{
  "best_metric": 0.40593138337135315,
  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-7000",
  "epoch": 347.82608695652175,
  "eval_steps": 500,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 5.992666721343994,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.9248,
      "step": 50
    },
    {
      "epoch": 4.3478260869565215,
      "grad_norm": 3.4944448471069336,
      "learning_rate": 4.95e-07,
      "loss": 0.8709,
      "step": 100
    },
    {
      "epoch": 6.521739130434782,
      "grad_norm": 1.9902064800262451,
      "learning_rate": 7.450000000000001e-07,
      "loss": 0.8005,
      "step": 150
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 1.4649031162261963,
      "learning_rate": 9.950000000000002e-07,
      "loss": 0.733,
      "step": 200
    },
    {
      "epoch": 10.869565217391305,
      "grad_norm": 1.4645055532455444,
      "learning_rate": 1.2450000000000002e-06,
      "loss": 0.6758,
      "step": 250
    },
    {
      "epoch": 13.043478260869565,
      "grad_norm": 1.5862590074539185,
      "learning_rate": 1.495e-06,
      "loss": 0.6246,
      "step": 300
    },
    {
      "epoch": 15.217391304347826,
      "grad_norm": 1.198890209197998,
      "learning_rate": 1.745e-06,
      "loss": 0.5917,
      "step": 350
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 1.324212908744812,
      "learning_rate": 1.9950000000000004e-06,
      "loss": 0.5721,
      "step": 400
    },
    {
      "epoch": 19.565217391304348,
      "grad_norm": 1.1638662815093994,
      "learning_rate": 2.245e-06,
      "loss": 0.5635,
      "step": 450
    },
    {
      "epoch": 21.73913043478261,
      "grad_norm": 0.9293428063392639,
      "learning_rate": 2.4950000000000003e-06,
      "loss": 0.5541,
      "step": 500
    },
    {
      "epoch": 21.73913043478261,
      "eval_loss": 0.49769046902656555,
      "eval_runtime": 12.8521,
      "eval_samples_per_second": 12.527,
      "eval_steps_per_second": 1.634,
      "step": 500
    },
    {
      "epoch": 23.91304347826087,
      "grad_norm": 0.7982977628707886,
      "learning_rate": 2.7450000000000004e-06,
      "loss": 0.5487,
      "step": 550
    },
    {
      "epoch": 26.08695652173913,
      "grad_norm": 0.8491495251655579,
      "learning_rate": 2.995e-06,
      "loss": 0.5373,
      "step": 600
    },
    {
      "epoch": 28.26086956521739,
      "grad_norm": 0.7158030867576599,
      "learning_rate": 3.2450000000000003e-06,
      "loss": 0.5271,
      "step": 650
    },
    {
      "epoch": 30.434782608695652,
      "grad_norm": 1.2020411491394043,
      "learning_rate": 3.495e-06,
      "loss": 0.5201,
      "step": 700
    },
    {
      "epoch": 32.608695652173914,
      "grad_norm": 0.7339745163917542,
      "learning_rate": 3.745e-06,
      "loss": 0.5126,
      "step": 750
    },
    {
      "epoch": 34.78260869565217,
      "grad_norm": 1.1129218339920044,
      "learning_rate": 3.995000000000001e-06,
      "loss": 0.5136,
      "step": 800
    },
    {
      "epoch": 36.95652173913044,
      "grad_norm": 0.7733961939811707,
      "learning_rate": 4.245e-06,
      "loss": 0.5099,
      "step": 850
    },
    {
      "epoch": 39.130434782608695,
      "grad_norm": 1.1559356451034546,
      "learning_rate": 4.495e-06,
      "loss": 0.5047,
      "step": 900
    },
    {
      "epoch": 41.30434782608695,
      "grad_norm": 0.873518705368042,
      "learning_rate": 4.745e-06,
      "loss": 0.4955,
      "step": 950
    },
    {
      "epoch": 43.47826086956522,
      "grad_norm": 0.703760027885437,
      "learning_rate": 4.9950000000000005e-06,
      "loss": 0.4931,
      "step": 1000
    },
    {
      "epoch": 43.47826086956522,
      "eval_loss": 0.4528730511665344,
      "eval_runtime": 12.5404,
      "eval_samples_per_second": 12.838,
      "eval_steps_per_second": 1.675,
      "step": 1000
