{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.7915611293012912,
  "eval_steps": 500,
  "global_step": 14500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.5140399932861328,
      "learning_rate": 3.722126480018763e-05,
      "loss": 1.6355,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 10,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.3762837648391724,
      "learning_rate": 4.8425981981596005e-05,
      "loss": 1.3277,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 20,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.4097976684570312,
      "learning_rate": 5.4980321363905964e-05,
      "loss": 1.2103,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 30,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.8763718605041504,
      "learning_rate": 5.963069916300439e-05,
      "loss": 1.1981,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 40,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7860246300697327,
      "learning_rate": 6.323781241896688e-05,
      "loss": 1.1533,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 50,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8660722970962524,
      "learning_rate": 6.618503854531432e-05,
      "loss": 1.1533,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 60,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9666918516159058,
      "learning_rate": 6.867688272967785e-05,
      "loss": 1.1299,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.17,
      "step": 70,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9307446479797363,
      "learning_rate": 7.083541634441276e-05,
      "loss": 1.1127,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 80,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1029677391052246,
      "learning_rate": 7.273937792762429e-05,
      "loss": 1.0781,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 90,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.121899127960205,
      "learning_rate": 7.444252960037526e-05,
      "loss": 1.1192,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9134292602539062,
      "learning_rate": 7.598321769544036e-05,
      "loss": 1.1036,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.0284743309020996,
      "learning_rate": 7.738975572672271e-05,
      "loss": 1.0748,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9237496852874756,
      "learning_rate": 7.86836452888091e-05,
      "loss": 1.083,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0437508821487427,
      "learning_rate": 7.988159991108621e-05,
      "loss": 1.041,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0131335258483887,
      "learning_rate": 8.09968689826852e-05,
      "loss": 1.0847,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9940443634986877,
      "learning_rate": 8.204013352582114e-05,
      "loss": 1.0519,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9474312663078308,
      "learning_rate": 8.302012992591362e-05,
      "loss": 1.0354,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0124878883361816,
      "learning_rate": 8.394409510903266e-05,
      "loss": 1.0626,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8570514917373657,
      "learning_rate": 8.481809119544634e-05,
      "loss": 1.0267,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8720101118087769,
      "learning_rate": 8.564724678178364e-05,
      "loss": 1.0212,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8913308382034302,
      "learning_rate": 8.643593929339616e-05,
      "loss": 1.0065,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.7943713068962097,
      "learning_rate": 8.718793487684874e-05,
      "loss": 1.0079,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9145928025245667,
      "learning_rate": 8.790649717038493e-05,
      "loss": 1.0043,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0095473527908325,
      "learning_rate": 8.859447290813108e-05,
      "loss": 1.0172,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8466660380363464,
      "learning_rate": 8.925436003774613e-05,
      "loss": 1.0382,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8690722584724426,
      "learning_rate": 8.988836247021747e-05,
      "loss": 0.9887,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.901485025882721,
      "learning_rate": 9.04984344913426e-05,
      "loss": 1.0159,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.7459328174591064,
      "learning_rate": 9.108631709249458e-05,
      "loss": 0.9864,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8639645576477051,
      "learning_rate": 9.16535679232489e-05,
      "loss": 0.9763,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9529710412025452,
      "learning_rate": 9.220158616409358e-05,
      "loss": 0.9864,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.697185754776001,
      "learning_rate": 9.273163331924945e-05,
      "loss": 0.9927,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.7157967686653137,
      "learning_rate": 9.324485070722952e-05,
      "loss": 1.017,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.848825991153717,
      "learning_rate": 9.374227425915866e-05,
      "loss": 1.0055,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8817904591560364,
      "learning_rate": 9.4224847107322e-05,
      "loss": 0.9788,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.7796025276184082,
      "learning_rate": 9.469343034845709e-05,
      "loss": 0.999,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0431816577911377,
      "learning_rate": 9.514881229044104e-05,
      "loss": 0.969,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8022053241729736,
      "learning_rate": 9.559171643179602e-05,
      "loss": 1.0107,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7660630345344543,
      "learning_rate": 9.602280837685472e-05,
      "loss": 0.9832,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7798706293106079,
      "learning_rate": 9.644270185252743e-05,
      "loss": 0.9871,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.6778541207313538,
      "learning_rate": 9.685196396319201e-05,
      "loss": 0.9831,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7524168491363525,
      "learning_rate": 9.725111979662076e-05,
      "loss": 0.9801,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7357635498046875,
      "learning_rate": 9.764065647480455e-05,
      "loss": 0.9778,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7956339716911316,
      "learning_rate": 9.802102672806894e-05,
      "loss": 0.981,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.6962536573410034,
      "learning_rate": 9.83926520582571e-05,
      "loss": 0.9459,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.7492054104804993,
      "learning_rate": 9.875592554640354e-05,
      "loss": 0.9785,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6673711538314819,
      "learning_rate": 9.911121435179331e-05,
      "loss": 0.985,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6954076290130615,
      "learning_rate": 9.945886194223949e-05,
      "loss": 0.975,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.7642614841461182,
      "learning_rate": 9.979919008953946e-05,
      "loss": 0.9806,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.7033060193061829,
      "learning_rate": 9.998089171974523e-05,
      "loss": 0.9694,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6798336505889893,
      "learning_rate": 9.99171974522293e-05,
      "loss": 0.9697,
      "max_memory_allocated (GB)": 83.34,
      "memory_allocated (GB)": 14.97,
      "step": 500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.7565618753433228,
      "learning_rate": 9.985350318471339e-05,
      "loss": 0.9813,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.738100528717041,
      "learning_rate": 9.978980891719746e-05,
      "loss": 0.9562,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6972020864486694,
      "learning_rate": 9.972611464968154e-05,
      "loss": 0.9444,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7740529179573059,
      "learning_rate": 9.966242038216561e-05,
      "loss": 0.9693,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.713945209980011,
      "learning_rate": 9.95987261146497e-05,
      "loss": 0.9554,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6673797965049744,
      "learning_rate": 9.953503184713376e-05,
      "loss": 0.9745,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8715893626213074,
      "learning_rate": 9.947133757961783e-05,
      "loss": 0.9689,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7116804718971252,
      "learning_rate": 9.940764331210192e-05,
      "loss": 0.9527,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6342630386352539,
      "learning_rate": 9.934394904458599e-05,
      "loss": 0.9637,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6425337791442871,
      "learning_rate": 9.928025477707007e-05,
      "loss": 0.9807,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.72014981508255,
      "learning_rate": 9.921656050955414e-05,
      "loss": 0.9873,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.660266637802124,
      "learning_rate": 9.915286624203823e-05,
      "loss": 0.9898,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6311970949172974,
      "learning_rate": 9.90891719745223e-05,
      "loss": 0.9508,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.92966628074646,
      "learning_rate": 9.902547770700638e-05,
      "loss": 0.9427,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6435150504112244,
      "learning_rate": 9.896178343949045e-05,
      "loss": 0.9434,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6682836413383484,
      "learning_rate": 9.889808917197453e-05,
      "loss": 0.9619,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0358749628067017,
      "learning_rate": 9.88343949044586e-05,
      "loss": 0.9774,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6212781667709351,
      "learning_rate": 9.877070063694269e-05,
      "loss": 0.9387,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5973750948905945,
      "learning_rate": 9.870700636942676e-05,
      "loss": 0.9479,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6477555632591248,
      "learning_rate": 9.864331210191083e-05,
      "loss": 0.972,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6448289752006531,
      "learning_rate": 9.857961783439491e-05,
      "loss": 0.9639,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6806017160415649,
      "learning_rate": 9.851592356687898e-05,
      "loss": 0.9501,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6851417422294617,
      "learning_rate": 9.845222929936306e-05,
      "loss": 0.9411,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6251152753829956,
      "learning_rate": 9.838853503184714e-05,
      "loss": 0.9818,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6990805864334106,
      "learning_rate": 9.832484076433122e-05,
      "loss": 0.961,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.7680927515029907,
      "learning_rate": 9.826114649681529e-05,
      "loss": 0.9617,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6304144263267517,
      "learning_rate": 9.819745222929937e-05,
      "loss": 0.9731,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6871405839920044,
      "learning_rate": 9.813375796178344e-05,
      "loss": 0.9498,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.644216775894165,
      "learning_rate": 9.807006369426753e-05,
      "loss": 0.9698,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6013162732124329,
      "learning_rate": 9.80063694267516e-05,
      "loss": 0.9843,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8040651082992554,
      "learning_rate": 9.794267515923568e-05,
      "loss": 0.9436,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7139380574226379,
      "learning_rate": 9.787898089171975e-05,
      "loss": 0.9364,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6355264782905579,
      "learning_rate": 9.781528662420383e-05,
      "loss": 0.9567,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7494617700576782,
      "learning_rate": 9.77515923566879e-05,
      "loss": 0.9189,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6250610947608948,
      "learning_rate": 9.768789808917197e-05,
      "loss": 0.9525,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7264392971992493,
      "learning_rate": 9.762420382165606e-05,
      "loss": 0.9183,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.626250684261322,
      "learning_rate": 9.756050955414013e-05,
      "loss": 0.938,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5986000299453735,
      "learning_rate": 9.749681528662421e-05,
      "loss": 0.9471,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6429228782653809,
      "learning_rate": 9.743312101910828e-05,
      "loss": 0.9328,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5998768210411072,
      "learning_rate": 9.736942675159237e-05,
      "loss": 0.9378,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7069288492202759,
      "learning_rate": 9.730573248407644e-05,
      "loss": 0.9636,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6847216486930847,
      "learning_rate": 9.724203821656052e-05,
      "loss": 0.954,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5854921936988831,
      "learning_rate": 9.717834394904459e-05,
      "loss": 0.9331,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.693719208240509,
      "learning_rate": 9.711464968152867e-05,
      "loss": 0.9195,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.6119904518127441,
      "learning_rate": 9.705095541401274e-05,
      "loss": 0.9364,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.6169638633728027,
      "learning_rate": 9.698726114649683e-05,
      "loss": 0.9508,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.6071642637252808,
      "learning_rate": 9.69235668789809e-05,
      "loss": 0.9629,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5701698660850525,
      "learning_rate": 9.685987261146498e-05,
      "loss": 0.9256,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.6308379173278809,
      "learning_rate": 9.679617834394905e-05,
      "loss": 0.949,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5775448083877563,
      "learning_rate": 9.673248407643312e-05,
      "loss": 0.9397,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 1000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.6638918519020081,
      "learning_rate": 9.66687898089172e-05,
      "loss": 0.9468,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 1010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6384360194206238,
      "learning_rate": 9.660509554140128e-05,
      "loss": 0.9562,
      "max_memory_allocated (GB)": 94.36,
      "memory_allocated (GB)": 14.97,
      "step": 1020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6402448415756226,
      "learning_rate": 9.654140127388536e-05,
      "loss": 0.9366,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6450813412666321,
      "learning_rate": 9.647770700636943e-05,
      "loss": 0.9163,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6314339637756348,
      "learning_rate": 9.641401273885351e-05,
      "loss": 0.9425,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5695529580116272,
      "learning_rate": 9.635031847133758e-05,
      "loss": 0.938,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6315151453018188,
      "learning_rate": 9.628662420382167e-05,
      "loss": 0.9495,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6656625866889954,
      "learning_rate": 9.622292993630574e-05,
      "loss": 0.9412,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6286743879318237,
      "learning_rate": 9.615923566878982e-05,
      "loss": 0.9577,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.596004843711853,
      "learning_rate": 9.609554140127389e-05,
      "loss": 0.9758,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5832968354225159,
      "learning_rate": 9.603184713375797e-05,
      "loss": 0.9375,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5997422933578491,
      "learning_rate": 9.596815286624204e-05,
      "loss": 0.905,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6267014145851135,
      "learning_rate": 9.590445859872613e-05,
      "loss": 0.9496,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6150197386741638,
      "learning_rate": 9.58407643312102e-05,
      "loss": 0.9399,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5876752734184265,
      "learning_rate": 9.577707006369427e-05,
      "loss": 0.9193,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5847297310829163,
      "learning_rate": 9.571337579617835e-05,
      "loss": 0.9299,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6613140106201172,
      "learning_rate": 9.564968152866242e-05,
      "loss": 0.9367,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.625577986240387,
      "learning_rate": 9.55859872611465e-05,
      "loss": 0.9678,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6265081167221069,
      "learning_rate": 9.552229299363058e-05,
      "loss": 0.9267,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6331568956375122,
      "learning_rate": 9.545859872611466e-05,
      "loss": 0.9222,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5775994658470154,
      "learning_rate": 9.539490445859873e-05,
      "loss": 0.9283,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6138973236083984,
      "learning_rate": 9.533121019108281e-05,
      "loss": 0.9296,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6381206512451172,
      "learning_rate": 9.526751592356688e-05,
      "loss": 0.9388,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.604809045791626,
      "learning_rate": 9.520382165605097e-05,
      "loss": 0.9428,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6210758686065674,
      "learning_rate": 9.514012738853504e-05,
      "loss": 0.9438,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6653516292572021,
      "learning_rate": 9.507643312101912e-05,
      "loss": 0.9396,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5696619153022766,
      "learning_rate": 9.501273885350319e-05,
      "loss": 0.946,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6504731178283691,
      "learning_rate": 9.494904458598726e-05,
      "loss": 0.9603,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6198241710662842,
      "learning_rate": 9.488535031847135e-05,
      "loss": 0.9091,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6397411823272705,
      "learning_rate": 9.482165605095542e-05,
      "loss": 0.933,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6273577809333801,
      "learning_rate": 9.47579617834395e-05,
      "loss": 0.9171,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.677662193775177,
      "learning_rate": 9.469426751592357e-05,
      "loss": 0.9361,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6163740754127502,
      "learning_rate": 9.463057324840765e-05,
      "loss": 0.9396,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6019936203956604,
      "learning_rate": 9.456687898089172e-05,
      "loss": 0.9417,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6324911117553711,
      "learning_rate": 9.45031847133758e-05,
      "loss": 0.9254,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5347614288330078,
      "learning_rate": 9.443949044585988e-05,
      "loss": 0.9259,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5964996814727783,
      "learning_rate": 9.437579617834396e-05,
      "loss": 0.9164,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6841687560081482,
      "learning_rate": 9.431210191082803e-05,
      "loss": 0.9043,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6945049166679382,
      "learning_rate": 9.424840764331211e-05,
      "loss": 0.9339,
      "max_memory_allocated (GB)": 94.43,
      "memory_allocated (GB)": 14.97,
      "step": 1390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6683288812637329,
      "learning_rate": 9.418471337579618e-05,
      "loss": 0.929,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5797848701477051,
      "learning_rate": 9.412101910828027e-05,
      "loss": 0.9099,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6015276312828064,
      "learning_rate": 9.405732484076434e-05,
      "loss": 0.9411,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6140068769454956,
      "learning_rate": 9.399363057324841e-05,
      "loss": 0.9213,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6447374820709229,
      "learning_rate": 9.392993630573249e-05,
      "loss": 0.8924,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6505595445632935,
      "learning_rate": 9.386624203821656e-05,
      "loss": 0.9281,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6256375312805176,
      "learning_rate": 9.380254777070065e-05,
      "loss": 0.9577,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6421294808387756,
      "learning_rate": 9.373885350318472e-05,
      "loss": 0.9125,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.7359905242919922,
      "learning_rate": 9.36751592356688e-05,
      "loss": 0.909,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6073474884033203,
      "learning_rate": 9.361146496815287e-05,
      "loss": 0.9433,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5646604299545288,
      "learning_rate": 9.354777070063695e-05,
      "loss": 0.9174,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7164282202720642,
      "learning_rate": 9.348407643312102e-05,
      "loss": 0.934,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6293795108795166,
      "learning_rate": 9.342038216560511e-05,
      "loss": 0.9134,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6097316145896912,
      "learning_rate": 9.335668789808918e-05,
      "loss": 0.913,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6106476783752441,
      "learning_rate": 9.329299363057326e-05,
      "loss": 0.9029,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6438948512077332,
      "learning_rate": 9.322929936305733e-05,
      "loss": 0.9444,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.633876621723175,
      "learning_rate": 9.316560509554141e-05,
      "loss": 0.9239,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6850993633270264,
      "learning_rate": 9.310191082802549e-05,
      "loss": 0.9173,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5955223441123962,
      "learning_rate": 9.303821656050956e-05,
      "loss": 0.9402,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6091153025627136,
      "learning_rate": 9.297452229299364e-05,
      "loss": 0.9753,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5890401005744934,
      "learning_rate": 9.291082802547771e-05,
      "loss": 0.9372,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6389315724372864,
      "learning_rate": 9.284713375796179e-05,
      "loss": 0.8853,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5620580315589905,
      "learning_rate": 9.278343949044586e-05,
      "loss": 0.942,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5825024843215942,
      "learning_rate": 9.271974522292995e-05,
      "loss": 0.9107,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.544722318649292,
      "learning_rate": 9.265605095541402e-05,
      "loss": 0.907,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6667870879173279,
      "learning_rate": 9.25923566878981e-05,
      "loss": 0.9207,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.599666953086853,
