{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1225,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04081632653061224,
      "grad_norm": 9.589848518371582,
      "learning_rate": 9.918367346938776e-06,
      "loss": 0.2612,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 10,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08163265306122448,
      "grad_norm": 6.701302528381348,
      "learning_rate": 9.836734693877552e-06,
      "loss": 0.154,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 20,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12244897959183673,
      "grad_norm": 5.337311267852783,
      "learning_rate": 9.755102040816327e-06,
      "loss": 0.1235,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 30,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16326530612244897,
      "grad_norm": 4.5042338371276855,
      "learning_rate": 9.673469387755103e-06,
      "loss": 0.1096,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 40,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 4.461822032928467,
      "learning_rate": 9.591836734693878e-06,
      "loss": 0.1196,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 50,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24489795918367346,
      "grad_norm": 2.2825701236724854,
      "learning_rate": 9.510204081632653e-06,
      "loss": 0.0805,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 60,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 3.725268602371216,
      "learning_rate": 9.42857142857143e-06,
      "loss": 0.1026,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 70,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32653061224489793,
      "grad_norm": 1.707739233970642,
      "learning_rate": 9.346938775510204e-06,
      "loss": 0.1111,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 80,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3673469387755102,
      "grad_norm": 4.5863938331604,
      "learning_rate": 9.26530612244898e-06,
      "loss": 0.0856,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 90,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 11.972647666931152,
      "learning_rate": 9.183673469387756e-06,
      "loss": 0.0759,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4489795918367347,
      "grad_norm": 4.550654888153076,
      "learning_rate": 9.102040816326532e-06,
      "loss": 0.0717,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4897959183673469,
      "grad_norm": 4.418276786804199,
      "learning_rate": 9.020408163265307e-06,
      "loss": 0.0717,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5306122448979592,
      "grad_norm": 1.651443600654602,
      "learning_rate": 8.938775510204082e-06,
      "loss": 0.0581,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 1.5251814126968384,
      "learning_rate": 8.857142857142858e-06,
      "loss": 0.0481,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 1.7455183267593384,
      "learning_rate": 8.775510204081633e-06,
      "loss": 0.0625,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6530612244897959,
      "grad_norm": 1.7588891983032227,
      "learning_rate": 8.69387755102041e-06,
      "loss": 0.0711,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6938775510204082,
      "grad_norm": 2.7675328254699707,
      "learning_rate": 8.612244897959184e-06,
      "loss": 0.0747,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7346938775510204,
      "grad_norm": 1.781469464302063,
      "learning_rate": 8.530612244897961e-06,
      "loss": 0.061,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7755102040816326,
      "grad_norm": 2.3728435039520264,
      "learning_rate": 8.448979591836736e-06,
      "loss": 0.0588,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 0.8711996674537659,
      "learning_rate": 8.36734693877551e-06,
      "loss": 0.062,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 1.1986733675003052,
      "learning_rate": 8.285714285714287e-06,
      "loss": 0.0627,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8979591836734694,
      "grad_norm": 2.8968520164489746,
      "learning_rate": 8.204081632653062e-06,
      "loss": 0.0604,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9387755102040817,
      "grad_norm": 0.8414793610572815,
      "learning_rate": 8.122448979591837e-06,
      "loss": 0.0559,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9795918367346939,
      "grad_norm": 0.7434167861938477,
      "learning_rate": 8.040816326530613e-06,
      "loss": 0.0498,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0204081632653061,
      "grad_norm": 0.8703041076660156,
      "learning_rate": 7.959183673469388e-06,
      "loss": 0.0618,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0612244897959184,
      "grad_norm": 1.0856379270553589,
      "learning_rate": 7.877551020408164e-06,
      "loss": 0.056,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1020408163265305,
      "grad_norm": 0.8847401142120361,
      "learning_rate": 7.79591836734694e-06,
      "loss": 0.0625,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 1.5929882526397705,
      "learning_rate": 7.714285714285716e-06,
      "loss": 0.0571,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.183673469387755,
      "grad_norm": 0.8007532954216003,
      "learning_rate": 7.63265306122449e-06,
      "loss": 0.0511,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2244897959183674,
      "grad_norm": 1.2002859115600586,
