-06,
      "loss": 0.3899,
      "step": 5950
    },
    {
      "epoch": 237.62376237623764,
      "grad_norm": 0.6972154378890991,
      "learning_rate": 3.3350000000000003e-06,
      "loss": 0.3884,
      "step": 6000
    },
    {
      "epoch": 237.62376237623764,
      "eval_loss": 0.3934413194656372,
      "eval_runtime": 7.3097,
      "eval_samples_per_second": 24.625,
      "eval_steps_per_second": 3.146,
      "step": 6000
    },
    {
      "epoch": 239.6039603960396,
      "grad_norm": 0.7429386377334595,
      "learning_rate": 3.2516666666666667e-06,
      "loss": 0.3902,
      "step": 6050
    },
    {
      "epoch": 241.58415841584159,
      "grad_norm": 0.6832717061042786,
      "learning_rate": 3.1683333333333336e-06,
      "loss": 0.3896,
      "step": 6100
    },
    {
      "epoch": 243.56435643564356,
      "grad_norm": 0.7591320872306824,
      "learning_rate": 3.085e-06,
      "loss": 0.3905,
      "step": 6150
    },
    {
      "epoch": 245.54455445544554,
      "grad_norm": 0.5904133319854736,
      "learning_rate": 3.001666666666667e-06,
      "loss": 0.3876,
      "step": 6200
    },
    {
      "epoch": 247.52475247524754,
      "grad_norm": 0.6338399052619934,
      "learning_rate": 2.9183333333333335e-06,
      "loss": 0.3891,
      "step": 6250
    },
    {
      "epoch": 249.5049504950495,
      "grad_norm": 0.6359020471572876,
      "learning_rate": 2.835e-06,
      "loss": 0.3905,
      "step": 6300
    },
    {
      "epoch": 251.4851485148515,
      "grad_norm": 0.6320521831512451,
      "learning_rate": 2.751666666666667e-06,
      "loss": 0.3885,
      "step": 6350
    },
    {
      "epoch": 253.46534653465346,
      "grad_norm": 0.6006565690040588,
      "learning_rate": 2.6683333333333333e-06,
      "loss": 0.3861,
      "step": 6400
    },
    {
      "epoch": 255.44554455445544,
      "grad_norm": 0.5949265360832214,
      "learning_rate": 2.5866666666666667e-06,
      "loss": 0.3898,
      "step": 6450
    },
    {
      "epoch": 257.4257425742574,
      "grad_norm": 0.6407713294029236,
      "learning_rate": 2.5033333333333336e-06,
      "loss": 0.3889,
      "step": 6500
    },
    {
      "epoch": 257.4257425742574,
      "eval_loss": 0.3926040828227997,
      "eval_runtime": 7.8449,
      "eval_samples_per_second": 22.945,
      "eval_steps_per_second": 2.932,
      "step": 6500
    },
    {
      "epoch": 259.4059405940594,
      "grad_norm": 0.6776269674301147,
      "learning_rate": 2.42e-06,
      "loss": 0.3863,
      "step": 6550
    },
    {
      "epoch": 261.38613861386136,
      "grad_norm": 0.7183690667152405,
      "learning_rate": 2.3366666666666666e-06,
      "loss": 0.388,
      "step": 6600
    },
    {
      "epoch": 263.36633663366337,
      "grad_norm": 0.683351993560791,
      "learning_rate": 2.2533333333333335e-06,
      "loss": 0.3861,
      "step": 6650
    },
    {
      "epoch": 265.34653465346537,
      "grad_norm": 0.5622109770774841,
      "learning_rate": 2.17e-06,
      "loss": 0.3872,
      "step": 6700
    },
    {
      "epoch": 267.3267326732673,
      "grad_norm": 0.6307435035705566,
      "learning_rate": 2.086666666666667e-06,
      "loss": 0.3851,
      "step": 6750
    },
    {
      "epoch": 269.3069306930693,
      "grad_norm": 0.5858733654022217,
      "learning_rate": 2.0033333333333334e-06,
      "loss": 0.3847,
      "step": 6800
    },
    {
      "epoch": 271.28712871287127,
      "grad_norm": 0.7978197336196899,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.3876,
      "step": 6850
    },
    {
      "epoch": 273.26732673267327,
      "grad_norm": 0.6843935251235962,
      "learning_rate": 1.836666666666667e-06,
      "loss": 0.3871,
      "step": 6900
    },
    {
      "epoch": 275.2475247524753,
      "grad_norm": 0.8366850018501282,
      "learning_rate": 1.7533333333333336e-06,
      "loss": 0.3909,
      "step": 6950
    },
    {
      "epoch": 277.2277227722772,
      "grad_norm": 0.584172248840332,
      "learning_rate": 1.6700000000000003e-06,
      "loss": 0.3855,
      "step": 7000
    },
    {
      "epoch": 277.2277227722772,
      "eval_loss": 0.39400026202201843,