      "learning_rate": 9.252866242038217e-05,
      "loss": 0.9117,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6073837280273438,
      "learning_rate": 9.246496815286625e-05,
      "loss": 0.93,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6553378701210022,
      "learning_rate": 9.240127388535032e-05,
      "loss": 0.9076,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6511078476905823,
      "learning_rate": 9.233757961783441e-05,
      "loss": 0.9015,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5960680246353149,
      "learning_rate": 9.227388535031848e-05,
      "loss": 0.915,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6395820379257202,
      "learning_rate": 9.221019108280255e-05,
      "loss": 0.9336,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6158560514450073,
      "learning_rate": 9.214649681528663e-05,
      "loss": 0.9315,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5883167386054993,
      "learning_rate": 9.20828025477707e-05,
      "loss": 0.9243,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5660105347633362,
      "learning_rate": 9.201910828025479e-05,
      "loss": 0.9141,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5553457736968994,
      "learning_rate": 9.195541401273886e-05,
      "loss": 0.9309,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6037126183509827,
      "learning_rate": 9.189171974522294e-05,
      "loss": 0.9239,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6285903453826904,
      "learning_rate": 9.182802547770701e-05,
      "loss": 0.9223,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5817345976829529,
      "learning_rate": 9.17643312101911e-05,
      "loss": 0.9434,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6466053128242493,
      "learning_rate": 9.170063694267516e-05,
      "loss": 0.928,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6852598190307617,
      "learning_rate": 9.163694267515925e-05,
      "loss": 0.9415,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5549845099449158,
      "learning_rate": 9.157324840764332e-05,
      "loss": 0.902,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5786306262016296,
      "learning_rate": 9.15095541401274e-05,
      "loss": 0.9417,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6237860321998596,
      "learning_rate": 9.144585987261147e-05,
      "loss": 0.9348,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6448583006858826,
      "learning_rate": 9.138216560509555e-05,
      "loss": 0.972,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5415520668029785,
      "learning_rate": 9.131847133757963e-05,
      "loss": 0.9277,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5853114128112793,
      "learning_rate": 9.12547770700637e-05,
      "loss": 0.9421,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5774340033531189,
      "learning_rate": 9.119108280254778e-05,
      "loss": 0.9241,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6107351779937744,
      "learning_rate": 9.112738853503185e-05,
      "loss": 0.9133,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6424444317817688,
      "learning_rate": 9.106369426751593e-05,
      "loss": 0.9546,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5167615413665771,
      "learning_rate": 9.1e-05,
      "loss": 0.9397,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6501650214195251,
      "learning_rate": 9.093630573248409e-05,
      "loss": 0.9138,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6346274614334106,
      "learning_rate": 9.087261146496816e-05,
      "loss": 0.9447,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6025852560997009,
      "learning_rate": 9.080891719745224e-05,
      "loss": 0.8942,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5486747026443481,
      "learning_rate": 9.074522292993631e-05,
      "loss": 0.9147,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6243796348571777,
      "learning_rate": 9.06815286624204e-05,
      "loss": 0.9298,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6881577372550964,
      "learning_rate": 9.061783439490446e-05,
      "loss": 0.9086,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6005951166152954,
      "learning_rate": 9.055414012738855e-05,
      "loss": 0.9158,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5638641119003296,
      "learning_rate": 9.049044585987262e-05,
      "loss": 0.9297,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.505285918712616,
      "learning_rate": 9.04267515923567e-05,
      "loss": 0.9313,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 1990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6503742337226868,
      "learning_rate": 9.036305732484077e-05,
      "loss": 0.8957,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5132340788841248,
      "learning_rate": 9.029936305732484e-05,
      "loss": 0.9292,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6249287724494934,
      "learning_rate": 9.023566878980893e-05,
      "loss": 0.9238,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7638124823570251,
      "learning_rate": 9.0171974522293e-05,
      "loss": 0.9347,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5246148705482483,
      "learning_rate": 9.010828025477708e-05,
      "loss": 0.9245,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7826683521270752,
      "learning_rate": 9.004458598726115e-05,
      "loss": 0.9292,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5587345361709595,
      "learning_rate": 8.998089171974523e-05,
      "loss": 0.928,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5445036292076111,
      "learning_rate": 8.99171974522293e-05,
      "loss": 0.8869,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5840720534324646,
      "learning_rate": 8.985350318471339e-05,
      "loss": 0.9201,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.649189829826355,
      "learning_rate": 8.978980891719746e-05,
      "loss": 0.9178,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.6557360291481018,
      "learning_rate": 8.972611464968154e-05,
      "loss": 0.9426,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.6620033383369446,
      "learning_rate": 8.966242038216561e-05,
      "loss": 0.8703,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.6042346358299255,
      "learning_rate": 8.95987261146497e-05,
      "loss": 0.9214,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5340855121612549,
      "learning_rate": 8.953503184713377e-05,
      "loss": 0.9448,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5642148852348328,
      "learning_rate": 8.947133757961785e-05,
      "loss": 0.9163,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5716170072555542,
      "learning_rate": 8.940764331210192e-05,
      "loss": 0.9159,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5505129098892212,
      "learning_rate": 8.934394904458599e-05,
      "loss": 0.9149,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.6169817447662354,
      "learning_rate": 8.928025477707007e-05,
      "loss": 0.9258,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5575990676879883,
      "learning_rate": 8.921656050955414e-05,
      "loss": 0.916,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5870167016983032,
      "learning_rate": 8.915286624203823e-05,
      "loss": 0.8911,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5575961470603943,
      "learning_rate": 8.90891719745223e-05,
      "loss": 0.9301,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.6070531010627747,
      "learning_rate": 8.902547770700638e-05,
      "loss": 0.9031,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5783848762512207,
      "learning_rate": 8.896178343949045e-05,
      "loss": 0.8998,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6437045931816101,
      "learning_rate": 8.889808917197453e-05,
      "loss": 0.8992,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.608791708946228,
      "learning_rate": 8.88343949044586e-05,
      "loss": 0.9362,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6209393739700317,
      "learning_rate": 8.877070063694269e-05,
      "loss": 0.92,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5823927521705627,
      "learning_rate": 8.870700636942676e-05,
      "loss": 0.9126,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6524648666381836,
      "learning_rate": 8.864331210191084e-05,
      "loss": 0.9168,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5724565982818604,
      "learning_rate": 8.857961783439491e-05,
      "loss": 0.9302,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6382232904434204,
      "learning_rate": 8.851592356687898e-05,
      "loss": 0.9029,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6507483720779419,
      "learning_rate": 8.845222929936307e-05,
      "loss": 0.9335,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.6949073076248169,
      "learning_rate": 8.838853503184714e-05,
      "loss": 0.9127,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5478443503379822,
      "learning_rate": 8.832484076433122e-05,
      "loss": 0.9072,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5452962517738342,
      "learning_rate": 8.826114649681529e-05,
      "loss": 0.8968,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5772174000740051,
      "learning_rate": 8.819745222929937e-05,
      "loss": 0.9237,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5888592004776001,
      "learning_rate": 8.813375796178344e-05,
      "loss": 0.9021,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5698549747467041,
      "learning_rate": 8.807006369426753e-05,
      "loss": 0.9173,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5790499448776245,
      "learning_rate": 8.80063694267516e-05,
      "loss": 0.9431,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.6478803157806396,
      "learning_rate": 8.794267515923568e-05,
      "loss": 0.9167,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6243005394935608,
      "learning_rate": 8.787898089171975e-05,
      "loss": 0.9052,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6000833511352539,
      "learning_rate": 8.781528662420383e-05,
      "loss": 0.8992,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5554850101470947,
      "learning_rate": 8.77515923566879e-05,
      "loss": 0.9277,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6706589460372925,
      "learning_rate": 8.768789808917199e-05,
      "loss": 0.9232,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6054849624633789,
      "learning_rate": 8.762420382165606e-05,
      "loss": 0.8939,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5416914224624634,
      "learning_rate": 8.756050955414013e-05,
      "loss": 0.8967,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6584843993186951,
      "learning_rate": 8.749681528662421e-05,
      "loss": 0.9156,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6535277366638184,
      "learning_rate": 8.743312101910828e-05,
      "loss": 0.9269,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.6150842905044556,
      "learning_rate": 8.736942675159237e-05,
      "loss": 0.9336,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5612450838088989,
      "learning_rate": 8.730573248407644e-05,
      "loss": 0.897,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5459498167037964,
      "learning_rate": 8.724203821656052e-05,
      "loss": 0.9181,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5725070238113403,
      "learning_rate": 8.717834394904459e-05,
      "loss": 0.9039,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5893815755844116,
      "learning_rate": 8.711464968152867e-05,
      "loss": 0.8934,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.6075538992881775,
      "learning_rate": 8.705095541401274e-05,
      "loss": 0.9199,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.7179542183876038,
      "learning_rate": 8.698726114649683e-05,
      "loss": 0.8964,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.6916159987449646,
      "learning_rate": 8.69235668789809e-05,
      "loss": 0.9228,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5973563194274902,
      "learning_rate": 8.685987261146498e-05,
      "loss": 0.9109,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5797328352928162,
      "learning_rate": 8.679617834394905e-05,
      "loss": 0.9006,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6468456983566284,
      "learning_rate": 8.673248407643314e-05,
      "loss": 0.9306,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5766948461532593,
      "learning_rate": 8.66687898089172e-05,
      "loss": 0.8813,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5691495537757874,
      "learning_rate": 8.660509554140128e-05,
      "loss": 0.9204,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6611936092376709,
      "learning_rate": 8.654140127388536e-05,
      "loss": 0.935,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5880815386772156,
      "learning_rate": 8.647770700636943e-05,
      "loss": 0.9257,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.581061840057373,
      "learning_rate": 8.641401273885351e-05,
      "loss": 0.8615,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5449854731559753,
      "learning_rate": 8.635031847133758e-05,
      "loss": 0.8968,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5454363822937012,
      "learning_rate": 8.628662420382167e-05,
      "loss": 0.9069,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.567973792552948,
      "learning_rate": 8.622292993630574e-05,
      "loss": 0.9045,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5044744610786438,
      "learning_rate": 8.615923566878982e-05,
      "loss": 0.9307,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5225447416305542,
      "learning_rate": 8.609554140127389e-05,
      "loss": 0.9395,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.6234666109085083,
      "learning_rate": 8.603184713375798e-05,
      "loss": 0.9077,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5319868922233582,
      "learning_rate": 8.596815286624205e-05,
      "loss": 0.9213,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5291190147399902,
      "learning_rate": 8.590445859872613e-05,
      "loss": 0.909,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.6539480686187744,
      "learning_rate": 8.58407643312102e-05,
      "loss": 0.8908,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6586701273918152,
      "learning_rate": 8.577707006369428e-05,
      "loss": 0.9112,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5719462633132935,
      "learning_rate": 8.571337579617835e-05,
      "loss": 0.8926,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6857839822769165,
      "learning_rate": 8.564968152866242e-05,
      "loss": 0.9244,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5097509026527405,
      "learning_rate": 8.55859872611465e-05,
      "loss": 0.9195,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.626815676689148,
      "learning_rate": 8.552229299363058e-05,
      "loss": 0.9034,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5483399033546448,
      "learning_rate": 8.545859872611466e-05,
      "loss": 0.9295,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5154463648796082,
      "learning_rate": 8.539490445859873e-05,
      "loss": 0.936,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5392872095108032,
      "learning_rate": 8.533121019108281e-05,
      "loss": 0.914,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5366926193237305,
      "learning_rate": 8.526751592356688e-05,
      "loss": 0.9067,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5744665861129761,
      "learning_rate": 8.520382165605097e-05,
      "loss": 0.9105,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5835170745849609,
      "learning_rate": 8.514012738853504e-05,
      "loss": 0.917,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.6016770601272583,
      "learning_rate": 8.507643312101912e-05,
      "loss": 0.9023,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5826542973518372,
      "learning_rate": 8.501273885350319e-05,
      "loss": 0.8775,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5647284388542175,
      "learning_rate": 8.494904458598728e-05,
      "loss": 0.8926,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5165654420852661,
      "learning_rate": 8.488535031847135e-05,
      "loss": 0.9084,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5151524543762207,
      "learning_rate": 8.482165605095542e-05,
      "loss": 0.9008,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.581031322479248,
      "learning_rate": 8.47579617834395e-05,
      "loss": 0.9088,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5811273455619812,
      "learning_rate": 8.469426751592357e-05,
      "loss": 0.8841,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5163042545318604,
      "learning_rate": 8.463057324840765e-05,
      "loss": 0.8814,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5550446510314941,
      "learning_rate": 8.456687898089172e-05,
      "loss": 0.8873,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6043127775192261,
      "learning_rate": 8.450318471337581e-05,
      "loss": 0.9096,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5609890818595886,
      "learning_rate": 8.443949044585988e-05,
      "loss": 0.9039,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.578622579574585,
      "learning_rate": 8.437579617834396e-05,
      "loss": 0.9134,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5179211497306824,
      "learning_rate": 8.431210191082803e-05,
      "loss": 0.8848,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5544730424880981,
      "learning_rate": 8.424840764331212e-05,
      "loss": 0.881,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5601953864097595,
      "learning_rate": 8.418471337579619e-05,
      "loss": 0.9211,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5997374653816223,
      "learning_rate": 8.412101910828027e-05,
      "loss": 0.9003,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5613746643066406,
      "learning_rate": 8.405732484076434e-05,
      "loss": 0.922,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 2990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5812062621116638,
      "learning_rate": 8.399363057324842e-05,
      "loss": 0.9153,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.6788537502288818,
      "learning_rate": 8.392993630573249e-05,
      "loss": 0.898,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5436647534370422,
      "learning_rate": 8.386624203821656e-05,
      "loss": 0.9003,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.7214387655258179,
      "learning_rate": 8.380254777070065e-05,
      "loss": 0.912,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5388379096984863,
      "learning_rate": 8.373885350318472e-05,
      "loss": 0.9196,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5177743434906006,
      "learning_rate": 8.36751592356688e-05,
      "loss": 0.8901,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.584418535232544,
      "learning_rate": 8.361146496815287e-05,
      "loss": 0.8822,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.6588314771652222,
      "learning_rate": 8.354777070063695e-05,
      "loss": 0.9292,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.6969806551933289,
      "learning_rate": 8.348407643312102e-05,
      "loss": 0.897,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.567183256149292,
      "learning_rate": 8.342038216560511e-05,
      "loss": 0.9072,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5422699451446533,
      "learning_rate": 8.335668789808918e-05,
      "loss": 0.9206,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.606003999710083,
      "learning_rate": 8.329299363057326e-05,
      "loss": 0.9029,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5588945150375366,
      "learning_rate": 8.322929936305733e-05,
      "loss": 0.8914,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.7723147869110107,
      "learning_rate": 8.316560509554142e-05,
      "loss": 0.9052,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5690342783927917,
      "learning_rate": 8.310191082802549e-05,
      "loss": 0.8959,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4744355082511902,
      "learning_rate": 8.303821656050957e-05,
      "loss": 0.9225,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5398902893066406,
      "learning_rate": 8.297452229299364e-05,
      "loss": 0.8994,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5458796620368958,
      "learning_rate": 8.291082802547771e-05,
      "loss": 0.9018,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.6232591867446899,
      "learning_rate": 8.28471337579618e-05,
      "loss": 0.9131,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5887508988380432,
      "learning_rate": 8.278343949044586e-05,
      "loss": 0.8985,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5530678629875183,
      "learning_rate": 8.271974522292995e-05,
      "loss": 0.8922,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5570492744445801,
      "learning_rate": 8.265605095541402e-05,
      "loss": 0.9188,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5902305841445923,
      "learning_rate": 8.25923566878981e-05,
      "loss": 0.8917,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5945755243301392,
      "learning_rate": 8.252866242038217e-05,
      "loss": 0.8966,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5670562982559204,
      "learning_rate": 8.246496815286626e-05,
      "loss": 0.9239,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.542846143245697,
      "learning_rate": 8.240127388535033e-05,
      "loss": 0.9198,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5747506618499756,
      "learning_rate": 8.233757961783441e-05,
      "loss": 0.8817,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5135393738746643,
      "learning_rate": 8.227388535031848e-05,
      "loss": 0.9082,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5435426831245422,
      "learning_rate": 8.221019108280256e-05,
      "loss": 0.9095,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5716527700424194,
      "learning_rate": 8.214649681528663e-05,
      "loss": 0.9126,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5149152278900146,
      "learning_rate": 8.20828025477707e-05,
      "loss": 0.8981,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.6820017695426941,
      "learning_rate": 8.201910828025479e-05,
      "loss": 0.8949,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5794596076011658,
      "learning_rate": 8.195541401273886e-05,
      "loss": 0.9184,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5630435943603516,
      "learning_rate": 8.189171974522294e-05,
      "loss": 0.9107,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5356283187866211,
      "learning_rate": 8.182802547770701e-05,
      "loss": 0.909,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5827937126159668,
      "learning_rate": 8.17643312101911e-05,
      "loss": 0.9021,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.7308997511863708,
      "learning_rate": 8.170063694267516e-05,
      "loss": 0.8983,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5804486274719238,
      "learning_rate": 8.163694267515925e-05,
      "loss": 0.8959,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.6027332544326782,