      "learning_rate": 7.551020408163265e-06,
      "loss": 0.065,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2653061224489797,
      "grad_norm": 12.871713638305664,
      "learning_rate": 7.469387755102041e-06,
      "loss": 0.0664,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.306122448979592,
      "grad_norm": 2.46173357963562,
      "learning_rate": 7.387755102040817e-06,
      "loss": 0.0495,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.346938775510204,
      "grad_norm": 0.860598087310791,
      "learning_rate": 7.306122448979592e-06,
      "loss": 0.0603,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3877551020408163,
      "grad_norm": 2.5583598613739014,
      "learning_rate": 7.224489795918368e-06,
      "loss": 0.0547,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.37155964970588684,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.048,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.469387755102041,
      "grad_norm": 1.808316707611084,
      "learning_rate": 7.061224489795919e-06,
      "loss": 0.0462,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.510204081632653,
      "grad_norm": 1.0183931589126587,
      "learning_rate": 6.979591836734695e-06,
      "loss": 0.0594,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5510204081632653,
      "grad_norm": 0.5249583721160889,
      "learning_rate": 6.8979591836734705e-06,
      "loss": 0.0479,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5918367346938775,
      "grad_norm": 1.1005572080612183,
      "learning_rate": 6.816326530612245e-06,
      "loss": 0.0649,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6326530612244898,
      "grad_norm": 0.6047573089599609,
      "learning_rate": 6.734693877551021e-06,
      "loss": 0.0607,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6734693877551021,
      "grad_norm": 0.7261654734611511,
      "learning_rate": 6.653061224489797e-06,
      "loss": 0.0606,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.848527193069458,
      "learning_rate": 6.571428571428572e-06,
      "loss": 0.0532,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7551020408163265,
      "grad_norm": 0.23483288288116455,
      "learning_rate": 6.489795918367348e-06,
      "loss": 0.068,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7959183673469388,
      "grad_norm": 2.0767459869384766,
      "learning_rate": 6.408163265306124e-06,
      "loss": 0.0617,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.836734693877551,
      "grad_norm": 0.5654011368751526,
      "learning_rate": 6.326530612244899e-06,
      "loss": 0.044,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.8775510204081631,
      "grad_norm": 0.7382919788360596,
      "learning_rate": 6.244897959183675e-06,
      "loss": 0.0537,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.9183673469387754,
      "grad_norm": 1.3547204732894897,
      "learning_rate": 6.163265306122449e-06,
      "loss": 0.0432,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.9591836734693877,
      "grad_norm": 0.19681082665920258,
      "learning_rate": 6.0816326530612245e-06,
      "loss": 0.0498,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.109737515449524,
      "learning_rate": 6e-06,
      "loss": 0.0639,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.0408163265306123,
      "grad_norm": 0.5894625782966614,
      "learning_rate": 5.918367346938776e-06,
      "loss": 0.0593,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.0816326530612246,
      "grad_norm": 0.7122555375099182,
      "learning_rate": 5.8367346938775515e-06,
      "loss": 0.0498,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.122448979591837,
      "grad_norm": 0.8958902955055237,
      "learning_rate": 5.755102040816327e-06,
      "loss": 0.0457,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.163265306122449,
      "grad_norm": 11.620415687561035,
      "learning_rate": 5.673469387755103e-06,
      "loss": 0.0626,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.204081632653061,
      "grad_norm": 0.3538230061531067,
      "learning_rate": 5.591836734693878e-06,
      "loss": 0.0584,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.2448979591836733,
      "grad_norm": 1.5313146114349365,
      "learning_rate": 5.510204081632653e-06,
      "loss": 0.0627,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 1.3519809246063232,
      "learning_rate": 5.428571428571429e-06,
      "loss": 0.0572,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.326530612244898,
      "grad_norm": 1.0263270139694214,
      "learning_rate": 5.3469387755102045e-06,
      "loss": 0.0585,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.36734693877551,
      "grad_norm": 0.8926671147346497,
      "learning_rate": 5.26530612244898e-06,
      "loss": 0.0673,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.4081632653061225,
      "grad_norm": 0.3185974955558777,
      "learning_rate": 5.183673469387756e-06,
      "loss": 0.0537,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.4489795918367347,
      "grad_norm": 0.944624662399292,
      "learning_rate": 5.1020408163265315e-06,
      "loss": 0.0442,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.489795918367347,
      "grad_norm": 0.32796111702919006,
      "learning_rate": 5.020408163265307e-06,
      "loss": 0.0413,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.5306122448979593,
      "grad_norm": 0.7929801940917969,
      "learning_rate": 4.938775510204082e-06,