    },
    {
      "epoch": 45.65217391304348,
      "grad_norm": 0.9317522644996643,
      "learning_rate": 5.245e-06,
      "loss": 0.4922,
      "step": 1050
    },
    {
      "epoch": 47.82608695652174,
      "grad_norm": 0.9118973016738892,
      "learning_rate": 5.495000000000001e-06,
      "loss": 0.4855,
      "step": 1100
    },
    {
      "epoch": 50.0,
      "grad_norm": 2.0197670459747314,
      "learning_rate": 5.745000000000001e-06,
      "loss": 0.485,
      "step": 1150
    },
    {
      "epoch": 52.17391304347826,
      "grad_norm": 1.1067094802856445,
      "learning_rate": 5.995000000000001e-06,
      "loss": 0.4872,
      "step": 1200
    },
    {
      "epoch": 54.34782608695652,
      "grad_norm": 1.20389723777771,
      "learning_rate": 6.245000000000001e-06,
      "loss": 0.4836,
      "step": 1250
    },
    {
      "epoch": 56.52173913043478,
      "grad_norm": 0.9784926176071167,
      "learning_rate": 6.4950000000000005e-06,
      "loss": 0.4793,
      "step": 1300
    },
    {
      "epoch": 58.69565217391305,
      "grad_norm": 1.4276039600372314,
      "learning_rate": 6.745000000000001e-06,
      "loss": 0.4774,
      "step": 1350
    },
    {
      "epoch": 60.869565217391305,
      "grad_norm": 1.117233157157898,
      "learning_rate": 6.995000000000001e-06,
      "loss": 0.4776,
      "step": 1400
    },
    {
      "epoch": 63.04347826086956,
      "grad_norm": 0.9267759919166565,
      "learning_rate": 7.245000000000001e-06,
      "loss": 0.4682,
      "step": 1450
    },
    {
      "epoch": 65.21739130434783,
      "grad_norm": 0.8514929413795471,
      "learning_rate": 7.495000000000001e-06,
      "loss": 0.4695,
      "step": 1500
    },
    {
      "epoch": 65.21739130434783,
      "eval_loss": 0.43296581506729126,
      "eval_runtime": 12.858,
      "eval_samples_per_second": 12.521,
      "eval_steps_per_second": 1.633,
      "step": 1500
    },
    {
      "epoch": 67.3913043478261,
      "grad_norm": 1.6638849973678589,
      "learning_rate": 7.745e-06,
      "loss": 0.4674,
      "step": 1550
    },
    {
      "epoch": 69.56521739130434,
      "grad_norm": 0.8030112981796265,
      "learning_rate": 7.995e-06,
      "loss": 0.4652,
      "step": 1600
    },
    {
      "epoch": 71.73913043478261,
      "grad_norm": 1.2407337427139282,
      "learning_rate": 8.245000000000002e-06,
      "loss": 0.4638,
      "step": 1650
    },
    {
      "epoch": 73.91304347826087,
      "grad_norm": 1.0331224203109741,
      "learning_rate": 8.495e-06,
      "loss": 0.4685,
      "step": 1700
    },
    {
      "epoch": 76.08695652173913,
      "grad_norm": 2.4681615829467773,
      "learning_rate": 8.745000000000002e-06,
      "loss": 0.4627,
      "step": 1750
    },
    {
      "epoch": 78.26086956521739,
      "grad_norm": 0.9344178438186646,
      "learning_rate": 8.995000000000001e-06,
      "loss": 0.4575,
      "step": 1800
    },
    {
      "epoch": 80.43478260869566,
      "grad_norm": 1.0561341047286987,
      "learning_rate": 9.245e-06,
      "loss": 0.4621,
      "step": 1850
    },
    {
      "epoch": 82.6086956521739,
      "grad_norm": 0.9816989302635193,
      "learning_rate": 9.495000000000001e-06,
      "loss": 0.4544,
      "step": 1900
    },
    {
      "epoch": 84.78260869565217,
      "grad_norm": 1.4779256582260132,
      "learning_rate": 9.745e-06,
      "loss": 0.4579,
      "step": 1950
    },
    {
      "epoch": 86.95652173913044,
      "grad_norm": 1.1634950637817383,
      "learning_rate": 9.995000000000002e-06,
      "loss": 0.4518,
      "step": 2000
    },
    {
      "epoch": 86.95652173913044,
      "eval_loss": 0.42301931977272034,
      "eval_runtime": 12.725,
      "eval_samples_per_second": 12.652,
      "eval_steps_per_second": 1.65,
      "step": 2000
    },
    {
      "epoch": 89.1304347826087,