      "learning_rate": 8.157324840764332e-05,
      "loss": 0.9061,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5659801959991455,
      "learning_rate": 8.15095541401274e-05,
      "loss": 0.893,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.6081441044807434,
      "learning_rate": 8.144585987261147e-05,
      "loss": 0.8899,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.703420102596283,
      "learning_rate": 8.138216560509556e-05,
      "loss": 0.9179,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5405531525611877,
      "learning_rate": 8.131847133757963e-05,
      "loss": 0.9216,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.6257489919662476,
      "learning_rate": 8.125477707006371e-05,
      "loss": 0.9134,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5303512215614319,
      "learning_rate": 8.119108280254778e-05,
      "loss": 0.8945,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.7373655438423157,
      "learning_rate": 8.112738853503185e-05,
      "loss": 0.8916,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5460899472236633,
      "learning_rate": 8.106369426751593e-05,
      "loss": 0.8931,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5521990060806274,
      "learning_rate": 8.1e-05,
      "loss": 0.8566,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5547364950180054,
      "learning_rate": 8.093630573248409e-05,
      "loss": 0.897,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5535370111465454,
      "learning_rate": 8.087261146496816e-05,
      "loss": 0.8899,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5962079763412476,
      "learning_rate": 8.080891719745224e-05,
      "loss": 0.9031,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5219169855117798,
      "learning_rate": 8.074522292993631e-05,
      "loss": 0.9025,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.6031842827796936,
      "learning_rate": 8.06815286624204e-05,
      "loss": 0.8659,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5150040984153748,
      "learning_rate": 8.061783439490447e-05,
      "loss": 0.9043,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5902099013328552,
      "learning_rate": 8.055414012738855e-05,
      "loss": 0.9039,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5089321732521057,
      "learning_rate": 8.049044585987262e-05,
      "loss": 0.8955,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5322747826576233,
      "learning_rate": 8.04267515923567e-05,
      "loss": 0.896,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5516270995140076,
      "learning_rate": 8.036305732484077e-05,
      "loss": 0.9095,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5034855604171753,
      "learning_rate": 8.029936305732486e-05,
      "loss": 0.9156,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.6564231514930725,
      "learning_rate": 8.023566878980893e-05,
      "loss": 0.8812,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.530149519443512,
      "learning_rate": 8.0171974522293e-05,
      "loss": 0.9207,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5480339527130127,
      "learning_rate": 8.010828025477708e-05,
      "loss": 0.8734,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5565770864486694,
      "learning_rate": 8.004458598726115e-05,
      "loss": 0.8867,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5027381181716919,
      "learning_rate": 7.998089171974522e-05,
      "loss": 0.8744,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5091252326965332,
      "learning_rate": 7.991719745222929e-05,
      "loss": 0.9363,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5606957077980042,
      "learning_rate": 7.985350318471337e-05,
      "loss": 0.9081,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5584906935691833,
      "learning_rate": 7.978980891719744e-05,
      "loss": 0.8972,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5510685443878174,
      "learning_rate": 7.972611464968153e-05,
      "loss": 0.9095,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5009828209877014,
      "learning_rate": 7.96624203821656e-05,
      "loss": 0.8797,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5698838829994202,
      "learning_rate": 7.959872611464968e-05,
      "loss": 0.9026,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5882912874221802,
      "learning_rate": 7.953503184713375e-05,
      "loss": 0.9035,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5969718098640442,
      "learning_rate": 7.947133757961784e-05,
      "loss": 0.8707,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5464320182800293,
      "learning_rate": 7.94076433121019e-05,
      "loss": 0.8946,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5218807458877563,
      "learning_rate": 7.934394904458599e-05,
      "loss": 0.893,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5073594450950623,
      "learning_rate": 7.928025477707006e-05,
      "loss": 0.9295,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5315717458724976,
      "learning_rate": 7.921656050955414e-05,
      "loss": 0.874,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.640521764755249,
      "learning_rate": 7.915286624203821e-05,
      "loss": 0.8927,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6283209323883057,
      "learning_rate": 7.90891719745223e-05,
      "loss": 0.8859,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5683282017707825,
      "learning_rate": 7.902547770700637e-05,
      "loss": 0.922,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5616042613983154,
      "learning_rate": 7.896178343949044e-05,
      "loss": 0.9092,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.582125186920166,
      "learning_rate": 7.889808917197452e-05,
      "loss": 0.8862,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5669243335723877,
      "learning_rate": 7.883439490445859e-05,
      "loss": 0.9283,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6402415633201599,
      "learning_rate": 7.877070063694268e-05,
      "loss": 0.9037,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5741037726402283,
      "learning_rate": 7.870700636942675e-05,
      "loss": 0.9033,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5755082368850708,
      "learning_rate": 7.864331210191083e-05,
      "loss": 0.8608,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5075112581253052,
      "learning_rate": 7.85796178343949e-05,
      "loss": 0.8774,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5267348289489746,
      "learning_rate": 7.851592356687898e-05,
      "loss": 0.8916,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5770332217216492,
      "learning_rate": 7.845222929936305e-05,
      "loss": 0.8453,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5677030086517334,
      "learning_rate": 7.838853503184714e-05,
      "loss": 0.895,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.577979564666748,
      "learning_rate": 7.832484076433121e-05,
      "loss": 0.9107,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5809403657913208,
      "learning_rate": 7.826114649681529e-05,
      "loss": 0.8957,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5649895668029785,
      "learning_rate": 7.819745222929936e-05,
      "loss": 0.8991,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5361667275428772,
      "learning_rate": 7.813375796178344e-05,
      "loss": 0.8966,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5365725159645081,
      "learning_rate": 7.807006369426751e-05,
      "loss": 0.8991,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5496806502342224,
      "learning_rate": 7.800636942675158e-05,
      "loss": 0.8738,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5392926931381226,
      "learning_rate": 7.794267515923567e-05,
      "loss": 0.8847,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5666927099227905,
      "learning_rate": 7.787898089171974e-05,
      "loss": 0.8732,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5134681463241577,
      "learning_rate": 7.781528662420382e-05,
      "loss": 0.896,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4999007284641266,
      "learning_rate": 7.775159235668789e-05,
      "loss": 0.8835,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.6310585141181946,
      "learning_rate": 7.768789808917198e-05,
      "loss": 0.8467,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 3990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.6398172974586487,
      "learning_rate": 7.762420382165605e-05,
      "loss": 0.9006,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5285815000534058,
      "learning_rate": 7.756050955414013e-05,
      "loss": 0.8968,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5276922583580017,
      "learning_rate": 7.74968152866242e-05,
      "loss": 0.8926,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5247613787651062,
      "learning_rate": 7.743312101910828e-05,
      "loss": 0.8891,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5284489393234253,
      "learning_rate": 7.736942675159235e-05,
      "loss": 0.9014,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6317757964134216,
      "learning_rate": 7.730573248407644e-05,
      "loss": 0.8714,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5911426544189453,
      "learning_rate": 7.724203821656051e-05,
      "loss": 0.8982,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5255340337753296,
      "learning_rate": 7.717834394904459e-05,
      "loss": 0.8884,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5683048963546753,
      "learning_rate": 7.711464968152866e-05,
      "loss": 0.8908,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5523961782455444,
      "learning_rate": 7.705095541401273e-05,
      "loss": 0.8745,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5336757302284241,
      "learning_rate": 7.698726114649682e-05,
      "loss": 0.8728,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5282894372940063,
      "learning_rate": 7.692356687898089e-05,
      "loss": 0.8826,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5190914869308472,
      "learning_rate": 7.685987261146497e-05,
      "loss": 0.8914,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5775402188301086,
      "learning_rate": 7.679617834394904e-05,
      "loss": 0.9145,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5857258439064026,
      "learning_rate": 7.673248407643312e-05,
      "loss": 0.8854,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.567958652973175,
      "learning_rate": 7.666878980891719e-05,
      "loss": 0.8926,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.576958954334259,
      "learning_rate": 7.660509554140128e-05,
      "loss": 0.906,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5861126184463501,
      "learning_rate": 7.654140127388535e-05,
      "loss": 0.898,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5804855227470398,
      "learning_rate": 7.647770700636943e-05,
      "loss": 0.8925,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5505447387695312,
      "learning_rate": 7.64140127388535e-05,
      "loss": 0.8918,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5158912539482117,
      "learning_rate": 7.635031847133758e-05,
      "loss": 0.9067,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.6285148859024048,
      "learning_rate": 7.628662420382165e-05,
      "loss": 0.8937,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.60187828540802,
      "learning_rate": 7.622292993630572e-05,
      "loss": 0.9046,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5804048776626587,
      "learning_rate": 7.615923566878981e-05,
      "loss": 0.8849,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5304928421974182,
      "learning_rate": 7.609554140127388e-05,
      "loss": 0.89,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5056673288345337,
      "learning_rate": 7.603184713375796e-05,
      "loss": 0.9253,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5457956194877625,
      "learning_rate": 7.596815286624203e-05,
      "loss": 0.8704,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5564907789230347,
      "learning_rate": 7.590445859872612e-05,
      "loss": 0.8779,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5742987990379333,
      "learning_rate": 7.584076433121019e-05,
      "loss": 0.8948,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5531176328659058,
      "learning_rate": 7.577707006369427e-05,
      "loss": 0.875,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5973535180091858,
      "learning_rate": 7.571337579617834e-05,
      "loss": 0.8809,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5450004935264587,
      "learning_rate": 7.564968152866242e-05,
      "loss": 0.8827,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.6022359728813171,
      "learning_rate": 7.55859872611465e-05,
      "loss": 0.8934,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5830252170562744,
      "learning_rate": 7.552229299363058e-05,
      "loss": 0.9135,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6169658303260803,
      "learning_rate": 7.545859872611465e-05,
      "loss": 0.8866,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5571846961975098,
      "learning_rate": 7.539490445859873e-05,
      "loss": 0.8698,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6459663510322571,
      "learning_rate": 7.53312101910828e-05,
      "loss": 0.8953,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.629196047782898,
      "learning_rate": 7.526751592356687e-05,
      "loss": 0.8968,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6803416013717651,
      "learning_rate": 7.520382165605096e-05,
      "loss": 0.8984,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5971508026123047,
      "learning_rate": 7.514012738853503e-05,
      "loss": 0.9216,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5492663979530334,
      "learning_rate": 7.507643312101911e-05,
      "loss": 0.8535,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.581967294216156,
      "learning_rate": 7.501273885350318e-05,
      "loss": 0.8945,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5171147584915161,
      "learning_rate": 7.494904458598726e-05,
      "loss": 0.8509,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5407620668411255,
      "learning_rate": 7.488535031847133e-05,
      "loss": 0.879,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5820145606994629,
      "learning_rate": 7.482165605095542e-05,
      "loss": 0.9072,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5734186768531799,
      "learning_rate": 7.475796178343949e-05,
      "loss": 0.9084,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5551985502243042,
      "learning_rate": 7.469426751592357e-05,
      "loss": 0.869,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.6070141196250916,
      "learning_rate": 7.463057324840764e-05,
      "loss": 0.8956,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5512224435806274,
      "learning_rate": 7.456687898089172e-05,
      "loss": 0.8824,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5932292342185974,
      "learning_rate": 7.45031847133758e-05,
      "loss": 0.8911,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5633782744407654,
      "learning_rate": 7.443949044585988e-05,
      "loss": 0.884,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5297742486000061,
      "learning_rate": 7.437579617834395e-05,
      "loss": 0.8981,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.49679142236709595,
      "learning_rate": 7.431210191082802e-05,
      "loss": 0.8798,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6201618909835815,
      "learning_rate": 7.42484076433121e-05,
      "loss": 0.8898,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5749346017837524,
      "learning_rate": 7.418471337579617e-05,
      "loss": 0.8833,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.539679229259491,
      "learning_rate": 7.412101910828026e-05,
      "loss": 0.9201,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5480098128318787,
      "learning_rate": 7.405732484076433e-05,
      "loss": 0.8727,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6261775493621826,
      "learning_rate": 7.399363057324841e-05,
      "loss": 0.9153,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5050274729728699,
      "learning_rate": 7.392993630573248e-05,
      "loss": 0.9331,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.6069965958595276,
      "learning_rate": 7.386624203821656e-05,
      "loss": 0.8867,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.573829174041748,
      "learning_rate": 7.380254777070063e-05,
      "loss": 0.9034,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5090081095695496,
      "learning_rate": 7.373885350318472e-05,
      "loss": 0.8707,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5630531311035156,
      "learning_rate": 7.367515923566879e-05,
      "loss": 0.8904,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.559716522693634,
      "learning_rate": 7.361146496815287e-05,
      "loss": 0.8998,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.49673840403556824,
      "learning_rate": 7.354777070063694e-05,
      "loss": 0.9104,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5938435792922974,
      "learning_rate": 7.348407643312101e-05,
      "loss": 0.8874,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5486466884613037,
      "learning_rate": 7.34203821656051e-05,
      "loss": 0.8756,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5997244119644165,
      "learning_rate": 7.335668789808917e-05,
      "loss": 0.8907,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5366829633712769,
      "learning_rate": 7.329299363057325e-05,
      "loss": 0.9097,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5445237755775452,
      "learning_rate": 7.322929936305732e-05,
      "loss": 0.8628,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5116376876831055,
      "learning_rate": 7.31656050955414e-05,
      "loss": 0.86,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.579694390296936,
      "learning_rate": 7.310191082802547e-05,
      "loss": 0.9075,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5973986387252808,
      "learning_rate": 7.303821656050956e-05,
      "loss": 0.8966,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5258294343948364,
      "learning_rate": 7.297452229299363e-05,
      "loss": 0.8845,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4957127571105957,
      "learning_rate": 7.291082802547771e-05,
      "loss": 0.8919,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5788044929504395,
      "learning_rate": 7.284713375796178e-05,
      "loss": 0.867,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.6266525983810425,
      "learning_rate": 7.278343949044586e-05,
      "loss": 0.9157,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5797058939933777,
      "learning_rate": 7.271974522292993e-05,
      "loss": 0.8995,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5284995436668396,
      "learning_rate": 7.265605095541402e-05,
      "loss": 0.8798,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5411540269851685,
      "learning_rate": 7.259235668789809e-05,
      "loss": 0.86,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.7692059874534607,
      "learning_rate": 7.252866242038216e-05,
      "loss": 0.8797,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5444372892379761,
      "learning_rate": 7.246496815286624e-05,
      "loss": 0.9047,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.511956512928009,
      "learning_rate": 7.240127388535031e-05,
      "loss": 0.8824,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6188622713088989,
      "learning_rate": 7.23375796178344e-05,
      "loss": 0.87,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5659939050674438,
      "learning_rate": 7.227388535031847e-05,
      "loss": 0.8822,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5688533186912537,
      "learning_rate": 7.221019108280255e-05,
      "loss": 0.892,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.627356767654419,
      "learning_rate": 7.214649681528662e-05,
      "loss": 0.8911,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6712976098060608,
      "learning_rate": 7.20828025477707e-05,
      "loss": 0.8704,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5349586606025696,
      "learning_rate": 7.201910828025477e-05,
      "loss": 0.8951,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.608233630657196,
      "learning_rate": 7.195541401273886e-05,
      "loss": 0.8927,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5030810832977295,
      "learning_rate": 7.189171974522293e-05,
      "loss": 0.8674,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5219730734825134,
      "learning_rate": 7.182802547770701e-05,
      "loss": 0.893,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5737751722335815,
      "learning_rate": 7.176433121019108e-05,
      "loss": 0.9008,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5733808875083923,
      "learning_rate": 7.170063694267517e-05,
      "loss": 0.8723,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5448349118232727,
      "learning_rate": 7.163694267515924e-05,
      "loss": 0.9006,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.6214298009872437,
      "learning_rate": 7.15732484076433e-05,
      "loss": 0.8829,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5896222591400146,
      "learning_rate": 7.150955414012739e-05,
      "loss": 0.9047,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5450072884559631,
      "learning_rate": 7.144585987261146e-05,
      "loss": 0.8964,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5411213636398315,
      "learning_rate": 7.138216560509554e-05,
      "loss": 0.8717,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.4791530966758728,
      "learning_rate": 7.131847133757961e-05,
      "loss": 0.8823,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 4990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5541684627532959,
      "learning_rate": 7.12547770700637e-05,
      "loss": 0.8912,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6349590420722961,
      "learning_rate": 7.119108280254777e-05,
      "loss": 0.9002,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5731371641159058,
      "learning_rate": 7.112738853503185e-05,
      "loss": 0.8675,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5998538732528687,
      "learning_rate": 7.106369426751592e-05,
      "loss": 0.8972,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.605619490146637,
      "learning_rate": 7.1e-05,
      "loss": 0.8887,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5330061316490173,
      "learning_rate": 7.093630573248407e-05,
      "loss": 0.8896,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5728380084037781,
      "learning_rate": 7.087261146496816e-05,
      "loss": 0.8993,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.6582812070846558,
      "learning_rate": 7.080891719745223e-05,
      "loss": 0.8896,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5117426514625549,
      "learning_rate": 7.074522292993631e-05,
      "loss": 0.9062,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5359607338905334,
      "learning_rate": 7.068152866242038e-05,
      "loss": 0.8734,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5526896715164185,
      "learning_rate": 7.061783439490445e-05,