      "loss": 0.0428,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.910254955291748,
      "learning_rate": 4.857142857142858e-06,
      "loss": 0.0813,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.612244897959184,
      "grad_norm": 1.101942539215088,
      "learning_rate": 4.775510204081633e-06,
      "loss": 0.0495,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.6530612244897958,
      "grad_norm": 0.7182526588439941,
      "learning_rate": 4.693877551020409e-06,
      "loss": 0.0471,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.693877551020408,
      "grad_norm": 0.8068158626556396,
      "learning_rate": 4.612244897959184e-06,
      "loss": 0.0469,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.7346938775510203,
      "grad_norm": 1.2375913858413696,
      "learning_rate": 4.530612244897959e-06,
      "loss": 0.0857,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.7755102040816326,
      "grad_norm": 1.1524357795715332,
      "learning_rate": 4.448979591836735e-06,
      "loss": 0.0488,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.816326530612245,
      "grad_norm": 0.3913586437702179,
      "learning_rate": 4.367346938775511e-06,
      "loss": 0.0451,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.47935113310813904,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.0433,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.8979591836734695,
      "grad_norm": 0.8084143996238708,
      "learning_rate": 4.204081632653061e-06,
      "loss": 0.0548,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.938775510204082,
      "grad_norm": 1.7315497398376465,
      "learning_rate": 4.122448979591837e-06,
      "loss": 0.0587,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.979591836734694,
      "grad_norm": 0.20743349194526672,
      "learning_rate": 4.040816326530612e-06,
      "loss": 0.0342,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.020408163265306,
      "grad_norm": 0.8024761080741882,
      "learning_rate": 3.959183673469388e-06,
      "loss": 0.053,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.061224489795918,
      "grad_norm": 0.45326006412506104,
      "learning_rate": 3.877551020408164e-06,
      "loss": 0.0619,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.1020408163265305,
      "grad_norm": 0.6953087449073792,
      "learning_rate": 3.795918367346939e-06,
      "loss": 0.0527,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 1.2290390729904175,
      "learning_rate": 3.7142857142857146e-06,
      "loss": 0.0689,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.183673469387755,
      "grad_norm": 0.6281890869140625,
      "learning_rate": 3.6326530612244903e-06,
      "loss": 0.0647,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.2244897959183674,
      "grad_norm": 0.3096281588077545,
      "learning_rate": 3.5510204081632655e-06,
      "loss": 0.0522,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.2653061224489797,
      "grad_norm": 0.9390127062797546,
      "learning_rate": 3.469387755102041e-06,
      "loss": 0.0432,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.306122448979592,
      "grad_norm": 0.87565016746521,
      "learning_rate": 3.3877551020408168e-06,
      "loss": 0.0555,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.3469387755102042,
      "grad_norm": 1.0797837972640991,
      "learning_rate": 3.3061224489795924e-06,
      "loss": 0.0455,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.387755102040816,
      "grad_norm": 0.3658354878425598,
      "learning_rate": 3.2244897959183672e-06,
      "loss": 0.0487,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.4766336977481842,
      "learning_rate": 3.142857142857143e-06,
      "loss": 0.053,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.4693877551020407,
      "grad_norm": 0.49318933486938477,
      "learning_rate": 3.0612244897959185e-06,
      "loss": 0.0812,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.510204081632653,
      "grad_norm": 1.3475311994552612,
      "learning_rate": 2.979591836734694e-06,
      "loss": 0.0451,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.5510204081632653,
      "grad_norm": 0.36763882637023926,
      "learning_rate": 2.8979591836734694e-06,
      "loss": 0.0646,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.5918367346938775,
      "grad_norm": 3.085198402404785,
      "learning_rate": 2.816326530612245e-06,
      "loss": 0.0439,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.63265306122449,
      "grad_norm": 0.17229312658309937,
      "learning_rate": 2.7346938775510207e-06,
      "loss": 0.0288,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.673469387755102,
      "grad_norm": 1.0760900974273682,
      "learning_rate": 2.6530612244897964e-06,
      "loss": 0.0514,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 0.45855164527893066,
      "learning_rate": 2.571428571428571e-06,
      "loss": 0.0602,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.7551020408163263,
      "grad_norm": 0.15575875341892242,
      "learning_rate": 2.489795918367347e-06,
      "loss": 0.0543,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.795918367346939,
      "grad_norm": 0.779755175113678,
      "learning_rate": 2.4081632653061225e-06,
      "loss": 0.0497,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.836734693877551,
      "grad_norm": 0.7307060956954956,
      "learning_rate": 2.326530612244898e-06,