      "grad_norm": 1.0197867155075073,
      "learning_rate": 9.918333333333335e-06,
      "loss": 0.4501,
      "step": 2050
    },
    {
      "epoch": 91.30434782608695,
      "grad_norm": 1.2543045282363892,
      "learning_rate": 9.835000000000002e-06,
      "loss": 0.4499,
      "step": 2100
    },
    {
      "epoch": 93.47826086956522,
      "grad_norm": 0.8574827909469604,
      "learning_rate": 9.751666666666667e-06,
      "loss": 0.4485,
      "step": 2150
    },
    {
      "epoch": 95.65217391304348,
      "grad_norm": 1.3784998655319214,
      "learning_rate": 9.668333333333334e-06,
      "loss": 0.4507,
      "step": 2200
    },
    {
      "epoch": 97.82608695652173,
      "grad_norm": 0.9518347978591919,
      "learning_rate": 9.585e-06,
      "loss": 0.4486,
      "step": 2250
    },
    {
      "epoch": 100.0,
      "grad_norm": 2.422320604324341,
      "learning_rate": 9.501666666666667e-06,
      "loss": 0.4455,
      "step": 2300
    },
    {
      "epoch": 102.17391304347827,
      "grad_norm": 1.3441377878189087,
      "learning_rate": 9.418333333333334e-06,
      "loss": 0.45,
      "step": 2350
    },
    {
      "epoch": 104.34782608695652,
      "grad_norm": 0.969160795211792,
      "learning_rate": 9.335000000000001e-06,
      "loss": 0.4454,
      "step": 2400
    },
    {
      "epoch": 106.52173913043478,
      "grad_norm": 2.1275179386138916,
      "learning_rate": 9.251666666666668e-06,
      "loss": 0.4459,
      "step": 2450
    },
    {
      "epoch": 108.69565217391305,
      "grad_norm": 1.3673362731933594,
      "learning_rate": 9.168333333333333e-06,
      "loss": 0.4442,
      "step": 2500
    },
    {
      "epoch": 108.69565217391305,
      "eval_loss": 0.4178585410118103,
      "eval_runtime": 13.0534,
      "eval_samples_per_second": 12.334,
      "eval_steps_per_second": 1.609,
      "step": 2500
    },
    {
      "epoch": 110.8695652173913,
      "grad_norm": 1.3769398927688599,
      "learning_rate": 9.085e-06,
      "loss": 0.4406,
      "step": 2550
    },
    {
      "epoch": 113.04347826086956,
      "grad_norm": 0.9913681745529175,
      "learning_rate": 9.001666666666667e-06,
      "loss": 0.4389,
      "step": 2600
    },
    {
      "epoch": 115.21739130434783,
      "grad_norm": 1.1747106313705444,
      "learning_rate": 8.918333333333334e-06,
      "loss": 0.4386,
      "step": 2650
    },
    {
      "epoch": 117.3913043478261,
      "grad_norm": 1.0514781475067139,
      "learning_rate": 8.835000000000001e-06,
      "loss": 0.4393,
      "step": 2700
    },
    {
      "epoch": 119.56521739130434,
      "grad_norm": 1.8967915773391724,
      "learning_rate": 8.751666666666668e-06,
      "loss": 0.4421,
      "step": 2750
    },
    {
      "epoch": 121.73913043478261,
      "grad_norm": 0.8795832395553589,
      "learning_rate": 8.668333333333335e-06,
      "loss": 0.436,
      "step": 2800
    },
    {
      "epoch": 123.91304347826087,
      "grad_norm": 0.7928704023361206,
      "learning_rate": 8.585000000000002e-06,
      "loss": 0.4333,
      "step": 2850
    },
    {
      "epoch": 126.08695652173913,
      "grad_norm": 1.2805510759353638,
      "learning_rate": 8.501666666666667e-06,
      "loss": 0.4326,
      "step": 2900
    },
    {
      "epoch": 128.2608695652174,
      "grad_norm": 1.420920968055725,
      "learning_rate": 8.418333333333334e-06,
      "loss": 0.4317,
      "step": 2950
    },
    {
      "epoch": 130.43478260869566,
      "grad_norm": 0.7063888907432556,
      "learning_rate": 8.335e-06,
      "loss": 0.4344,
      "step": 3000
    },
    {
      "epoch": 130.43478260869566,
      "eval_loss": 0.41350311040878296,
      "eval_runtime": 12.8066,
      "eval_samples_per_second": 12.572,
      "eval_steps_per_second": 1.64,
      "step": 3000
    },
    {
      "epoch": 132.6086956521739,
      "grad_norm": 1.8065855503082275,