      "loss": 0.8754,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5385130643844604,
      "learning_rate": 7.055414012738854e-05,
      "loss": 0.8882,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5222037434577942,
      "learning_rate": 7.04904458598726e-05,
      "loss": 0.8818,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.6482069492340088,
      "learning_rate": 7.042675159235669e-05,
      "loss": 0.8702,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.504509687423706,
      "learning_rate": 7.036305732484076e-05,
      "loss": 0.8957,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5588439106941223,
      "learning_rate": 7.029936305732484e-05,
      "loss": 0.8873,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.610141396522522,
      "learning_rate": 7.023566878980891e-05,
      "loss": 0.897,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5777451992034912,
      "learning_rate": 7.0171974522293e-05,
      "loss": 0.9118,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.6056225895881653,
      "learning_rate": 7.010828025477707e-05,
      "loss": 0.9095,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5970057845115662,
      "learning_rate": 7.004458598726115e-05,
      "loss": 0.8801,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.6030189394950867,
      "learning_rate": 6.998089171974522e-05,
      "loss": 0.8945,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5925199389457703,
      "learning_rate": 6.99171974522293e-05,
      "loss": 0.8782,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5551109313964844,
      "learning_rate": 6.985350318471338e-05,
      "loss": 0.8821,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5022830963134766,
      "learning_rate": 6.978980891719745e-05,
      "loss": 0.8675,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5868439674377441,
      "learning_rate": 6.972611464968153e-05,
      "loss": 0.8728,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5532885193824768,
      "learning_rate": 6.96624203821656e-05,
      "loss": 0.8811,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5532572865486145,
      "learning_rate": 6.959872611464968e-05,
      "loss": 0.901,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.594550371170044,
      "learning_rate": 6.953503184713375e-05,
      "loss": 0.9012,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5424590110778809,
      "learning_rate": 6.947133757961784e-05,
      "loss": 0.8559,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5972351431846619,
      "learning_rate": 6.940764331210191e-05,
      "loss": 0.8713,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.6569409966468811,
      "learning_rate": 6.934394904458599e-05,
      "loss": 0.8878,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.597029447555542,
      "learning_rate": 6.928025477707006e-05,
      "loss": 0.8804,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5740005373954773,
      "learning_rate": 6.921656050955414e-05,
      "loss": 0.8933,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.548782467842102,
      "learning_rate": 6.915286624203821e-05,
      "loss": 0.8706,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5242761969566345,
      "learning_rate": 6.90891719745223e-05,
      "loss": 0.8872,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.6307296752929688,
      "learning_rate": 6.902547770700637e-05,
      "loss": 0.8865,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5386359095573425,
      "learning_rate": 6.896178343949045e-05,
      "loss": 0.8894,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5430437326431274,
      "learning_rate": 6.889808917197452e-05,
      "loss": 0.8823,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.49671536684036255,
      "learning_rate": 6.883439490445859e-05,
      "loss": 0.8665,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.6163560748100281,
      "learning_rate": 6.877070063694268e-05,
      "loss": 0.8845,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5099320411682129,
      "learning_rate": 6.870700636942675e-05,
      "loss": 0.8801,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5513092279434204,
      "learning_rate": 6.864331210191083e-05,
      "loss": 0.8994,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.49879205226898193,
      "learning_rate": 6.85796178343949e-05,
      "loss": 0.8943,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5449738502502441,
      "learning_rate": 6.851592356687898e-05,
      "loss": 0.8962,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.46554479002952576,
      "learning_rate": 6.845222929936305e-05,
      "loss": 0.888,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5366774201393127,
      "learning_rate": 6.838853503184714e-05,
      "loss": 0.8937,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.614962637424469,
      "learning_rate": 6.832484076433121e-05,
      "loss": 0.8691,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.4922376275062561,
      "learning_rate": 6.826114649681529e-05,
      "loss": 0.8842,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.6051047444343567,
      "learning_rate": 6.819745222929936e-05,
      "loss": 0.9041,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5461781024932861,
      "learning_rate": 6.813375796178345e-05,
      "loss": 0.9112,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5873090028762817,
      "learning_rate": 6.807006369426752e-05,
      "loss": 0.888,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5473777651786804,
      "learning_rate": 6.80063694267516e-05,
      "loss": 0.9067,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5596609115600586,
      "learning_rate": 6.794267515923567e-05,
      "loss": 0.8925,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.48069074749946594,
      "learning_rate": 6.787898089171974e-05,
      "loss": 0.8738,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.6374889016151428,
      "learning_rate": 6.781528662420382e-05,
      "loss": 0.8667,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5413127541542053,
      "learning_rate": 6.77515923566879e-05,
      "loss": 0.886,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4677344858646393,
      "learning_rate": 6.768789808917198e-05,
      "loss": 0.8752,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5758306980133057,
      "learning_rate": 6.762420382165605e-05,
      "loss": 0.9012,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4969116151332855,
      "learning_rate": 6.756050955414013e-05,
      "loss": 0.9112,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5981295108795166,
      "learning_rate": 6.74968152866242e-05,
      "loss": 0.9015,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5720081329345703,
      "learning_rate": 6.743312101910828e-05,
      "loss": 0.9096,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.569769024848938,
      "learning_rate": 6.736942675159235e-05,
      "loss": 0.8706,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5617003440856934,
      "learning_rate": 6.730573248407644e-05,
      "loss": 0.87,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5824927687644958,
      "learning_rate": 6.724203821656051e-05,
      "loss": 0.9072,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5539324283599854,
      "learning_rate": 6.717834394904459e-05,
      "loss": 0.8934,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5366997122764587,
      "learning_rate": 6.711464968152866e-05,
      "loss": 0.8784,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5396209359169006,
      "learning_rate": 6.705095541401275e-05,
      "loss": 0.9027,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5348963141441345,
      "learning_rate": 6.698726114649682e-05,
      "loss": 0.866,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5780596137046814,
      "learning_rate": 6.692356687898089e-05,
      "loss": 0.8894,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.6637961268424988,
      "learning_rate": 6.685987261146497e-05,
      "loss": 0.8851,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.864047110080719,
      "learning_rate": 6.679617834394904e-05,
      "loss": 0.8916,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5552239418029785,
      "learning_rate": 6.673248407643312e-05,
      "loss": 0.9218,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5573702454566956,
      "learning_rate": 6.66687898089172e-05,
      "loss": 0.8956,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5623142719268799,
      "learning_rate": 6.660509554140128e-05,
      "loss": 0.8905,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5609655380249023,
      "learning_rate": 6.654140127388535e-05,
      "loss": 0.8929,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5581201314926147,
      "learning_rate": 6.647770700636943e-05,
      "loss": 0.8699,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5425609946250916,
      "learning_rate": 6.64140127388535e-05,
      "loss": 0.8707,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5254411697387695,
      "learning_rate": 6.635031847133759e-05,
      "loss": 0.8608,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5528599619865417,
      "learning_rate": 6.628662420382166e-05,
      "loss": 0.8677,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5470523238182068,
      "learning_rate": 6.622292993630574e-05,
      "loss": 0.9024,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5529558062553406,
      "learning_rate": 6.615923566878981e-05,
      "loss": 0.888,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.6355565786361694,
      "learning_rate": 6.609554140127388e-05,
      "loss": 0.8635,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5301648378372192,
      "learning_rate": 6.603184713375796e-05,
      "loss": 0.9049,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5182161927223206,
      "learning_rate": 6.596815286624203e-05,
      "loss": 0.8601,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.6221626400947571,
      "learning_rate": 6.590445859872612e-05,
      "loss": 0.8746,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5320441722869873,
      "learning_rate": 6.584076433121019e-05,
      "loss": 0.892,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5502296686172485,
      "learning_rate": 6.577707006369427e-05,
      "loss": 0.8951,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5433371067047119,
      "learning_rate": 6.571337579617834e-05,
      "loss": 0.881,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.565592885017395,
      "learning_rate": 6.564968152866242e-05,
      "loss": 0.8857,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5991478562355042,
      "learning_rate": 6.55859872611465e-05,
      "loss": 0.8834,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5779725909233093,
      "learning_rate": 6.552229299363058e-05,
      "loss": 0.9006,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5620351433753967,
      "learning_rate": 6.545859872611465e-05,
      "loss": 0.8641,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5702232718467712,
      "learning_rate": 6.539490445859873e-05,
      "loss": 0.8617,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5302250385284424,
      "learning_rate": 6.53312101910828e-05,
      "loss": 0.8962,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5041064620018005,
      "learning_rate": 6.526751592356689e-05,
      "loss": 0.8696,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5761558413505554,
      "learning_rate": 6.520382165605096e-05,
      "loss": 0.862,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5102734565734863,
      "learning_rate": 6.514012738853503e-05,
      "loss": 0.8937,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.49469760060310364,
      "learning_rate": 6.507643312101911e-05,
      "loss": 0.8656,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5822766423225403,
      "learning_rate": 6.501273885350318e-05,
      "loss": 0.8728,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5382155776023865,
      "learning_rate": 6.494904458598726e-05,
      "loss": 0.884,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 5990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5767959952354431,
      "learning_rate": 6.488535031847133e-05,
      "loss": 0.8785,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5036946535110474,
      "learning_rate": 6.482165605095542e-05,
      "loss": 0.891,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5634387135505676,
      "learning_rate": 6.475796178343949e-05,
      "loss": 0.8866,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6301183700561523,
      "learning_rate": 6.469426751592357e-05,
      "loss": 0.8764,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6013332009315491,
      "learning_rate": 6.463057324840764e-05,
      "loss": 0.9015,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5528202056884766,
      "learning_rate": 6.456687898089173e-05,
      "loss": 0.8977,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.568086564540863,
      "learning_rate": 6.45031847133758e-05,
      "loss": 0.8669,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6520410776138306,
      "learning_rate": 6.443949044585988e-05,
      "loss": 0.8907,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5794298648834229,
      "learning_rate": 6.437579617834395e-05,
      "loss": 0.8712,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5090157985687256,
      "learning_rate": 6.431210191082803e-05,
      "loss": 0.878,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6202682852745056,
      "learning_rate": 6.42484076433121e-05,
      "loss": 0.8534,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5463171005249023,
      "learning_rate": 6.418471337579617e-05,
      "loss": 0.852,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5335412621498108,
      "learning_rate": 6.412101910828026e-05,
      "loss": 0.8868,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5212587118148804,
      "learning_rate": 6.405732484076433e-05,
      "loss": 0.9003,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5295570492744446,
      "learning_rate": 6.399363057324841e-05,
      "loss": 0.9077,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5488661527633667,
      "learning_rate": 6.392993630573248e-05,
      "loss": 0.9139,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.548369824886322,
      "learning_rate": 6.386624203821656e-05,
      "loss": 0.8591,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5016814470291138,
      "learning_rate": 6.380254777070063e-05,
      "loss": 0.8944,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5105929374694824,
      "learning_rate": 6.373885350318472e-05,
      "loss": 0.8983,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.521321177482605,
      "learning_rate": 6.367515923566879e-05,
      "loss": 0.8745,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5385405421257019,
      "learning_rate": 6.361146496815287e-05,
      "loss": 0.8836,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5338440537452698,
      "learning_rate": 6.354777070063694e-05,
      "loss": 0.9225,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5258084535598755,
      "learning_rate": 6.348407643312103e-05,
      "loss": 0.8843,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5688164830207825,
      "learning_rate": 6.34203821656051e-05,
      "loss": 0.8958,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5619838237762451,
      "learning_rate": 6.335668789808917e-05,
      "loss": 0.8578,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5111443400382996,
      "learning_rate": 6.329299363057325e-05,
      "loss": 0.8763,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5403845310211182,
      "learning_rate": 6.322929936305732e-05,
      "loss": 0.8788,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5215356945991516,
      "learning_rate": 6.31656050955414e-05,
      "loss": 0.8656,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5903708934783936,
      "learning_rate": 6.310191082802547e-05,
      "loss": 0.8882,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6022163033485413,
      "learning_rate": 6.303821656050956e-05,
      "loss": 0.8575,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5367187857627869,
      "learning_rate": 6.297452229299363e-05,
      "loss": 0.9121,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5830124020576477,
      "learning_rate": 6.291082802547771e-05,
      "loss": 0.8605,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5332449078559875,
      "learning_rate": 6.284713375796178e-05,
      "loss": 0.9164,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5586867928504944,
      "learning_rate": 6.278343949044587e-05,
      "loss": 0.8642,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5963526964187622,
      "learning_rate": 6.271974522292994e-05,
      "loss": 0.8693,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6327101588249207,
      "learning_rate": 6.265605095541402e-05,
      "loss": 0.8843,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5778951048851013,
      "learning_rate": 6.259235668789809e-05,
      "loss": 0.8837,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5711691379547119,
      "learning_rate": 6.252866242038217e-05,
      "loss": 0.9106,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5818186402320862,
      "learning_rate": 6.246496815286624e-05,
      "loss": 0.8736,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.6049737930297852,
      "learning_rate": 6.240127388535031e-05,
      "loss": 0.9108,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5459161996841431,
      "learning_rate": 6.23375796178344e-05,
      "loss": 0.8905,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.510424017906189,
      "learning_rate": 6.227388535031847e-05,
      "loss": 0.9056,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5309520363807678,
      "learning_rate": 6.221019108280255e-05,
      "loss": 0.8693,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5280246138572693,
      "learning_rate": 6.214649681528662e-05,
      "loss": 0.8727,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5652052164077759,
      "learning_rate": 6.20828025477707e-05,
      "loss": 0.8704,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5129522681236267,
      "learning_rate": 6.201910828025477e-05,
      "loss": 0.8761,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5963137149810791,
      "learning_rate": 6.195541401273886e-05,
      "loss": 0.8544,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5989205241203308,
      "learning_rate": 6.189171974522293e-05,
      "loss": 0.89,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5194454193115234,
      "learning_rate": 6.182802547770701e-05,
      "loss": 0.8731,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4959360957145691,
      "learning_rate": 6.176433121019108e-05,
      "loss": 0.8929,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.551125705242157,
      "learning_rate": 6.170063694267517e-05,
      "loss": 0.8616,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5451139211654663,
      "learning_rate": 6.163694267515924e-05,
      "loss": 0.8643,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5444925427436829,
      "learning_rate": 6.157324840764332e-05,
      "loss": 0.8831,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5570912957191467,
      "learning_rate": 6.150955414012739e-05,
      "loss": 0.8836,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5151851773262024,
      "learning_rate": 6.144585987261146e-05,
      "loss": 0.8639,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.49056312441825867,
      "learning_rate": 6.138216560509554e-05,
      "loss": 0.8785,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4749782979488373,
      "learning_rate": 6.131847133757961e-05,
      "loss": 0.8987,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5096744298934937,
      "learning_rate": 6.12547770700637e-05,
      "loss": 0.8636,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5247639417648315,
      "learning_rate": 6.119108280254777e-05,
      "loss": 0.8822,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5125365257263184,
      "learning_rate": 6.112738853503185e-05,
      "loss": 0.8612,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6495028138160706,
      "learning_rate": 6.106369426751592e-05,
      "loss": 0.8804,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6130849123001099,
      "learning_rate": 6.1e-05,
      "loss": 0.8603,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.570441484451294,
      "learning_rate": 6.0936305732484076e-05,
      "loss": 0.8976,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5690892934799194,
      "learning_rate": 6.087261146496815e-05,
      "loss": 0.843,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6043858528137207,
      "learning_rate": 6.080891719745223e-05,
      "loss": 0.8872,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.521172821521759,
      "learning_rate": 6.0745222929936306e-05,
      "loss": 0.8623,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6279696822166443,
      "learning_rate": 6.068152866242038e-05,
      "loss": 0.8774,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5914869904518127,
      "learning_rate": 6.061783439490446e-05,
      "loss": 0.8738,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5309963822364807,
      "learning_rate": 6.055414012738854e-05,
      "loss": 0.8961,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.49851086735725403,
      "learning_rate": 6.0490445859872614e-05,
      "loss": 0.8677,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5877674221992493,
      "learning_rate": 6.042675159235669e-05,
      "loss": 0.8684,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.48666176199913025,
      "learning_rate": 6.036305732484077e-05,
      "loss": 0.8774,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6143956184387207,
      "learning_rate": 6.0299363057324845e-05,
      "loss": 0.8747,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5709465742111206,
      "learning_rate": 6.023566878980892e-05,
      "loss": 0.8844,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.56462162733078,
      "learning_rate": 6.0171974522293e-05,
      "loss": 0.8559,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5849558711051941,
      "learning_rate": 6.010828025477707e-05,
      "loss": 0.8786,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5224260091781616,
      "learning_rate": 6.0044585987261146e-05,
      "loss": 0.8532,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5706215500831604,
      "learning_rate": 5.998089171974522e-05,
      "loss": 0.8549,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5041067004203796,
      "learning_rate": 5.99171974522293e-05,
      "loss": 0.8703,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.6147697567939758,
      "learning_rate": 5.9853503184713376e-05,
      "loss": 0.8786,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.6441134214401245,
      "learning_rate": 5.978980891719745e-05,
      "loss": 0.8992,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.6264499425888062,
      "learning_rate": 5.972611464968153e-05,
      "loss": 0.8845,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5924432277679443,
      "learning_rate": 5.966242038216561e-05,
      "loss": 0.8828,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.6159273982048035,