      "eval_runtime": 6.8593,
      "eval_samples_per_second": 26.242,
      "eval_steps_per_second": 3.353,
      "step": 7000
    },
    {
      "epoch": 279.2079207920792,
      "grad_norm": 0.7005228996276855,
      "learning_rate": 1.586666666666667e-06,
      "loss": 0.3858,
      "step": 7050
    },
    {
      "epoch": 281.18811881188117,
      "grad_norm": 0.5294095277786255,
      "learning_rate": 1.5033333333333337e-06,
      "loss": 0.3845,
      "step": 7100
    },
    {
      "epoch": 283.16831683168317,
      "grad_norm": 0.5473897457122803,
      "learning_rate": 1.42e-06,
      "loss": 0.3846,
      "step": 7150
    },
    {
      "epoch": 285.1485148514852,
      "grad_norm": 0.6222283840179443,
      "learning_rate": 1.3366666666666666e-06,
      "loss": 0.3849,
      "step": 7200
    },
    {
      "epoch": 287.1287128712871,
      "grad_norm": 0.6396065354347229,
      "learning_rate": 1.2533333333333333e-06,
      "loss": 0.3878,
      "step": 7250
    },
    {
      "epoch": 289.1089108910891,
      "grad_norm": 0.6903032064437866,
      "learning_rate": 1.1700000000000002e-06,
      "loss": 0.3865,
      "step": 7300
    },
    {
      "epoch": 291.08910891089107,
      "grad_norm": 0.6112281680107117,
      "learning_rate": 1.0866666666666667e-06,
      "loss": 0.3888,
      "step": 7350
    },
    {
      "epoch": 293.0693069306931,
      "grad_norm": 0.6655446887016296,
      "learning_rate": 1.0033333333333334e-06,
      "loss": 0.3844,
      "step": 7400
    },
    {
      "epoch": 295.0495049504951,
      "grad_norm": 0.5784509181976318,
      "learning_rate": 9.200000000000001e-07,
      "loss": 0.3862,
      "step": 7450
    },
    {
      "epoch": 297.029702970297,
      "grad_norm": 0.6216564774513245,
      "learning_rate": 8.366666666666668e-07,
      "loss": 0.3892,
      "step": 7500
    },
    {
      "epoch": 297.029702970297,
      "eval_loss": 0.3934391140937805,
      "eval_runtime": 7.8487,
      "eval_samples_per_second": 22.934,
      "eval_steps_per_second": 2.93,
      "step": 7500
    },
    {
      "epoch": 299.009900990099,
      "grad_norm": 0.9473925232887268,
      "learning_rate": 7.533333333333335e-07,
      "loss": 0.3873,
      "step": 7550
    },
    {
      "epoch": 300.990099009901,
      "grad_norm": 0.5682732462882996,
      "learning_rate": 6.7e-07,
      "loss": 0.3874,
      "step": 7600
    },
    {
      "epoch": 302.970297029703,
      "grad_norm": 0.787708580493927,
      "learning_rate": 5.866666666666667e-07,
      "loss": 0.3892,
      "step": 7650
    },
    {
      "epoch": 304.9504950495049,
      "grad_norm": 0.7085989713668823,
      "learning_rate": 5.033333333333334e-07,
      "loss": 0.3851,
      "step": 7700
    },
    {
      "epoch": 306.9306930693069,
      "grad_norm": 0.6272616386413574,
      "learning_rate": 4.2000000000000006e-07,
      "loss": 0.3891,
      "step": 7750
    },
    {
      "epoch": 308.91089108910893,
      "grad_norm": 0.6887345314025879,
      "learning_rate": 3.366666666666667e-07,
      "loss": 0.3874,
      "step": 7800
    },
    {
      "epoch": 310.8910891089109,
      "grad_norm": 0.6206291913986206,
      "learning_rate": 2.533333333333333e-07,
      "loss": 0.3853,
      "step": 7850
    },
    {
      "epoch": 312.8712871287129,
      "grad_norm": 0.6549195051193237,
      "learning_rate": 1.7000000000000001e-07,
      "loss": 0.3867,
      "step": 7900
    },
    {
      "epoch": 314.8514851485148,
      "grad_norm": 0.7210911512374878,
      "learning_rate": 8.666666666666668e-08,
      "loss": 0.3845,
      "step": 7950
    },
    {
      "epoch": 316.83168316831683,
      "grad_norm": 0.6023419499397278,
      "learning_rate": 3.3333333333333334e-09,
      "loss": 0.3899,
      "step": 8000
    },
    {
      "epoch": 316.83168316831683,
      "eval_loss": 0.39394611120224,
      "eval_runtime": 7.5969,
      "eval_samples_per_second": 23.694,
      "eval_steps_per_second": 3.028,
      "step": 8000
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 320,
  "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": 8.643923525044128e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}