      "learning_rate": 8.251666666666668e-06,
      "loss": 0.4361,
      "step": 3050
    },
    {
      "epoch": 134.7826086956522,
      "grad_norm": 0.8073704838752747,
      "learning_rate": 8.168333333333334e-06,
      "loss": 0.4339,
      "step": 3100
    },
    {
      "epoch": 136.95652173913044,
      "grad_norm": 1.2890065908432007,
      "learning_rate": 8.085000000000001e-06,
      "loss": 0.4325,
      "step": 3150
    },
    {
      "epoch": 139.1304347826087,
      "grad_norm": 1.336401104927063,
      "learning_rate": 8.001666666666668e-06,
      "loss": 0.4334,
      "step": 3200
    },
    {
      "epoch": 141.30434782608697,
      "grad_norm": 1.2965891361236572,
      "learning_rate": 7.918333333333333e-06,
      "loss": 0.4298,
      "step": 3250
    },
    {
      "epoch": 143.47826086956522,
      "grad_norm": 0.8761409521102905,
      "learning_rate": 7.835e-06,
      "loss": 0.4231,
      "step": 3300
    },
    {
      "epoch": 145.65217391304347,
      "grad_norm": 1.1475930213928223,
      "learning_rate": 7.751666666666667e-06,
      "loss": 0.4316,
      "step": 3350
    },
    {
      "epoch": 147.82608695652175,
      "grad_norm": 0.8305974006652832,
      "learning_rate": 7.668333333333334e-06,
      "loss": 0.4277,
      "step": 3400
    },
    {
      "epoch": 150.0,
      "grad_norm": 1.6335935592651367,
      "learning_rate": 7.585e-06,
      "loss": 0.4248,
      "step": 3450
    },
    {
      "epoch": 152.17391304347825,
      "grad_norm": 1.1171984672546387,
      "learning_rate": 7.501666666666667e-06,
      "loss": 0.4318,
      "step": 3500
    },
    {
      "epoch": 152.17391304347825,
      "eval_loss": 0.4111216962337494,
      "eval_runtime": 13.0991,
      "eval_samples_per_second": 12.291,
      "eval_steps_per_second": 1.603,
      "step": 3500
    },
    {
      "epoch": 154.34782608695653,
      "grad_norm": 0.8932999968528748,
      "learning_rate": 7.418333333333334e-06,
      "loss": 0.4287,
      "step": 3550
    },
    {
      "epoch": 156.52173913043478,
      "grad_norm": 0.9644981622695923,
      "learning_rate": 7.335000000000001e-06,
      "loss": 0.4253,
      "step": 3600
    },
    {
      "epoch": 158.69565217391303,
      "grad_norm": 1.5559085607528687,
      "learning_rate": 7.251666666666667e-06,
      "loss": 0.4282,
      "step": 3650
    },
    {
      "epoch": 160.8695652173913,
      "grad_norm": 1.0696306228637695,
      "learning_rate": 7.168333333333334e-06,
      "loss": 0.4221,
      "step": 3700
    },
    {
      "epoch": 163.04347826086956,
      "grad_norm": 1.0170419216156006,
      "learning_rate": 7.085000000000001e-06,
      "loss": 0.4254,
      "step": 3750
    },
    {
      "epoch": 165.2173913043478,
      "grad_norm": 0.7487155795097351,
      "learning_rate": 7.001666666666668e-06,
      "loss": 0.4269,
      "step": 3800
    },
    {
      "epoch": 167.3913043478261,
      "grad_norm": 0.7589255571365356,
      "learning_rate": 6.918333333333334e-06,
      "loss": 0.4252,
      "step": 3850
    },
    {
      "epoch": 169.56521739130434,
      "grad_norm": 0.9557852745056152,
      "learning_rate": 6.835000000000001e-06,
      "loss": 0.4224,
      "step": 3900
    },
    {
      "epoch": 171.7391304347826,
      "grad_norm": 0.7511025667190552,
      "learning_rate": 6.7516666666666675e-06,
      "loss": 0.4201,
      "step": 3950
    },
    {
      "epoch": 173.91304347826087,
      "grad_norm": 0.9032562375068665,
      "learning_rate": 6.668333333333334e-06,
      "loss": 0.4201,
      "step": 4000
    },
    {
      "epoch": 173.91304347826087,
      "eval_loss": 0.41097140312194824,
      "eval_runtime": 12.8806,
      "eval_samples_per_second": 12.499,
      "eval_steps_per_second": 1.63,
      "step": 4000
    },
    {
      "epoch": 176.08695652173913,
      "grad_norm": 0.9379887580871582,