      "learning_rate": 5.9598726114649684e-05,
      "loss": 0.8686,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5710685849189758,
      "learning_rate": 5.953503184713376e-05,
      "loss": 0.894,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5524258613586426,
      "learning_rate": 5.947133757961784e-05,
      "loss": 0.8899,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5787354111671448,
      "learning_rate": 5.9407643312101915e-05,
      "loss": 0.8546,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5477069020271301,
      "learning_rate": 5.934394904458599e-05,
      "loss": 0.8787,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.6147220730781555,
      "learning_rate": 5.928025477707007e-05,
      "loss": 0.8879,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5261998176574707,
      "learning_rate": 5.921656050955414e-05,
      "loss": 0.9033,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5160650610923767,
      "learning_rate": 5.9152866242038216e-05,
      "loss": 0.8578,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.6226661205291748,
      "learning_rate": 5.908917197452229e-05,
      "loss": 0.8895,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4986201226711273,
      "learning_rate": 5.902547770700637e-05,
      "loss": 0.8696,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5445514917373657,
      "learning_rate": 5.8961783439490446e-05,
      "loss": 0.8936,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5744483470916748,
      "learning_rate": 5.889808917197452e-05,
      "loss": 0.8541,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5101067423820496,
      "learning_rate": 5.88343949044586e-05,
      "loss": 0.8972,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.54083651304245,
      "learning_rate": 5.877070063694268e-05,
      "loss": 0.8575,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5288860201835632,
      "learning_rate": 5.8707006369426754e-05,
      "loss": 0.8859,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5304664373397827,
      "learning_rate": 5.864331210191083e-05,
      "loss": 0.9006,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5717320442199707,
      "learning_rate": 5.857961783439491e-05,
      "loss": 0.8684,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 6990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.521751880645752,
      "learning_rate": 5.8515923566878985e-05,
      "loss": 0.8773,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5460343360900879,
      "learning_rate": 5.845222929936306e-05,
      "loss": 0.8676,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5551382303237915,
      "learning_rate": 5.838853503184714e-05,
      "loss": 0.8459,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5307554602622986,
      "learning_rate": 5.8324840764331216e-05,
      "loss": 0.8972,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5830961465835571,
      "learning_rate": 5.8261146496815286e-05,
      "loss": 0.8601,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5745844841003418,
      "learning_rate": 5.819745222929936e-05,
      "loss": 0.8742,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5107027888298035,
      "learning_rate": 5.813375796178344e-05,
      "loss": 0.878,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.7025989890098572,
      "learning_rate": 5.8070063694267516e-05,
      "loss": 0.8642,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.6303359866142273,
      "learning_rate": 5.800636942675159e-05,
      "loss": 0.8602,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5137262344360352,
      "learning_rate": 5.794267515923567e-05,
      "loss": 0.8707,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.554009735584259,
      "learning_rate": 5.787898089171975e-05,
      "loss": 0.8572,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5526527166366577,
      "learning_rate": 5.7815286624203824e-05,
      "loss": 0.8688,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.9742128252983093,
      "learning_rate": 5.77515923566879e-05,
      "loss": 0.8824,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5232354998588562,
      "learning_rate": 5.768789808917198e-05,
      "loss": 0.8901,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5452825427055359,
      "learning_rate": 5.7624203821656055e-05,
      "loss": 0.872,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5305787324905396,
      "learning_rate": 5.756050955414013e-05,
      "loss": 0.9046,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5473122000694275,
      "learning_rate": 5.749681528662421e-05,
      "loss": 0.8804,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.6258863210678101,
      "learning_rate": 5.7433121019108286e-05,
      "loss": 0.8933,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5017375946044922,
      "learning_rate": 5.7369426751592356e-05,
      "loss": 0.8921,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5204535126686096,
      "learning_rate": 5.730573248407643e-05,
      "loss": 0.8694,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5876609086990356,
      "learning_rate": 5.724203821656051e-05,
      "loss": 0.8849,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5704731345176697,
      "learning_rate": 5.7178343949044586e-05,
      "loss": 0.8844,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.49567875266075134,
      "learning_rate": 5.711464968152866e-05,
      "loss": 0.872,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5774997472763062,
      "learning_rate": 5.705095541401274e-05,
      "loss": 0.8693,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5102982521057129,
      "learning_rate": 5.698726114649682e-05,
      "loss": 0.8591,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5977953672409058,
      "learning_rate": 5.6923566878980894e-05,
      "loss": 0.8624,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.664761483669281,
      "learning_rate": 5.685987261146497e-05,
      "loss": 0.8862,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5394580960273743,
      "learning_rate": 5.679617834394905e-05,
      "loss": 0.8796,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5266260504722595,
      "learning_rate": 5.6732484076433125e-05,
      "loss": 0.8768,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.571178138256073,
      "learning_rate": 5.66687898089172e-05,
      "loss": 0.8949,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5399868488311768,
      "learning_rate": 5.660509554140128e-05,
      "loss": 0.8656,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6034677028656006,
      "learning_rate": 5.6541401273885356e-05,
      "loss": 0.8687,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.537738025188446,
      "learning_rate": 5.647770700636943e-05,
      "loss": 0.8712,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5704653263092041,
      "learning_rate": 5.64140127388535e-05,
      "loss": 0.8606,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5335469245910645,
      "learning_rate": 5.635031847133758e-05,
      "loss": 0.8853,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5269140005111694,
      "learning_rate": 5.6286624203821656e-05,
      "loss": 0.8588,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.602112889289856,
      "learning_rate": 5.622292993630573e-05,
      "loss": 0.8652,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5989587903022766,
      "learning_rate": 5.615923566878981e-05,
      "loss": 0.8425,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.6427887082099915,
      "learning_rate": 5.609554140127389e-05,
      "loss": 0.8702,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5564777255058289,
      "learning_rate": 5.6031847133757964e-05,
      "loss": 0.8633,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5418772101402283,
      "learning_rate": 5.596815286624204e-05,
      "loss": 0.8957,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5809254050254822,
      "learning_rate": 5.590445859872612e-05,
      "loss": 0.8711,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5313619375228882,
      "learning_rate": 5.5840764331210195e-05,
      "loss": 0.8698,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5611067414283752,
      "learning_rate": 5.577707006369427e-05,
      "loss": 0.8674,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5353450775146484,
      "learning_rate": 5.571337579617835e-05,
      "loss": 0.8638,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5202521681785583,
      "learning_rate": 5.5649681528662426e-05,
      "loss": 0.8724,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5684260725975037,
      "learning_rate": 5.55859872611465e-05,
      "loss": 0.8813,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5583212375640869,
      "learning_rate": 5.552229299363057e-05,
      "loss": 0.8937,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.569209098815918,
      "learning_rate": 5.545859872611465e-05,
      "loss": 0.8703,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5366013050079346,
      "learning_rate": 5.5394904458598726e-05,
      "loss": 0.8725,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5401418805122375,
      "learning_rate": 5.53312101910828e-05,
      "loss": 0.8772,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5705187320709229,
      "learning_rate": 5.526751592356688e-05,
      "loss": 0.8893,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5963625907897949,
      "learning_rate": 5.520382165605096e-05,
      "loss": 0.8797,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5229465961456299,
      "learning_rate": 5.5140127388535034e-05,
      "loss": 0.8889,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5574226975440979,
      "learning_rate": 5.507643312101911e-05,
      "loss": 0.8738,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.6087568998336792,
      "learning_rate": 5.501273885350319e-05,
      "loss": 0.8904,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5513929128646851,
      "learning_rate": 5.4949044585987265e-05,
      "loss": 0.9023,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5194180011749268,
      "learning_rate": 5.488535031847134e-05,
      "loss": 0.8888,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5172723531723022,
      "learning_rate": 5.482165605095542e-05,
      "loss": 0.8731,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5943834185600281,
      "learning_rate": 5.4757961783439496e-05,
      "loss": 0.8772,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.536495566368103,
      "learning_rate": 5.469426751592357e-05,
      "loss": 0.8665,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5841646194458008,
      "learning_rate": 5.463057324840764e-05,
      "loss": 0.8712,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5669993162155151,
      "learning_rate": 5.456687898089172e-05,
      "loss": 0.8597,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.6329278349876404,
      "learning_rate": 5.4503184713375796e-05,
      "loss": 0.8751,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5109385251998901,
      "learning_rate": 5.443949044585987e-05,
      "loss": 0.8942,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5742037892341614,
      "learning_rate": 5.437579617834395e-05,
      "loss": 0.9001,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.661705493927002,
      "learning_rate": 5.431210191082803e-05,
      "loss": 0.8458,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5366435050964355,
      "learning_rate": 5.4248407643312104e-05,
      "loss": 0.86,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6003596186637878,
      "learning_rate": 5.418471337579618e-05,
      "loss": 0.8696,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.533245325088501,
      "learning_rate": 5.412101910828026e-05,
      "loss": 0.8522,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5306530594825745,
      "learning_rate": 5.4057324840764335e-05,
      "loss": 0.8847,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6527922749519348,
      "learning_rate": 5.399363057324841e-05,
      "loss": 0.8789,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.595000147819519,
      "learning_rate": 5.392993630573249e-05,
      "loss": 0.8355,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5394546985626221,
      "learning_rate": 5.3866242038216566e-05,
      "loss": 0.8548,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5733621716499329,
      "learning_rate": 5.380254777070064e-05,
      "loss": 0.8653,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5508468747138977,
      "learning_rate": 5.373885350318472e-05,
      "loss": 0.8945,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.610894501209259,
      "learning_rate": 5.367515923566879e-05,
      "loss": 0.8976,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5553873777389526,
      "learning_rate": 5.3611464968152866e-05,
      "loss": 0.867,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5115483999252319,
      "learning_rate": 5.354777070063694e-05,
      "loss": 0.8523,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.513729989528656,
      "learning_rate": 5.348407643312102e-05,
      "loss": 0.8988,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5983762741088867,
      "learning_rate": 5.34203821656051e-05,
      "loss": 0.8499,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5823088884353638,
      "learning_rate": 5.3356687898089174e-05,
      "loss": 0.866,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.573600709438324,
      "learning_rate": 5.329299363057325e-05,
      "loss": 0.8701,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5541461706161499,
      "learning_rate": 5.322929936305733e-05,
      "loss": 0.8724,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5558524131774902,
      "learning_rate": 5.3165605095541405e-05,
      "loss": 0.8657,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5285508036613464,
      "learning_rate": 5.310191082802548e-05,
      "loss": 0.8425,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5635260939598083,
      "learning_rate": 5.303821656050956e-05,
      "loss": 0.89,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5565703511238098,
      "learning_rate": 5.2974522292993636e-05,
      "loss": 0.8612,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5692863464355469,
      "learning_rate": 5.291082802547771e-05,
      "loss": 0.8719,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5936442613601685,
      "learning_rate": 5.284713375796179e-05,
      "loss": 0.8786,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5127562284469604,
      "learning_rate": 5.278343949044586e-05,
      "loss": 0.8467,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5491020083427429,
      "learning_rate": 5.2719745222929936e-05,
      "loss": 0.8602,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5560590624809265,
      "learning_rate": 5.2656050955414013e-05,
      "loss": 0.8702,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.6465674638748169,
      "learning_rate": 5.259235668789809e-05,
      "loss": 0.8869,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.6076213717460632,
      "learning_rate": 5.252866242038217e-05,
      "loss": 0.8791,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5880044102668762,
      "learning_rate": 5.2464968152866244e-05,
      "loss": 0.8703,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5579984784126282,
      "learning_rate": 5.240127388535032e-05,
      "loss": 0.8722,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5295410752296448,
      "learning_rate": 5.23375796178344e-05,
      "loss": 0.8454,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5681312680244446,
      "learning_rate": 5.2273885350318475e-05,
      "loss": 0.8592,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.6963818669319153,
      "learning_rate": 5.221019108280255e-05,
      "loss": 0.8483,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 7990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.6498535871505737,
      "learning_rate": 5.214649681528663e-05,
      "loss": 0.8913,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5272806286811829,
      "learning_rate": 5.2082802547770706e-05,
      "loss": 0.8522,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.524757444858551,
      "learning_rate": 5.201910828025478e-05,
      "loss": 0.9146,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.8040810227394104,
      "learning_rate": 5.195541401273886e-05,
      "loss": 0.8646,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.637153148651123,
      "learning_rate": 5.1891719745222936e-05,
      "loss": 0.8595,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.6900390386581421,
      "learning_rate": 5.1828025477707006e-05,
      "loss": 0.8787,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5754966139793396,
      "learning_rate": 5.1764331210191083e-05,
      "loss": 0.8605,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5973451137542725,
      "learning_rate": 5.170063694267516e-05,
      "loss": 0.8691,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.6348568797111511,
      "learning_rate": 5.163694267515924e-05,
      "loss": 0.8745,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.540838360786438,
      "learning_rate": 5.1573248407643314e-05,
      "loss": 0.8585,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.6029226183891296,
      "learning_rate": 5.150955414012739e-05,
      "loss": 0.857,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5446135997772217,
      "learning_rate": 5.144585987261147e-05,
      "loss": 0.8467,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5528589487075806,
      "learning_rate": 5.1382165605095545e-05,
      "loss": 0.8552,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5417741537094116,
      "learning_rate": 5.131847133757962e-05,
      "loss": 0.8544,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.5020217895507812,
      "learning_rate": 5.12547770700637e-05,
      "loss": 0.8615,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.5379871726036072,
      "learning_rate": 5.1191082802547776e-05,
      "loss": 0.853,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.6144492030143738,
      "learning_rate": 5.112738853503185e-05,
      "loss": 0.8443,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.566996693611145,
      "learning_rate": 5.106369426751593e-05,
      "loss": 0.8595,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.5716440677642822,
      "learning_rate": 5.1000000000000006e-05,
      "loss": 0.8758,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.54944908618927,
      "learning_rate": 5.0936305732484077e-05,
      "loss": 0.8385,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.5852215886116028,
      "learning_rate": 5.0872611464968153e-05,
      "loss": 0.874,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.6428478360176086,
      "learning_rate": 5.080891719745223e-05,
      "loss": 0.8527,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.5332466959953308,
      "learning_rate": 5.074522292993631e-05,
      "loss": 0.8697,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.6330180168151855,
      "learning_rate": 5.0681528662420384e-05,
      "loss": 0.8447,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.6093716025352478,
      "learning_rate": 5.061783439490446e-05,
      "loss": 0.8576,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.5464934706687927,
      "learning_rate": 5.055414012738854e-05,
      "loss": 0.8563,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.5934207439422607,
      "learning_rate": 5.0490445859872615e-05,
      "loss": 0.8379,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.5612914562225342,
      "learning_rate": 5.042675159235669e-05,
      "loss": 0.8502,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.545253574848175,
      "learning_rate": 5.036305732484077e-05,
      "loss": 0.8509,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.631450891494751,
      "learning_rate": 5.0299363057324846e-05,
      "loss": 0.8606,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.5951800346374512,
      "learning_rate": 5.023566878980892e-05,
      "loss": 0.8716,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.6067357063293457,
      "learning_rate": 5.0171974522293e-05,
      "loss": 0.8929,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.5289549827575684,
      "learning_rate": 5.0108280254777076e-05,
      "loss": 0.865,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.5379547476768494,
      "learning_rate": 5.004458598726115e-05,
      "loss": 0.8649,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.5237485766410828,
      "learning_rate": 4.9980891719745223e-05,
      "loss": 0.8835,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.5215060710906982,
      "learning_rate": 4.99171974522293e-05,
      "loss": 0.8549,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.7047269940376282,
      "learning_rate": 4.985350318471338e-05,
      "loss": 0.8595,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.5435574054718018,
      "learning_rate": 4.9789808917197454e-05,
      "loss": 0.8762,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.5348605513572693,
      "learning_rate": 4.972611464968153e-05,
      "loss": 0.8524,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.6164429187774658,
      "learning_rate": 4.966242038216561e-05,
      "loss": 0.8243,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.5418780446052551,
      "learning_rate": 4.9598726114649685e-05,
      "loss": 0.8464,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.5288802981376648,
      "learning_rate": 4.953503184713376e-05,
      "loss": 0.834,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.5178465843200684,
      "learning_rate": 4.947133757961784e-05,
      "loss": 0.869,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.5777456760406494,
      "learning_rate": 4.9407643312101916e-05,
      "loss": 0.8598,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.6265929937362671,
      "learning_rate": 4.934394904458599e-05,
      "loss": 0.8382,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.5676342248916626,
      "learning_rate": 4.928025477707007e-05,
      "loss": 0.8545,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.6321595311164856,
      "learning_rate": 4.9216560509554146e-05,
      "loss": 0.8541,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.5599589943885803,
      "learning_rate": 4.915286624203822e-05,
      "loss": 0.8798,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.5524100661277771,
      "learning_rate": 4.9089171974522293e-05,
      "loss": 0.8318,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.5421915650367737,
      "learning_rate": 4.902547770700637e-05,
      "loss": 0.8652,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.7382321357727051,
      "learning_rate": 4.896178343949045e-05,
      "loss": 0.8511,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.5415564179420471,
      "learning_rate": 4.8898089171974524e-05,
      "loss": 0.8542,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.5908517241477966,
      "learning_rate": 4.88343949044586e-05,
      "loss": 0.8717,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.5226035714149475,
      "learning_rate": 4.877070063694268e-05,
      "loss": 0.8306,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5351910591125488,