      "loss": 0.0486,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.877551020408163,
      "grad_norm": 1.062565803527832,
      "learning_rate": 2.244897959183674e-06,
      "loss": 0.0594,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.9183673469387754,
      "grad_norm": 0.3031039535999298,
      "learning_rate": 2.1632653061224495e-06,
      "loss": 0.0497,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 3.9591836734693877,
      "grad_norm": 2.310593843460083,
      "learning_rate": 2.0816326530612247e-06,
      "loss": 0.0746,
      "max_memory_allocated (GB)": 57.18,
      "memory_allocated (GB)": 50.57,
      "step": 970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6998704075813293,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0703,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.040816326530612,
      "grad_norm": 0.7492395639419556,
      "learning_rate": 1.9183673469387756e-06,
      "loss": 0.0486,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.081632653061225,
      "grad_norm": 0.7633445858955383,
      "learning_rate": 1.8367346938775512e-06,
      "loss": 0.0625,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.122448979591836,
      "grad_norm": 0.6911561489105225,
      "learning_rate": 1.7551020408163267e-06,
      "loss": 0.0632,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.163265306122449,
      "grad_norm": 0.33521902561187744,
      "learning_rate": 1.6734693877551023e-06,
      "loss": 0.0406,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.204081632653061,
      "grad_norm": 0.7509037852287292,
      "learning_rate": 1.5918367346938775e-06,
      "loss": 0.0531,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.244897959183674,
      "grad_norm": 0.5234070420265198,
      "learning_rate": 1.5102040816326532e-06,
      "loss": 0.0396,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.7997304797172546,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.05,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.326530612244898,
      "grad_norm": 0.2255077213048935,
      "learning_rate": 1.3469387755102043e-06,
      "loss": 0.0457,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.36734693877551,
      "grad_norm": 0.5182124376296997,
      "learning_rate": 1.2653061224489795e-06,
      "loss": 0.0485,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.408163265306122,
      "grad_norm": 0.35046374797821045,
      "learning_rate": 1.1836734693877552e-06,
      "loss": 0.0519,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.448979591836735,
      "grad_norm": 0.3923434615135193,
      "learning_rate": 1.1020408163265308e-06,
      "loss": 0.0507,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.489795918367347,
      "grad_norm": 0.23866137862205505,
      "learning_rate": 1.0204081632653063e-06,
      "loss": 0.0362,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.530612244897959,
      "grad_norm": 0.15117916464805603,
      "learning_rate": 9.387755102040817e-07,
      "loss": 0.0464,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 0.5993088483810425,
      "learning_rate": 8.571428571428572e-07,
      "loss": 0.0404,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.612244897959184,
      "grad_norm": 0.30265432596206665,
      "learning_rate": 7.755102040816327e-07,
      "loss": 0.0545,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.653061224489796,
      "grad_norm": 0.6385183334350586,
      "learning_rate": 6.938775510204082e-07,
      "loss": 0.0731,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.6938775510204085,
      "grad_norm": 1.128566026687622,
      "learning_rate": 6.122448979591837e-07,
      "loss": 0.0516,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.73469387755102,
      "grad_norm": 1.1660116910934448,
      "learning_rate": 5.306122448979592e-07,
      "loss": 0.0611,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.775510204081632,
      "grad_norm": 0.5327439904212952,
      "learning_rate": 4.489795918367347e-07,
      "loss": 0.0549,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.816326530612245,
      "grad_norm": 0.8764423131942749,
      "learning_rate": 3.6734693877551025e-07,
      "loss": 0.0441,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 0.47835007309913635,
      "learning_rate": 2.8571428571428575e-07,
      "loss": 0.0541,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.8979591836734695,
      "grad_norm": 1.048047661781311,
      "learning_rate": 2.0408163265306121e-07,
      "loss": 0.0731,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.938775510204081,
      "grad_norm": 0.3101171851158142,
      "learning_rate": 1.2244897959183673e-07,
      "loss": 0.0648,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 4.979591836734694,
      "grad_norm": 0.76802659034729,
      "learning_rate": 4.0816326530612253e-08,
      "loss": 0.0418,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 5.0,
      "max_memory_allocated (GB)": 60.52,
      "memory_allocated (GB)": 50.57,
      "step": 1225,
      "total_flos": 3.0598946525952e+16,
      "total_memory_available (GB)": 94.62,
      "train_loss": 0.06085505417415074,
      "train_runtime": 1020.8061,
      "train_samples_per_second": 55.51,
      "train_steps_per_second": 1.389
    }
  ],
  "logging_steps": 10,
  "max_steps": 1225,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 3.0598946525952e+16,
  "train_batch_size": 40,
  "trial_name": null,
  "trial_params": null
}