      "learning_rate": 6.5866666666666666e-06,
      "loss": 0.426,
      "step": 4050
    },
    {
      "epoch": 178.2608695652174,
      "grad_norm": 1.2516026496887207,
      "learning_rate": 6.5033333333333335e-06,
      "loss": 0.4244,
      "step": 4100
    },
    {
      "epoch": 180.43478260869566,
      "grad_norm": 0.898951530456543,
      "learning_rate": 6.42e-06,
      "loss": 0.4212,
      "step": 4150
    },
    {
      "epoch": 182.6086956521739,
      "grad_norm": 0.810229480266571,
      "learning_rate": 6.336666666666667e-06,
      "loss": 0.4188,
      "step": 4200
    },
    {
      "epoch": 184.7826086956522,
      "grad_norm": 0.8796959519386292,
      "learning_rate": 6.253333333333333e-06,
      "loss": 0.4196,
      "step": 4250
    },
    {
      "epoch": 186.95652173913044,
      "grad_norm": 0.8476919531822205,
      "learning_rate": 6.17e-06,
      "loss": 0.4215,
      "step": 4300
    },
    {
      "epoch": 189.1304347826087,
      "grad_norm": 0.8933534622192383,
      "learning_rate": 6.086666666666667e-06,
      "loss": 0.4209,
      "step": 4350
    },
    {
      "epoch": 191.30434782608697,
      "grad_norm": 0.965189516544342,
      "learning_rate": 6.003333333333334e-06,
      "loss": 0.4157,
      "step": 4400
    },
    {
      "epoch": 193.47826086956522,
      "grad_norm": 0.6911047101020813,
      "learning_rate": 5.92e-06,
      "loss": 0.4156,
      "step": 4450
    },
    {
      "epoch": 195.65217391304347,
      "grad_norm": 0.98875492811203,
      "learning_rate": 5.836666666666667e-06,
      "loss": 0.4185,
      "step": 4500
    },
    {
      "epoch": 195.65217391304347,
      "eval_loss": 0.40913259983062744,
      "eval_runtime": 13.1748,
      "eval_samples_per_second": 12.22,
      "eval_steps_per_second": 1.594,
      "step": 4500
    },
    {
      "epoch": 197.82608695652175,
      "grad_norm": 0.8193331360816956,
      "learning_rate": 5.753333333333334e-06,
      "loss": 0.4171,
      "step": 4550
    },
    {
      "epoch": 200.0,
      "grad_norm": 2.4391345977783203,
      "learning_rate": 5.67e-06,
      "loss": 0.4156,
      "step": 4600
    },
    {
      "epoch": 202.17391304347825,
      "grad_norm": 0.7653400301933289,
      "learning_rate": 5.586666666666667e-06,
      "loss": 0.4198,
      "step": 4650
    },
    {
      "epoch": 204.34782608695653,
      "grad_norm": 0.9531933069229126,
      "learning_rate": 5.503333333333334e-06,
      "loss": 0.4165,
      "step": 4700
    },
    {
      "epoch": 206.52173913043478,
      "grad_norm": 1.7444751262664795,
      "learning_rate": 5.420000000000001e-06,
      "loss": 0.417,
      "step": 4750
    },
    {
      "epoch": 208.69565217391303,
      "grad_norm": 0.9747681021690369,
      "learning_rate": 5.336666666666667e-06,
      "loss": 0.4133,
      "step": 4800
    },
    {
      "epoch": 210.8695652173913,
      "grad_norm": 1.063536524772644,
      "learning_rate": 5.2533333333333336e-06,
      "loss": 0.4153,
      "step": 4850
    },
    {
      "epoch": 213.04347826086956,
      "grad_norm": 0.8719898462295532,
      "learning_rate": 5.1700000000000005e-06,
      "loss": 0.4104,
      "step": 4900
    },
    {
      "epoch": 215.2173913043478,
      "grad_norm": 0.7953347563743591,
      "learning_rate": 5.086666666666667e-06,
      "loss": 0.4204,
      "step": 4950
    },
    {
      "epoch": 217.3913043478261,
      "grad_norm": 0.9870523810386658,
      "learning_rate": 5.0033333333333334e-06,
      "loss": 0.4153,
      "step": 5000
    },
    {
      "epoch": 217.3913043478261,
      "eval_loss": 0.4096957743167877,
      "eval_runtime": 13.2951,
      "eval_samples_per_second": 12.11,
      "eval_steps_per_second": 1.58,
      "step": 5000
    },
    {
      "epoch": 219.56521739130434,
      "grad_norm": 0.9140251874923706,
      "learning_rate": 4.92e-06,