      "learning_rate": 4.8707006369426755e-05,
      "loss": 0.841,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5828281044960022,
      "learning_rate": 4.864331210191083e-05,
      "loss": 0.8322,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5739461779594421,
      "learning_rate": 4.857961783439491e-05,
      "loss": 0.8474,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5811489820480347,
      "learning_rate": 4.8515923566878986e-05,
      "loss": 0.8686,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5632668137550354,
      "learning_rate": 4.845222929936306e-05,
      "loss": 0.8386,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.6513923406600952,
      "learning_rate": 4.838853503184714e-05,
      "loss": 0.8776,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5276843309402466,
      "learning_rate": 4.8324840764331216e-05,
      "loss": 0.847,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.5792731642723083,
      "learning_rate": 4.826114649681529e-05,
      "loss": 0.8615,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5763053894042969,
      "learning_rate": 4.819745222929937e-05,
      "loss": 0.8468,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5236169099807739,
      "learning_rate": 4.813375796178344e-05,
      "loss": 0.828,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5243390798568726,
      "learning_rate": 4.807006369426752e-05,
      "loss": 0.8745,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5692167282104492,
      "learning_rate": 4.8006369426751594e-05,
      "loss": 0.8349,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.7056101560592651,
      "learning_rate": 4.794267515923567e-05,
      "loss": 0.8303,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5863968133926392,
      "learning_rate": 4.787898089171975e-05,
      "loss": 0.8409,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5797470808029175,
      "learning_rate": 4.7815286624203825e-05,
      "loss": 0.8504,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.6207801699638367,
      "learning_rate": 4.77515923566879e-05,
      "loss": 0.8596,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.5625149011611938,
      "learning_rate": 4.768789808917198e-05,
      "loss": 0.836,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.5690678358078003,
      "learning_rate": 4.7624203821656056e-05,
      "loss": 0.8604,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.518943727016449,
      "learning_rate": 4.756050955414013e-05,
      "loss": 0.8846,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.5701858401298523,
      "learning_rate": 4.749681528662421e-05,
      "loss": 0.8752,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.5501466989517212,
      "learning_rate": 4.7433121019108286e-05,
      "loss": 0.8431,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.692132830619812,
      "learning_rate": 4.736942675159236e-05,
      "loss": 0.8645,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.5666806697845459,
      "learning_rate": 4.730573248407644e-05,
      "loss": 0.8563,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.5708613991737366,
      "learning_rate": 4.724203821656051e-05,
      "loss": 0.8322,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.5461627244949341,
      "learning_rate": 4.717834394904459e-05,
      "loss": 0.8367,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.506532609462738,
      "learning_rate": 4.7114649681528664e-05,
      "loss": 0.8346,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.5624059438705444,
      "learning_rate": 4.705095541401274e-05,
      "loss": 0.8164,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.5525660514831543,
      "learning_rate": 4.698726114649682e-05,
      "loss": 0.8566,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.5464274883270264,
      "learning_rate": 4.6923566878980895e-05,
      "loss": 0.8471,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.5622477531433105,
      "learning_rate": 4.685987261146497e-05,
      "loss": 0.8667,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.608081042766571,
      "learning_rate": 4.679617834394905e-05,
      "loss": 0.8832,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.608669638633728,
      "learning_rate": 4.6732484076433126e-05,
      "loss": 0.8612,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.6040840148925781,
      "learning_rate": 4.66687898089172e-05,
      "loss": 0.8529,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.6103033423423767,
      "learning_rate": 4.660509554140128e-05,
      "loss": 0.8572,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5756964683532715,
      "learning_rate": 4.6541401273885356e-05,
      "loss": 0.8712,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5669918656349182,
      "learning_rate": 4.647770700636943e-05,
      "loss": 0.8262,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.6642317175865173,
      "learning_rate": 4.641401273885351e-05,
      "loss": 0.8717,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5726878643035889,
      "learning_rate": 4.635031847133758e-05,
      "loss": 0.8703,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5667121410369873,
      "learning_rate": 4.628662420382166e-05,
      "loss": 0.8469,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5571203231811523,
      "learning_rate": 4.6222929936305734e-05,
      "loss": 0.8476,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5069302916526794,
      "learning_rate": 4.615923566878981e-05,
      "loss": 0.8566,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.517417311668396,
      "learning_rate": 4.609554140127389e-05,
      "loss": 0.8732,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.5386157631874084,
      "learning_rate": 4.6031847133757965e-05,
      "loss": 0.8324,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.6298375725746155,
      "learning_rate": 4.596815286624204e-05,
      "loss": 0.7908,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.5578718185424805,
      "learning_rate": 4.590445859872612e-05,
      "loss": 0.8724,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.5550695657730103,
      "learning_rate": 4.5840764331210196e-05,
      "loss": 0.8522,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 8990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.5744197964668274,
      "learning_rate": 4.577707006369427e-05,
      "loss": 0.8519,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.5899156332015991,
      "learning_rate": 4.571337579617835e-05,
      "loss": 0.8519,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.6079444289207458,
      "learning_rate": 4.5649681528662426e-05,
      "loss": 0.8505,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.5901761054992676,
      "learning_rate": 4.55859872611465e-05,
      "loss": 0.8574,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.5695390701293945,
      "learning_rate": 4.552229299363058e-05,
      "loss": 0.85,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.5685015320777893,
      "learning_rate": 4.545859872611466e-05,
      "loss": 0.8223,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.6257502436637878,
      "learning_rate": 4.539490445859873e-05,
      "loss": 0.8676,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.5832564234733582,
      "learning_rate": 4.5331210191082804e-05,
      "loss": 0.8564,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.6030582785606384,
      "learning_rate": 4.526751592356688e-05,
      "loss": 0.8492,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.6871100068092346,
      "learning_rate": 4.520382165605096e-05,
      "loss": 0.8662,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.53432297706604,
      "learning_rate": 4.5140127388535035e-05,
      "loss": 0.8542,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.5585580468177795,
      "learning_rate": 4.507643312101911e-05,
      "loss": 0.7867,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.5414518117904663,
      "learning_rate": 4.501273885350319e-05,
      "loss": 0.8376,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.6098955273628235,
      "learning_rate": 4.4949044585987266e-05,
      "loss": 0.8725,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.5669853091239929,
      "learning_rate": 4.488535031847134e-05,
      "loss": 0.8332,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.5340268611907959,
      "learning_rate": 4.482165605095542e-05,
      "loss": 0.8553,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.5587529540061951,
      "learning_rate": 4.4757961783439496e-05,
      "loss": 0.8343,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.6109263896942139,
      "learning_rate": 4.469426751592357e-05,
      "loss": 0.8252,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.6285218596458435,
      "learning_rate": 4.463057324840765e-05,
      "loss": 0.8346,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.5180321931838989,
      "learning_rate": 4.456687898089173e-05,
      "loss": 0.8289,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.6208270788192749,
      "learning_rate": 4.45031847133758e-05,
      "loss": 0.8531,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.6124671697616577,
      "learning_rate": 4.4439490445859874e-05,
      "loss": 0.8379,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.5522865653038025,
      "learning_rate": 4.437579617834395e-05,
      "loss": 0.8407,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.6094708442687988,
      "learning_rate": 4.431210191082803e-05,
      "loss": 0.8587,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.5936129689216614,
      "learning_rate": 4.4248407643312105e-05,
      "loss": 0.8488,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.515078604221344,
      "learning_rate": 4.418471337579618e-05,
      "loss": 0.8551,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.622717022895813,
      "learning_rate": 4.412101910828026e-05,
      "loss": 0.8645,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5956246256828308,
      "learning_rate": 4.4057324840764336e-05,
      "loss": 0.8643,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5683963894844055,
      "learning_rate": 4.399363057324841e-05,
      "loss": 0.8389,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5116027593612671,
      "learning_rate": 4.392993630573249e-05,
      "loss": 0.8504,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5620011687278748,
      "learning_rate": 4.3866242038216566e-05,
      "loss": 0.8764,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5658126473426819,
      "learning_rate": 4.380254777070064e-05,
      "loss": 0.8733,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.6153251528739929,
      "learning_rate": 4.373885350318472e-05,
      "loss": 0.8496,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.6150373220443726,
      "learning_rate": 4.36751592356688e-05,
      "loss": 0.8285,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5434516072273254,
      "learning_rate": 4.3611464968152874e-05,
      "loss": 0.863,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.6302858591079712,
      "learning_rate": 4.3547770700636944e-05,
      "loss": 0.8682,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.61929851770401,
      "learning_rate": 4.348407643312102e-05,
      "loss": 0.886,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5373562574386597,
      "learning_rate": 4.34203821656051e-05,
      "loss": 0.8298,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5400745868682861,
      "learning_rate": 4.3356687898089175e-05,
      "loss": 0.838,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5738040804862976,
      "learning_rate": 4.329299363057325e-05,
      "loss": 0.8631,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5795983672142029,
      "learning_rate": 4.322929936305733e-05,
      "loss": 0.8565,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.55351722240448,
      "learning_rate": 4.3165605095541406e-05,
      "loss": 0.8902,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5806913375854492,
      "learning_rate": 4.310191082802548e-05,
      "loss": 0.881,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.6037645936012268,
      "learning_rate": 4.303821656050956e-05,
      "loss": 0.8624,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.5471616387367249,
      "learning_rate": 4.2974522292993636e-05,
      "loss": 0.8388,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.5801625847816467,
      "learning_rate": 4.2910828025477713e-05,
      "loss": 0.8382,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.544641375541687,
      "learning_rate": 4.284713375796179e-05,
      "loss": 0.8442,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.5814592838287354,
      "learning_rate": 4.278343949044587e-05,
      "loss": 0.8339,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.5295001268386841,
      "learning_rate": 4.2719745222929944e-05,
      "loss": 0.8476,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.6271728277206421,
      "learning_rate": 4.2656050955414014e-05,
      "loss": 0.8626,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.6040265560150146,
      "learning_rate": 4.259235668789809e-05,
      "loss": 0.8506,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5334275960922241,
      "learning_rate": 4.252866242038217e-05,
      "loss": 0.8227,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.647208034992218,
      "learning_rate": 4.246496815286624e-05,
      "loss": 0.8627,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5936146378517151,
      "learning_rate": 4.2401273885350315e-05,
      "loss": 0.888,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5416791439056396,
      "learning_rate": 4.233757961783439e-05,
      "loss": 0.8823,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5779086351394653,
      "learning_rate": 4.227388535031847e-05,
      "loss": 0.8465,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5653147101402283,
      "learning_rate": 4.2210191082802546e-05,
      "loss": 0.8227,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5841782093048096,
      "learning_rate": 4.214649681528662e-05,
      "loss": 0.8703,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5924214720726013,
      "learning_rate": 4.20828025477707e-05,
      "loss": 0.8601,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.5881546139717102,
      "learning_rate": 4.201910828025478e-05,
      "loss": 0.8651,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5617165565490723,
      "learning_rate": 4.1955414012738854e-05,
      "loss": 0.8447,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5984334945678711,
      "learning_rate": 4.189171974522293e-05,
      "loss": 0.8694,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5753591656684875,
      "learning_rate": 4.182802547770701e-05,
      "loss": 0.8598,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5675076246261597,
      "learning_rate": 4.1764331210191084e-05,
      "loss": 0.8652,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5429391860961914,
      "learning_rate": 4.170063694267516e-05,
      "loss": 0.8722,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.6187253594398499,
      "learning_rate": 4.163694267515924e-05,
      "loss": 0.8256,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5988272428512573,
      "learning_rate": 4.157324840764331e-05,
      "loss": 0.8266,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.5375120043754578,
      "learning_rate": 4.1509554140127385e-05,
      "loss": 0.8702,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.584695041179657,
      "learning_rate": 4.144585987261146e-05,
      "loss": 0.8497,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5349233150482178,
      "learning_rate": 4.138216560509554e-05,
      "loss": 0.8539,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.6189013719558716,
      "learning_rate": 4.1318471337579616e-05,
      "loss": 0.8756,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5804276466369629,
      "learning_rate": 4.125477707006369e-05,
      "loss": 0.8553,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.6997498869895935,
      "learning_rate": 4.119108280254777e-05,
      "loss": 0.87,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5317782759666443,
      "learning_rate": 4.112738853503185e-05,
      "loss": 0.8476,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5482212901115417,
      "learning_rate": 4.1063694267515924e-05,
      "loss": 0.8163,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5931942462921143,
      "learning_rate": 4.1e-05,
      "loss": 0.8261,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.5837103724479675,
      "learning_rate": 4.093630573248408e-05,
      "loss": 0.8701,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.5434841513633728,
      "learning_rate": 4.0872611464968154e-05,
      "loss": 0.8242,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.629984974861145,
      "learning_rate": 4.080891719745223e-05,
      "loss": 0.8705,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.6990328431129456,
      "learning_rate": 4.074522292993631e-05,
      "loss": 0.8586,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.5907762050628662,
      "learning_rate": 4.0681528662420385e-05,
      "loss": 0.8655,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.5725825428962708,
      "learning_rate": 4.0617834394904455e-05,
      "loss": 0.8504,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.5525303483009338,
      "learning_rate": 4.055414012738853e-05,
      "loss": 0.8422,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.601482629776001,
      "learning_rate": 4.049044585987261e-05,
      "loss": 0.8686,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.5497685670852661,
      "learning_rate": 4.0426751592356686e-05,
      "loss": 0.8397,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.5859342813491821,
      "learning_rate": 4.036305732484076e-05,
      "loss": 0.8906,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.5428025722503662,
      "learning_rate": 4.029936305732484e-05,
      "loss": 0.8413,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.6050729751586914,
      "learning_rate": 4.023566878980892e-05,
      "loss": 0.8755,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.5653709173202515,
      "learning_rate": 4.0171974522292994e-05,
      "loss": 0.8666,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.5361793041229248,
      "learning_rate": 4.010828025477707e-05,
      "loss": 0.8218,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.6140942573547363,
      "learning_rate": 4.004458598726115e-05,
      "loss": 0.8487,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.5644840002059937,
      "learning_rate": 3.9980891719745224e-05,
      "loss": 0.857,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.5795630812644958,
      "learning_rate": 3.99171974522293e-05,
      "loss": 0.8377,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.5633305311203003,
      "learning_rate": 3.985350318471338e-05,
      "loss": 0.8679,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.60102778673172,
      "learning_rate": 3.9789808917197455e-05,
      "loss": 0.8748,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.6072971224784851,
      "learning_rate": 3.9726114649681525e-05,
      "loss": 0.8537,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.6720110774040222,
      "learning_rate": 3.96624203821656e-05,
      "loss": 0.8485,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.6156752705574036,
      "learning_rate": 3.959872611464968e-05,
      "loss": 0.8478,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.5758206844329834,
      "learning_rate": 3.9535031847133756e-05,
      "loss": 0.8449,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.5701586008071899,
      "learning_rate": 3.947133757961783e-05,
      "loss": 0.8346,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 9990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.52237868309021,
      "learning_rate": 3.940764331210191e-05,
      "loss": 0.8253,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.5953871607780457,
      "learning_rate": 3.934394904458599e-05,
      "loss": 0.8402,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.6187193989753723,
      "learning_rate": 3.9280254777070064e-05,
      "loss": 0.84,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.5820570588111877,
      "learning_rate": 3.921656050955414e-05,
      "loss": 0.8452,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.5865253806114197,
      "learning_rate": 3.915286624203822e-05,
      "loss": 0.8622,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.594710648059845,
      "learning_rate": 3.9089171974522294e-05,
      "loss": 0.8575,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.5631176233291626,
      "learning_rate": 3.902547770700637e-05,
      "loss": 0.8374,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.5939119458198547,
      "learning_rate": 3.896178343949045e-05,
      "loss": 0.8763,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.573266327381134,
      "learning_rate": 3.8898089171974525e-05,
      "loss": 0.8529,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.49202847480773926,
      "learning_rate": 3.88343949044586e-05,
      "loss": 0.8263,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5387163162231445,
      "learning_rate": 3.877070063694267e-05,
      "loss": 0.8625,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5848548412322998,
      "learning_rate": 3.870700636942675e-05,
      "loss": 0.8307,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5548748970031738,
      "learning_rate": 3.8643312101910826e-05,
      "loss": 0.8285,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5727143287658691,
      "learning_rate": 3.85796178343949e-05,
      "loss": 0.8509,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5815598964691162,
      "learning_rate": 3.851592356687898e-05,
      "loss": 0.8836,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5907566547393799,
      "learning_rate": 3.845222929936306e-05,
      "loss": 0.8398,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.6278606653213501,
      "learning_rate": 3.8388535031847134e-05,
      "loss": 0.8435,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.5442222952842712,
      "learning_rate": 3.832484076433121e-05,
      "loss": 0.8505,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.5859748721122742,
      "learning_rate": 3.826114649681529e-05,
      "loss": 0.8566,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.6682499647140503,
      "learning_rate": 3.8197452229299364e-05,
      "loss": 0.8712,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.5633233785629272,
      "learning_rate": 3.813375796178344e-05,
      "loss": 0.8656,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.6651773452758789,
      "learning_rate": 3.807006369426752e-05,
      "loss": 0.8707,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.5781067609786987,
      "learning_rate": 3.8006369426751595e-05,
      "loss": 0.8598,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.6810473799705505,
      "learning_rate": 3.794267515923567e-05,
      "loss": 0.8373,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.6265770792961121,
      "learning_rate": 3.787898089171974e-05,
      "loss": 0.8729,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.5550416707992554,
      "learning_rate": 3.781528662420382e-05,
      "loss": 0.861,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.5630444884300232,
      "learning_rate": 3.7751592356687896e-05,
      "loss": 0.8614,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.5981015563011169,
      "learning_rate": 3.768789808917197e-05,
      "loss": 0.8417,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.5597262978553772,
      "learning_rate": 3.762420382165605e-05,
      "loss": 0.8552,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.5736472606658936,
      "learning_rate": 3.756050955414013e-05,
      "loss": 0.8444,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.5704269409179688,
      "learning_rate": 3.7496815286624204e-05,
      "loss": 0.8333,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.549435555934906,
      "learning_rate": 3.743312101910828e-05,
      "loss": 0.849,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5624418258666992,
      "learning_rate": 3.736942675159236e-05,
      "loss": 0.84,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5569450259208679,
      "learning_rate": 3.7305732484076434e-05,
      "loss": 0.82,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5782328844070435,
      "learning_rate": 3.724203821656051e-05,
      "loss": 0.8365,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.6573158502578735,
      "learning_rate": 3.717834394904459e-05,
      "loss": 0.835,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5627766847610474,
      "learning_rate": 3.7114649681528665e-05,
      "loss": 0.8338,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5935114026069641,
      "learning_rate": 3.705095541401274e-05,
      "loss": 0.8528,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.6130989193916321,
      "learning_rate": 3.698726114649681e-05,
      "loss": 0.8499,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5816967487335205,
      "learning_rate": 3.692356687898089e-05,
      "loss": 0.8541,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.5552312135696411,
      "learning_rate": 3.6859872611464966e-05,
      "loss": 0.8494,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.5926358103752136,
      "learning_rate": 3.679617834394904e-05,
      "loss": 0.8701,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.6585802435874939,
      "learning_rate": 3.673248407643312e-05,
      "loss": 0.8697,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.5160651206970215,
      "learning_rate": 3.66687898089172e-05,
      "loss": 0.8637,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.8365841507911682,
      "learning_rate": 3.6605095541401274e-05,
      "loss": 0.8572,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.6559211611747742,
      "learning_rate": 3.654140127388535e-05,
      "loss": 0.8221,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.603316068649292,
      "learning_rate": 3.647770700636943e-05,
      "loss": 0.8237,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.5332236289978027,
      "learning_rate": 3.6414012738853504e-05,
      "loss": 0.8604,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.6046949028968811,
      "learning_rate": 3.635031847133758e-05,
      "loss": 0.8606,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.6080541014671326,
      "learning_rate": 3.628662420382166e-05,
      "loss": 0.8231,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.5540225505828857,
      "learning_rate": 3.6222929936305735e-05,
      "loss": 0.8622,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.6116626858711243,
      "learning_rate": 3.615923566878981e-05,
      "loss": 0.8452,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.5268499255180359,
      "learning_rate": 3.609554140127389e-05,
      "loss": 0.8643,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.593361496925354,
      "learning_rate": 3.603184713375796e-05,
      "loss": 0.8679,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.6413514614105225,
      "learning_rate": 3.5968152866242036e-05,
      "loss": 0.8436,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.6277458667755127,
      "learning_rate": 3.590445859872611e-05,
      "loss": 0.8149,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.5620959401130676,
      "learning_rate": 3.584076433121019e-05,
      "loss": 0.8427,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.7740691900253296,
      "learning_rate": 3.577707006369427e-05,
      "loss": 0.8194,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.5390883088111877,
      "learning_rate": 3.5713375796178344e-05,
      "loss": 0.8462,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.6066960096359253,
      "learning_rate": 3.564968152866242e-05,
      "loss": 0.8438,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.5721805691719055,
      "learning_rate": 3.55859872611465e-05,
      "loss": 0.8852,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.5543267130851746,
      "learning_rate": 3.5522292993630574e-05,
      "loss": 0.8418,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.551352858543396,
      "learning_rate": 3.545859872611465e-05,
      "loss": 0.8575,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.5895906090736389,
      "learning_rate": 3.539490445859873e-05,
      "loss": 0.8322,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.6208714842796326,
      "learning_rate": 3.5331210191082805e-05,
      "loss": 0.8592,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.6484043002128601,
      "learning_rate": 3.526751592356688e-05,
      "loss": 0.8437,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5603862404823303,
      "learning_rate": 3.520382165605096e-05,
      "loss": 0.8414,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.6064278483390808,
      "learning_rate": 3.514012738853503e-05,
      "loss": 0.851,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5677962303161621,
      "learning_rate": 3.5076433121019106e-05,
      "loss": 0.8463,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5643174648284912,
      "learning_rate": 3.501273885350318e-05,
      "loss": 0.8446,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5769677758216858,
      "learning_rate": 3.494904458598726e-05,
      "loss": 0.8244,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.6470271348953247,
      "learning_rate": 3.488535031847134e-05,
      "loss": 0.8484,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5943115949630737,
      "learning_rate": 3.4821656050955414e-05,
      "loss": 0.8719,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.5173988342285156,
      "learning_rate": 3.475796178343949e-05,
      "loss": 0.8202,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.5133519172668457,
      "learning_rate": 3.469426751592357e-05,
      "loss": 0.8607,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.6479659676551819,
      "learning_rate": 3.4630573248407644e-05,
      "loss": 0.8357,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.6060785055160522,
      "learning_rate": 3.456687898089172e-05,
      "loss": 0.8443,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.5761645436286926,
      "learning_rate": 3.45031847133758e-05,
      "loss": 0.8721,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.5355122685432434,
      "learning_rate": 3.4439490445859875e-05,
      "loss": 0.8414,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.6006566286087036,
      "learning_rate": 3.437579617834395e-05,
      "loss": 0.8738,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.5849606394767761,
      "learning_rate": 3.431210191082803e-05,
      "loss": 0.8332,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.5841776132583618,
      "learning_rate": 3.4248407643312106e-05,
      "loss": 0.852,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.5938459634780884,
      "learning_rate": 3.4184713375796176e-05,
      "loss": 0.8796,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.5698296427726746,
      "learning_rate": 3.412101910828025e-05,
      "loss": 0.8333,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.580596923828125,
      "learning_rate": 3.405732484076433e-05,
      "loss": 0.8534,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.5333631038665771,
      "learning_rate": 3.399363057324841e-05,
      "loss": 0.8348,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.6496769189834595,
      "learning_rate": 3.3929936305732484e-05,
      "loss": 0.8932,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.7200677394866943,
      "learning_rate": 3.386624203821656e-05,
      "loss": 0.8346,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.5436977744102478,
      "learning_rate": 3.380254777070064e-05,
      "loss": 0.8464,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5795146226882935,
      "learning_rate": 3.3738853503184714e-05,
      "loss": 0.8072,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5177848935127258,
      "learning_rate": 3.367515923566879e-05,
      "loss": 0.8545,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.6202844381332397,
      "learning_rate": 3.361146496815287e-05,
      "loss": 0.851,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5789759159088135,
      "learning_rate": 3.3547770700636945e-05,
      "loss": 0.8693,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5165707468986511,
      "learning_rate": 3.348407643312102e-05,
      "loss": 0.8541,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5957597494125366,
      "learning_rate": 3.34203821656051e-05,
      "loss": 0.8739,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5551866888999939,
      "learning_rate": 3.3356687898089176e-05,
      "loss": 0.8696,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.6256672739982605,
      "learning_rate": 3.3292993630573246e-05,
      "loss": 0.8506,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.5564112663269043,
      "learning_rate": 3.322929936305732e-05,
      "loss": 0.8337,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.5981550216674805,
      "learning_rate": 3.31656050955414e-05,
      "loss": 0.8578,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.6769008040428162,
      "learning_rate": 3.310191082802548e-05,
      "loss": 0.8429,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 10990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.5731686353683472,
      "learning_rate": 3.3038216560509554e-05,
      "loss": 0.8609,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.6229621767997742,
      "learning_rate": 3.297452229299363e-05,
      "loss": 0.8435,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.6300867199897766,
      "learning_rate": 3.291082802547771e-05,
      "loss": 0.8797,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.6082947850227356,
      "learning_rate": 3.2847133757961784e-05,
      "loss": 0.8697,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.5586636662483215,
      "learning_rate": 3.278343949044586e-05,
      "loss": 0.8422,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.6251539587974548,
      "learning_rate": 3.271974522292994e-05,
      "loss": 0.8124,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.6733857989311218,
      "learning_rate": 3.2656050955414015e-05,
      "loss": 0.8382,
      "max_memory_allocated (GB)": 94.49,
      "memory_allocated (GB)": 14.97,
      "step": 11060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.601640522480011,
      "learning_rate": 3.259235668789809e-05,
      "loss": 0.8371,
      "max_memory_allocated (GB)": 94.55,
      "memory_allocated (GB)": 14.97,
      "step": 11070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.5989189147949219,
      "learning_rate": 3.252866242038217e-05,
      "loss": 0.8841,
      "max_memory_allocated (GB)": 94.55,
      "memory_allocated (GB)": 14.97,
      "step": 11080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.5607017278671265,
      "learning_rate": 3.2464968152866246e-05,
      "loss": 0.8503,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.5420870780944824,
      "learning_rate": 3.240127388535032e-05,
      "loss": 0.8607,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.6338402032852173,
      "learning_rate": 3.233757961783439e-05,
      "loss": 0.8442,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.5808389782905579,
      "learning_rate": 3.227388535031847e-05,
      "loss": 0.8361,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.5183432698249817,
      "learning_rate": 3.221019108280255e-05,
      "loss": 0.8474,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.6294568777084351,
      "learning_rate": 3.2146496815286624e-05,
      "loss": 0.8316,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.5947890877723694,
      "learning_rate": 3.20828025477707e-05,
      "loss": 0.821,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.6249579787254333,
      "learning_rate": 3.201910828025478e-05,
      "loss": 0.8496,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.5387454032897949,
      "learning_rate": 3.1955414012738854e-05,
      "loss": 0.8555,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.5490169525146484,
      "learning_rate": 3.189171974522293e-05,
      "loss": 0.8417,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.6309676170349121,
      "learning_rate": 3.182802547770701e-05,
      "loss": 0.83,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.6202732920646667,
      "learning_rate": 3.1764331210191085e-05,
      "loss": 0.8344,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.6259014010429382,
      "learning_rate": 3.170063694267516e-05,
      "loss": 0.8042,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.5590731501579285,
      "learning_rate": 3.163694267515924e-05,
      "loss": 0.831,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.6487732529640198,
      "learning_rate": 3.1573248407643316e-05,
      "loss": 0.8368,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.6921071410179138,
      "learning_rate": 3.150955414012739e-05,
      "loss": 0.855,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.592799186706543,
      "learning_rate": 3.144585987261146e-05,
      "loss": 0.8247,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.6593350172042847,
      "learning_rate": 3.138216560509554e-05,
      "loss": 0.8399,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.5151501297950745,
      "learning_rate": 3.131847133757962e-05,
      "loss": 0.8577,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.5430037975311279,
      "learning_rate": 3.1254777070063694e-05,
      "loss": 0.8286,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.5675473809242249,
      "learning_rate": 3.119108280254777e-05,
      "loss": 0.8413,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.6474435925483704,
      "learning_rate": 3.112738853503185e-05,
      "loss": 0.8435,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.6244317293167114,
      "learning_rate": 3.1063694267515924e-05,
      "loss": 0.815,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.5554672479629517,
      "learning_rate": 3.1e-05,
      "loss": 0.849,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.5643844604492188,
      "learning_rate": 3.093630573248408e-05,
      "loss": 0.849,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.5386666655540466,
      "learning_rate": 3.0872611464968155e-05,
      "loss": 0.8601,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.6414317488670349,
      "learning_rate": 3.080891719745223e-05,
      "loss": 0.8351,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.5491330623626709,
      "learning_rate": 3.074522292993631e-05,
      "loss": 0.8409,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.6073495745658875,
      "learning_rate": 3.0681528662420386e-05,
      "loss": 0.8514,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.5687406659126282,
      "learning_rate": 3.061783439490446e-05,
      "loss": 0.847,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.6221336126327515,
      "learning_rate": 3.055414012738854e-05,
      "loss": 0.86,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.6338231563568115,
      "learning_rate": 3.0490445859872613e-05,
      "loss": 0.8505,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.6311108469963074,
      "learning_rate": 3.042675159235669e-05,
      "loss": 0.8503,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.5485032796859741,
      "learning_rate": 3.0363057324840767e-05,
      "loss": 0.8311,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.8035585880279541,
      "learning_rate": 3.029936305732484e-05,
      "loss": 0.8514,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.5874172449111938,
      "learning_rate": 3.0235668789808918e-05,
      "loss": 0.8374,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.5448861122131348,
      "learning_rate": 3.0171974522292994e-05,
      "loss": 0.8591,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.6101541519165039,
      "learning_rate": 3.010828025477707e-05,
      "loss": 0.808,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.644763708114624,
      "learning_rate": 3.0044585987261148e-05,
      "loss": 0.8493,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.6366714239120483,
      "learning_rate": 2.9980891719745225e-05,
      "loss": 0.8473,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.6037055850028992,
      "learning_rate": 2.9917197452229302e-05,
      "loss": 0.8512,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.5885847210884094,
      "learning_rate": 2.9853503184713376e-05,
      "loss": 0.7974,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.5361979007720947,
      "learning_rate": 2.9789808917197453e-05,
      "loss": 0.8347,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.5696050524711609,
      "learning_rate": 2.972611464968153e-05,
      "loss": 0.8567,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.5758958458900452,
      "learning_rate": 2.9662420382165606e-05,
      "loss": 0.8504,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.613055944442749,
      "learning_rate": 2.9598726114649683e-05,
      "loss": 0.8399,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.6125010251998901,
      "learning_rate": 2.953503184713376e-05,
      "loss": 0.843,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.622986376285553,
      "learning_rate": 2.9471337579617837e-05,
      "loss": 0.822,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.5726191401481628,
      "learning_rate": 2.9407643312101914e-05,
      "loss": 0.8118,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.6309647560119629,
      "learning_rate": 2.9343949044585988e-05,
      "loss": 0.8407,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.5828757286071777,
      "learning_rate": 2.9280254777070064e-05,
      "loss": 0.8288,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.6801642179489136,
      "learning_rate": 2.921656050955414e-05,
      "loss": 0.8244,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.6346856355667114,
      "learning_rate": 2.9152866242038218e-05,
      "loss": 0.8625,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.6004509329795837,
      "learning_rate": 2.9089171974522295e-05,
      "loss": 0.8064,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.5791013836860657,
      "learning_rate": 2.9025477707006372e-05,
      "loss": 0.8469,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.6295746564865112,
      "learning_rate": 2.896178343949045e-05,
      "loss": 0.8381,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.5519148707389832,
      "learning_rate": 2.8898089171974523e-05,
      "loss": 0.8291,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.6016730666160583,
      "learning_rate": 2.88343949044586e-05,
      "loss": 0.8671,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.584075391292572,
      "learning_rate": 2.8770700636942676e-05,
      "loss": 0.8405,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.5806698799133301,
      "learning_rate": 2.8707006369426753e-05,
      "loss": 0.8348,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.6517837047576904,
      "learning_rate": 2.864331210191083e-05,
      "loss": 0.8385,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.5573362112045288,
      "learning_rate": 2.8579617834394907e-05,
      "loss": 0.8797,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.6296159029006958,
      "learning_rate": 2.8515923566878984e-05,
      "loss": 0.8179,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.5505685210227966,
      "learning_rate": 2.8452229299363058e-05,
      "loss": 0.8069,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.5616992115974426,
      "learning_rate": 2.8388535031847134e-05,
      "loss": 0.8455,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.5606623291969299,
      "learning_rate": 2.832484076433121e-05,
      "loss": 0.8513,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.5369701981544495,
      "learning_rate": 2.8261146496815288e-05,
      "loss": 0.8493,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.6416178345680237,
      "learning_rate": 2.8197452229299365e-05,
      "loss": 0.8528,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.517137885093689,
      "learning_rate": 2.8133757961783442e-05,
      "loss": 0.8564,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.5944896936416626,
      "learning_rate": 2.807006369426752e-05,
      "loss": 0.8143,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.5841711759567261,
      "learning_rate": 2.8006369426751593e-05,
      "loss": 0.852,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.5721305012702942,
      "learning_rate": 2.794267515923567e-05,
      "loss": 0.8448,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.6192442774772644,
      "learning_rate": 2.7878980891719746e-05,
      "loss": 0.834,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.669075071811676,
      "learning_rate": 2.7815286624203823e-05,
      "loss": 0.8484,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.5460517406463623,
      "learning_rate": 2.77515923566879e-05,
      "loss": 0.8693,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.5798333883285522,
      "learning_rate": 2.7687898089171977e-05,
      "loss": 0.8223,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.7171079516410828,
      "learning_rate": 2.7624203821656054e-05,
      "loss": 0.8239,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5694504976272583,
      "learning_rate": 2.7560509554140128e-05,
      "loss": 0.8415,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5613299608230591,
      "learning_rate": 2.7496815286624204e-05,
      "loss": 0.8432,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5206958651542664,
      "learning_rate": 2.743312101910828e-05,
      "loss": 0.8241,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5295774340629578,
      "learning_rate": 2.736942675159236e-05,
      "loss": 0.8602,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.64802086353302,
      "learning_rate": 2.7305732484076435e-05,
      "loss": 0.8171,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5394118428230286,
      "learning_rate": 2.7242038216560512e-05,
      "loss": 0.8296,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5935512781143188,
      "learning_rate": 2.717834394904459e-05,
      "loss": 0.8535,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.6226528882980347,
      "learning_rate": 2.7114649681528666e-05,
      "loss": 0.8272,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.6066229343414307,
      "learning_rate": 2.705095541401274e-05,