      "loss": 0.4142,
      "step": 5050
    },
    {
      "epoch": 221.7391304347826,
      "grad_norm": 0.7339205741882324,
      "learning_rate": 4.836666666666667e-06,
      "loss": 0.4126,
      "step": 5100
    },
    {
      "epoch": 223.91304347826087,
      "grad_norm": 0.8861171007156372,
      "learning_rate": 4.753333333333333e-06,
      "loss": 0.4189,
      "step": 5150
    },
    {
      "epoch": 226.08695652173913,
      "grad_norm": 0.7763547301292419,
      "learning_rate": 4.670000000000001e-06,
      "loss": 0.415,
      "step": 5200
    },
    {
      "epoch": 228.2608695652174,
      "grad_norm": 0.913635790348053,
      "learning_rate": 4.586666666666667e-06,
      "loss": 0.413,
      "step": 5250
    },
    {
      "epoch": 230.43478260869566,
      "grad_norm": 1.0177322626113892,
      "learning_rate": 4.503333333333333e-06,
      "loss": 0.4119,
      "step": 5300
    },
    {
      "epoch": 232.6086956521739,
      "grad_norm": 1.0655298233032227,
      "learning_rate": 4.42e-06,
      "loss": 0.4122,
      "step": 5350
    },
    {
      "epoch": 234.7826086956522,
      "grad_norm": 0.84236079454422,
      "learning_rate": 4.336666666666667e-06,
      "loss": 0.4104,
      "step": 5400
    },
    {
      "epoch": 236.95652173913044,
      "grad_norm": 1.0191460847854614,
      "learning_rate": 4.253333333333334e-06,
      "loss": 0.4128,
      "step": 5450
    },
    {
      "epoch": 239.1304347826087,
      "grad_norm": 0.7806400060653687,
      "learning_rate": 4.17e-06,
      "loss": 0.414,
      "step": 5500
    },
    {
      "epoch": 239.1304347826087,
      "eval_loss": 0.4069381058216095,
      "eval_runtime": 13.5672,
      "eval_samples_per_second": 11.867,
      "eval_steps_per_second": 1.548,
      "step": 5500
    },
    {
      "epoch": 241.30434782608697,
      "grad_norm": 0.866113007068634,
      "learning_rate": 4.086666666666667e-06,
      "loss": 0.4117,
      "step": 5550
    },
    {
      "epoch": 243.47826086956522,
      "grad_norm": 0.8896365165710449,
      "learning_rate": 4.003333333333334e-06,
      "loss": 0.4161,
      "step": 5600
    },
    {
      "epoch": 245.65217391304347,
      "grad_norm": 0.8311805129051208,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.4094,
      "step": 5650
    },
    {
      "epoch": 247.82608695652175,
      "grad_norm": 1.686919927597046,
      "learning_rate": 3.836666666666667e-06,
      "loss": 0.4128,
      "step": 5700
    },
    {
      "epoch": 250.0,
      "grad_norm": 1.9034643173217773,
      "learning_rate": 3.753333333333334e-06,
      "loss": 0.4143,
      "step": 5750
    },
    {
      "epoch": 252.17391304347825,
      "grad_norm": 0.8296219706535339,
      "learning_rate": 3.6700000000000004e-06,
      "loss": 0.4123,
      "step": 5800
    },
    {
      "epoch": 254.34782608695653,
      "grad_norm": 0.6855341792106628,
      "learning_rate": 3.5866666666666673e-06,
      "loss": 0.4099,
      "step": 5850
    },
    {
      "epoch": 256.5217391304348,
      "grad_norm": 1.0125635862350464,
      "learning_rate": 3.5033333333333334e-06,
      "loss": 0.4104,
      "step": 5900
    },
    {
      "epoch": 258.69565217391306,
      "grad_norm": 0.8589877486228943,
      "learning_rate": 3.4200000000000007e-06,
      "loss": 0.4069,
      "step": 5950
    },
    {
      "epoch": 260.8695652173913,
      "grad_norm": 0.9898667931556702,
      "learning_rate": 3.3366666666666668e-06,
      "loss": 0.4113,
      "step": 6000
    },
    {
      "epoch": 260.8695652173913,
      "eval_loss": 0.40804117918014526,
      "eval_runtime": 13.2074,
      "eval_samples_per_second": 12.19,
      "eval_steps_per_second": 1.59,
      "step": 6000
    },
    {
      "epoch": 263.04347826086956,
      "grad_norm": 1.1999317407608032,
      "learning_rate": 3.2533333333333332e-06,
      "loss": 0.4121,
      "step": 6050
    },
    {
      "epoch": 265.2173913043478,
      "grad_norm": 0.8878545761108398,
      "learning_rate": 3.17e-06,
      "loss": 0.4112,
      "step": 6100
    },
    {
      "epoch": 267.39130434782606,
      "grad_norm": 0.7173246145248413,
      "learning_rate": 3.0866666666666666e-06,
      "loss": 0.4079,
      "step": 6150
    },
    {
      "epoch": 269.5652173913044,
      "grad_norm": 0.8617014288902283,
      "learning_rate": 3.0033333333333335e-06,
      "loss": 0.4095,
      "step": 6200
    },
    {
      "epoch": 271.7391304347826,
      "grad_norm": 1.0037391185760498,
      "learning_rate": 2.92e-06,
      "loss": 0.4086,
      "step": 6250
    },
    {
      "epoch": 273.9130434782609,
      "grad_norm": 1.068651795387268,
      "learning_rate": 2.836666666666667e-06,
      "loss": 0.4136,
      "step": 6300
    },
    {
      "epoch": 276.0869565217391,
      "grad_norm": 0.7301667332649231,
      "learning_rate": 2.7533333333333334e-06,
      "loss": 0.4094,
      "step": 6350
    },
    {
      "epoch": 278.2608695652174,
      "grad_norm": 0.7777795195579529,
      "learning_rate": 2.6700000000000003e-06,
      "loss": 0.4099,
      "step": 6400
    },
    {
      "epoch": 280.4347826086956,
      "grad_norm": 2.096015453338623,
      "learning_rate": 2.5866666666666667e-06,
      "loss": 0.4142,
      "step": 6450
    },
    {
      "epoch": 282.60869565217394,
      "grad_norm": 1.2232917547225952,
      "learning_rate": 2.5033333333333336e-06,
      "loss": 0.4133,
      "step": 6500
    },
    {
      "epoch": 282.60869565217394,
      "eval_loss": 0.40729132294654846,
      "eval_runtime": 13.1317,
      "eval_samples_per_second": 12.26,
      "eval_steps_per_second": 1.599,
      "step": 6500
    },
    {
      "epoch": 284.7826086956522,
      "grad_norm": 0.8479878902435303,
      "learning_rate": 2.42e-06,
      "loss": 0.4069,
      "step": 6550
    },
    {
      "epoch": 286.95652173913044,
      "grad_norm": 0.9611658453941345,
      "learning_rate": 2.3366666666666666e-06,
      "loss": 0.4115,
      "step": 6600
    },
    {
      "epoch": 289.1304347826087,
      "grad_norm": 0.9590737819671631,
      "learning_rate": 2.2533333333333335e-06,
      "loss": 0.4115,
      "step": 6650
    },
    {
      "epoch": 291.30434782608694,
      "grad_norm": 0.8980545401573181,
      "learning_rate": 2.17e-06,
      "loss": 0.4075,
      "step": 6700
    },
    {
      "epoch": 293.4782608695652,
      "grad_norm": 1.0965752601623535,
      "learning_rate": 2.086666666666667e-06,
      "loss": 0.4097,
      "step": 6750
    },
    {
      "epoch": 295.6521739130435,
      "grad_norm": 0.7770458459854126,
      "learning_rate": 2.0033333333333334e-06,
      "loss": 0.4122,
      "step": 6800
    },
    {
      "epoch": 297.82608695652175,
      "grad_norm": 0.7583262920379639,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.4072,
      "step": 6850
    },
    {
      "epoch": 300.0,
      "grad_norm": 1.9386942386627197,
      "learning_rate": 1.836666666666667e-06,
      "loss": 0.4096,
      "step": 6900
    },
    {
      "epoch": 302.17391304347825,
      "grad_norm": 0.9514946341514587,
      "learning_rate": 1.7533333333333336e-06,
      "loss": 0.4101,
      "step": 6950
    },
    {
      "epoch": 304.3478260869565,
      "grad_norm": 1.0851093530654907,
      "learning_rate": 1.6716666666666666e-06,
      "loss": 0.4095,
      "step": 7000
    },
    {
      "epoch": 304.3478260869565,
      "eval_loss": 0.40593138337135315,
      "eval_runtime": 13.2236,
      "eval_samples_per_second": 12.175,
      "eval_steps_per_second": 1.588,
      "step": 7000
    },
    {
      "epoch": 306.5217391304348,
      "grad_norm": 0.8590102791786194,
      "learning_rate": 1.5883333333333333e-06,
      "loss": 0.4127,
      "step": 7050
    },
    {
      "epoch": 308.69565217391306,
      "grad_norm": 0.9405089616775513,
      "learning_rate": 1.505e-06,
      "loss": 0.413,
      "step": 7100
    },
    {
      "epoch": 310.8695652173913,
      "grad_norm": 0.7493429780006409,
      "learning_rate": 1.4216666666666667e-06,
      "loss": 0.4137,
      "step": 7150
    },
    {
      "epoch": 313.04347826086956,
      "grad_norm": 0.8461248278617859,
      "learning_rate": 1.3383333333333334e-06,
      "loss": 0.4104,
      "step": 7200
    },
    {
      "epoch": 315.2173913043478,
      "grad_norm": 0.6727828979492188,
      "learning_rate": 1.255e-06,
      "loss": 0.4071,
      "step": 7250
    },
    {
      "epoch": 317.39130434782606,
      "grad_norm": 0.7143705487251282,
      "learning_rate": 1.1716666666666667e-06,
      "loss": 0.4086,
      "step": 7300
    },
    {
      "epoch": 319.5652173913044,
      "grad_norm": 0.8034409284591675,
      "learning_rate": 1.0883333333333334e-06,
      "loss": 0.4051,
      "step": 7350
    },
    {
      "epoch": 321.7391304347826,
      "grad_norm": 0.7853546738624573,
      "learning_rate": 1.0050000000000001e-06,
      "loss": 0.4067,
      "step": 7400
    },
    {
      "epoch": 323.9130434782609,
      "grad_norm": 0.8830783367156982,
      "learning_rate": 9.216666666666667e-07,
      "loss": 0.4089,
      "step": 7450
    },
    {
      "epoch": 326.0869565217391,
      "grad_norm": 0.8115708231925964,
      "learning_rate": 8.383333333333334e-07,
      "loss": 0.4129,
      "step": 7500
    },
    {
      "epoch": 326.0869565217391,
      "eval_loss": 0.4083060622215271,
      "eval_runtime": 12.9398,
      "eval_samples_per_second": 12.442,
      "eval_steps_per_second": 1.623,
      "step": 7500
    },
    {
      "epoch": 328.2608695652174,
      "grad_norm": 0.6245155334472656,
      "learning_rate": 7.550000000000001e-07,
      "loss": 0.4067,
      "step": 7550
    },
    {
      "epoch": 330.4347826086956,
      "grad_norm": 0.7212103009223938,
      "learning_rate": 6.716666666666668e-07,
      "loss": 0.4038,
      "step": 7600
    },
    {
      "epoch": 332.60869565217394,
      "grad_norm": 0.8665474057197571,
      "learning_rate": 5.883333333333333e-07,
      "loss": 0.4048,
      "step": 7650
    },
    {
      "epoch": 334.7826086956522,
      "grad_norm": 0.7773168087005615,
      "learning_rate": 5.05e-07,
      "loss": 0.4054,
      "step": 7700
    },
    {
      "epoch": 336.95652173913044,
      "grad_norm": 0.7131752967834473,
      "learning_rate": 4.2166666666666667e-07,
      "loss": 0.4058,
      "step": 7750
    },
    {
      "epoch": 339.1304347826087,
      "grad_norm": 0.7420967221260071,
      "learning_rate": 3.3833333333333336e-07,
      "loss": 0.409,
      "step": 7800
    },
    {
      "epoch": 341.30434782608694,
      "grad_norm": 0.866454541683197,
      "learning_rate": 2.55e-07,
      "loss": 0.4071,
      "step": 7850
    },
    {
      "epoch": 343.4782608695652,
      "grad_norm": 0.6852492094039917,
      "learning_rate": 1.7166666666666668e-07,
      "loss": 0.4054,
      "step": 7900
    },
    {
      "epoch": 345.6521739130435,
      "grad_norm": 0.8399931788444519,
      "learning_rate": 8.833333333333334e-08,
      "loss": 0.4033,
      "step": 7950
    },
    {
      "epoch": 347.82608695652175,
      "grad_norm": 0.8192908763885498,
      "learning_rate": 5e-09,
      "loss": 0.4035,
      "step": 8000
    },
    {
      "epoch": 347.82608695652175,
      "eval_loss": 0.40772464871406555,
      "eval_runtime": 13.1445,
      "eval_samples_per_second": 12.248,
      "eval_steps_per_second": 1.598,
      "step": 8000
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 348,
  "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": 9.695483990986939e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}