      "loss": 0.8279,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.5835513472557068,
      "learning_rate": 2.6987261146496816e-05,
      "loss": 0.8512,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.5614724159240723,
      "learning_rate": 2.6923566878980893e-05,
      "loss": 0.8342,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.5533426403999329,
      "learning_rate": 2.685987261146497e-05,
      "loss": 0.8695,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.552813708782196,
      "learning_rate": 2.6796178343949047e-05,
      "loss": 0.8521,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.609652042388916,
      "learning_rate": 2.6732484076433124e-05,
      "loss": 0.8395,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 11990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.6274034976959229,
      "learning_rate": 2.66687898089172e-05,
      "loss": 0.7963,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.614339292049408,
      "learning_rate": 2.6605095541401274e-05,
      "loss": 0.848,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.5958476066589355,
      "learning_rate": 2.654140127388535e-05,
      "loss": 0.86,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.6635801792144775,
      "learning_rate": 2.647770700636943e-05,
      "loss": 0.828,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.6173161268234253,
      "learning_rate": 2.6414012738853505e-05,
      "loss": 0.8657,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.621300995349884,
      "learning_rate": 2.6350318471337582e-05,
      "loss": 0.8653,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.6438680291175842,
      "learning_rate": 2.628662420382166e-05,
      "loss": 0.8388,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.5824630856513977,
      "learning_rate": 2.6222929936305736e-05,
      "loss": 0.8658,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.5131130814552307,
      "learning_rate": 2.615923566878981e-05,
      "loss": 0.8505,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.6061905026435852,
      "learning_rate": 2.6095541401273886e-05,
      "loss": 0.8219,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6551744341850281,
      "learning_rate": 2.6031847133757963e-05,
      "loss": 0.8619,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6331368088722229,
      "learning_rate": 2.596815286624204e-05,
      "loss": 0.8692,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.644982099533081,
      "learning_rate": 2.5904458598726117e-05,
      "loss": 0.8454,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.48831579089164734,
      "learning_rate": 2.5840764331210194e-05,
      "loss": 0.8423,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.5983461737632751,
      "learning_rate": 2.577707006369427e-05,
      "loss": 0.8453,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6347223520278931,
      "learning_rate": 2.5713375796178345e-05,
      "loss": 0.8363,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6169897317886353,
      "learning_rate": 2.564968152866242e-05,
      "loss": 0.8327,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6166051030158997,
      "learning_rate": 2.55859872611465e-05,
      "loss": 0.858,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.5569881200790405,
      "learning_rate": 2.5522292993630575e-05,
      "loss": 0.8209,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.5771543979644775,
      "learning_rate": 2.5458598726114652e-05,
      "loss": 0.8205,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.5933178067207336,
      "learning_rate": 2.539490445859873e-05,
      "loss": 0.8283,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.5819362998008728,
      "learning_rate": 2.5331210191082806e-05,
      "loss": 0.8343,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.6056768298149109,
      "learning_rate": 2.5267515923566883e-05,
      "loss": 0.8516,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.5443101525306702,
      "learning_rate": 2.5203821656050956e-05,
      "loss": 0.8524,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.6425015330314636,
      "learning_rate": 2.5140127388535033e-05,
      "loss": 0.8496,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.6068335175514221,
      "learning_rate": 2.507643312101911e-05,
      "loss": 0.8592,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.5697997212409973,
      "learning_rate": 2.5012738853503187e-05,
      "loss": 0.8089,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.6474873423576355,
      "learning_rate": 2.4949044585987264e-05,
      "loss": 0.8211,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.5467383861541748,
      "learning_rate": 2.488535031847134e-05,
      "loss": 0.813,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.7291449904441833,
      "learning_rate": 2.4821656050955418e-05,
      "loss": 0.8483,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.5775637626647949,
      "learning_rate": 2.475796178343949e-05,
      "loss": 0.8565,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.5808078050613403,
      "learning_rate": 2.469426751592357e-05,
      "loss": 0.8448,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.5515039563179016,
      "learning_rate": 2.4630573248407645e-05,
      "loss": 0.8158,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.6430584788322449,
      "learning_rate": 2.4566878980891722e-05,
      "loss": 0.8353,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.5688273310661316,
      "learning_rate": 2.45031847133758e-05,
      "loss": 0.8614,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.5493985414505005,
      "learning_rate": 2.4439490445859876e-05,
      "loss": 0.7994,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.5262490510940552,
      "learning_rate": 2.4375796178343953e-05,
      "loss": 0.8378,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.5832075476646423,
      "learning_rate": 2.4312101910828026e-05,
      "loss": 0.8549,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.7101218104362488,
      "learning_rate": 2.4248407643312103e-05,
      "loss": 0.8381,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.5871150493621826,
      "learning_rate": 2.418471337579618e-05,
      "loss": 0.8461,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.6132311224937439,
      "learning_rate": 2.4121019108280257e-05,
      "loss": 0.8333,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.6356542706489563,
      "learning_rate": 2.4057324840764334e-05,
      "loss": 0.855,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.5895008444786072,
      "learning_rate": 2.399363057324841e-05,
      "loss": 0.8649,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.5542967319488525,
      "learning_rate": 2.3929936305732488e-05,
      "loss": 0.8778,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.6304234266281128,
      "learning_rate": 2.386624203821656e-05,
      "loss": 0.8381,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.561419665813446,
      "learning_rate": 2.380254777070064e-05,
      "loss": 0.8467,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.660615086555481,
      "learning_rate": 2.3738853503184712e-05,
      "loss": 0.8296,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.5472921133041382,
      "learning_rate": 2.367515923566879e-05,
      "loss": 0.8705,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.5949731469154358,
      "learning_rate": 2.3611464968152866e-05,
      "loss": 0.849,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.5476264953613281,
      "learning_rate": 2.3547770700636943e-05,
      "loss": 0.8479,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.5857336521148682,
      "learning_rate": 2.348407643312102e-05,
      "loss": 0.8332,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.5827224850654602,
      "learning_rate": 2.3420382165605096e-05,
      "loss": 0.8449,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.6607509851455688,
      "learning_rate": 2.3356687898089173e-05,
      "loss": 0.8388,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.6671987771987915,
      "learning_rate": 2.3292993630573247e-05,
      "loss": 0.8314,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.6405848264694214,
      "learning_rate": 2.3229299363057324e-05,
      "loss": 0.8386,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.6223413348197937,
      "learning_rate": 2.31656050955414e-05,
      "loss": 0.8247,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.5649201273918152,
      "learning_rate": 2.3101910828025478e-05,
      "loss": 0.8382,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.5927321910858154,
      "learning_rate": 2.3038216560509555e-05,
      "loss": 0.843,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.6821181774139404,
      "learning_rate": 2.297452229299363e-05,
      "loss": 0.8801,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.6321542859077454,
      "learning_rate": 2.291082802547771e-05,
      "loss": 0.837,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.5311897993087769,
      "learning_rate": 2.2847133757961782e-05,
      "loss": 0.8429,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.5849913358688354,
      "learning_rate": 2.278343949044586e-05,
      "loss": 0.8539,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.570947527885437,
      "learning_rate": 2.2719745222929936e-05,
      "loss": 0.8271,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.6576603651046753,
      "learning_rate": 2.2656050955414013e-05,
      "loss": 0.8354,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.5636218190193176,
      "learning_rate": 2.259235668789809e-05,
      "loss": 0.8522,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.5577971339225769,
      "learning_rate": 2.2528662420382166e-05,
      "loss": 0.819,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.6458442211151123,
      "learning_rate": 2.2464968152866243e-05,
      "loss": 0.8263,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5527836084365845,
      "learning_rate": 2.2401273885350317e-05,
      "loss": 0.8409,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5587246417999268,
      "learning_rate": 2.2337579617834394e-05,
      "loss": 0.8321,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5923686623573303,
      "learning_rate": 2.227388535031847e-05,
      "loss": 0.8603,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.6047967076301575,
      "learning_rate": 2.2210191082802548e-05,
      "loss": 0.8504,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5479164123535156,
      "learning_rate": 2.2146496815286625e-05,
      "loss": 0.8074,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5567662119865417,
      "learning_rate": 2.20828025477707e-05,
      "loss": 0.844,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5728209614753723,
      "learning_rate": 2.201910828025478e-05,
      "loss": 0.8236,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.592235267162323,
      "learning_rate": 2.1955414012738852e-05,
      "loss": 0.8385,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.6159526109695435,
      "learning_rate": 2.189171974522293e-05,
      "loss": 0.8438,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.7097925543785095,
      "learning_rate": 2.1828025477707006e-05,
      "loss": 0.8409,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.6174066662788391,
      "learning_rate": 2.1764331210191083e-05,
      "loss": 0.8257,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.6272559762001038,
      "learning_rate": 2.170063694267516e-05,
      "loss": 0.855,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.6165140271186829,
      "learning_rate": 2.1636942675159236e-05,
      "loss": 0.8345,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.5537904500961304,
      "learning_rate": 2.1573248407643313e-05,
      "loss": 0.8403,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.6132100820541382,
      "learning_rate": 2.150955414012739e-05,
      "loss": 0.8394,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.6333381533622742,
      "learning_rate": 2.1445859872611464e-05,
      "loss": 0.8373,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.5513577461242676,
      "learning_rate": 2.138216560509554e-05,
      "loss": 0.8583,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.552699625492096,
      "learning_rate": 2.1318471337579618e-05,
      "loss": 0.8555,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.636202871799469,
      "learning_rate": 2.1254777070063695e-05,
      "loss": 0.825,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.6300379037857056,
      "learning_rate": 2.119108280254777e-05,
      "loss": 0.8203,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.6774061918258667,
      "learning_rate": 2.112738853503185e-05,
      "loss": 0.8373,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.6430328488349915,
      "learning_rate": 2.1063694267515925e-05,
      "loss": 0.8227,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.6144772171974182,
      "learning_rate": 2.1e-05,
      "loss": 0.8402,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.5309426188468933,
      "learning_rate": 2.0936305732484076e-05,
      "loss": 0.8686,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5501454472541809,
      "learning_rate": 2.0872611464968153e-05,
      "loss": 0.8418,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.6468961238861084,
      "learning_rate": 2.080891719745223e-05,
      "loss": 0.8638,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.6229689717292786,
      "learning_rate": 2.0745222929936306e-05,
      "loss": 0.8216,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5711995363235474,
      "learning_rate": 2.0681528662420383e-05,
      "loss": 0.8599,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.6419293880462646,
      "learning_rate": 2.061783439490446e-05,
      "loss": 0.8342,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5840780138969421,
      "learning_rate": 2.0554140127388534e-05,
      "loss": 0.8778,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5459557771682739,
      "learning_rate": 2.049044585987261e-05,
      "loss": 0.8165,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5893337726593018,
      "learning_rate": 2.0426751592356688e-05,
      "loss": 0.8327,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5610139966011047,
      "learning_rate": 2.0363057324840765e-05,
      "loss": 0.8046,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 12990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.6202096939086914,
      "learning_rate": 2.029936305732484e-05,
      "loss": 0.8562,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.6327053308486938,
      "learning_rate": 2.023566878980892e-05,
      "loss": 0.8477,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.5899960398674011,
      "learning_rate": 2.0171974522292995e-05,
      "loss": 0.829,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.5683597922325134,
      "learning_rate": 2.010828025477707e-05,
      "loss": 0.8641,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.5415876507759094,
      "learning_rate": 2.0044585987261146e-05,
      "loss": 0.8223,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.6744927763938904,
      "learning_rate": 1.9980891719745223e-05,
      "loss": 0.8265,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.6498674154281616,
      "learning_rate": 1.99171974522293e-05,
      "loss": 0.8558,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.561626136302948,
      "learning_rate": 1.9853503184713377e-05,
      "loss": 0.8205,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.590421199798584,
      "learning_rate": 1.9789808917197453e-05,
      "loss": 0.8283,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.5505582094192505,
      "learning_rate": 1.972611464968153e-05,
      "loss": 0.8293,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.6818318963050842,
      "learning_rate": 1.9662420382165607e-05,
      "loss": 0.8231,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.6751272082328796,
      "learning_rate": 1.959872611464968e-05,
      "loss": 0.8447,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.6087584495544434,
      "learning_rate": 1.9535031847133758e-05,
      "loss": 0.8527,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.5452800989151001,
      "learning_rate": 1.9471337579617835e-05,
      "loss": 0.8528,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.5750834345817566,
      "learning_rate": 1.940764331210191e-05,
      "loss": 0.8695,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.5984904170036316,
      "learning_rate": 1.934394904458599e-05,
      "loss": 0.8419,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.5514894723892212,
      "learning_rate": 1.9280254777070065e-05,
      "loss": 0.8175,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.5440700650215149,
      "learning_rate": 1.9216560509554142e-05,
      "loss": 0.8151,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.6085617542266846,
      "learning_rate": 1.9152866242038216e-05,
      "loss": 0.8153,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.5366870164871216,
      "learning_rate": 1.9089171974522293e-05,
      "loss": 0.839,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.5688049793243408,
      "learning_rate": 1.902547770700637e-05,
      "loss": 0.8257,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.6022163033485413,
      "learning_rate": 1.8961783439490447e-05,
      "loss": 0.8301,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.6411393880844116,
      "learning_rate": 1.8898089171974523e-05,
      "loss": 0.8222,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.5853878855705261,
      "learning_rate": 1.88343949044586e-05,
      "loss": 0.8393,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.7118549942970276,
      "learning_rate": 1.8770700636942677e-05,
      "loss": 0.8529,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.6585959196090698,
      "learning_rate": 1.870700636942675e-05,
      "loss": 0.8589,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.6305626034736633,
      "learning_rate": 1.8643312101910828e-05,
      "loss": 0.8297,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.5811955332756042,
      "learning_rate": 1.8579617834394905e-05,
      "loss": 0.8325,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.640131413936615,
      "learning_rate": 1.851592356687898e-05,
      "loss": 0.8724,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.556799054145813,
      "learning_rate": 1.845222929936306e-05,
      "loss": 0.853,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.5864406228065491,
      "learning_rate": 1.8388535031847135e-05,
      "loss": 0.8407,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.5744174122810364,
      "learning_rate": 1.8324840764331212e-05,
      "loss": 0.8251,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.6507772207260132,
      "learning_rate": 1.8261146496815286e-05,
      "loss": 0.8154,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.8050383925437927,
      "learning_rate": 1.8197452229299363e-05,
      "loss": 0.847,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.5747182369232178,
      "learning_rate": 1.813375796178344e-05,
      "loss": 0.8322,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.5775649547576904,
      "learning_rate": 1.8070063694267517e-05,
      "loss": 0.8501,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.60750812292099,
      "learning_rate": 1.8006369426751593e-05,
      "loss": 0.8755,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.5230236649513245,
      "learning_rate": 1.794267515923567e-05,
      "loss": 0.8329,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.5921688079833984,
      "learning_rate": 1.7878980891719747e-05,
      "loss": 0.8172,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.49644508957862854,
      "learning_rate": 1.781528662420382e-05,
      "loss": 0.8517,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.6654114723205566,
      "learning_rate": 1.7751592356687898e-05,
      "loss": 0.8535,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.6532041430473328,
      "learning_rate": 1.7687898089171975e-05,
      "loss": 0.8325,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.5903627872467041,
      "learning_rate": 1.762420382165605e-05,
      "loss": 0.8121,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.6084394454956055,
      "learning_rate": 1.756050955414013e-05,
      "loss": 0.8374,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.6949676275253296,
      "learning_rate": 1.7496815286624205e-05,
      "loss": 0.7976,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.5996536612510681,
      "learning_rate": 1.7433121019108282e-05,
      "loss": 0.8629,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.5606485605239868,
      "learning_rate": 1.736942675159236e-05,
      "loss": 0.8258,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.6412463784217834,
      "learning_rate": 1.7305732484076433e-05,
      "loss": 0.8287,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.7513819932937622,
      "learning_rate": 1.724203821656051e-05,
      "loss": 0.842,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.5666155219078064,
      "learning_rate": 1.7178343949044587e-05,
      "loss": 0.8332,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.5912254452705383,
      "learning_rate": 1.7114649681528663e-05,
      "loss": 0.8699,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13500,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.6989673376083374,
      "learning_rate": 1.705095541401274e-05,
      "loss": 0.8553,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13510,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.5839787125587463,
      "learning_rate": 1.6987261146496817e-05,
      "loss": 0.8446,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13520,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.6057241559028625,
      "learning_rate": 1.6923566878980894e-05,
      "loss": 0.8351,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13530,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.6739611625671387,
      "learning_rate": 1.6859872611464968e-05,
      "loss": 0.8358,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13540,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.5866987705230713,
      "learning_rate": 1.6796178343949045e-05,
      "loss": 0.8403,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13550,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.6147080659866333,
      "learning_rate": 1.673248407643312e-05,
      "loss": 0.8591,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13560,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.5726975798606873,
      "learning_rate": 1.66687898089172e-05,
      "loss": 0.8434,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13570,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.6229400038719177,
      "learning_rate": 1.6605095541401275e-05,
      "loss": 0.8634,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13580,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.7118831276893616,
      "learning_rate": 1.6541401273885352e-05,
      "loss": 0.8281,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13590,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.5629773736000061,
      "learning_rate": 1.647770700636943e-05,
      "loss": 0.8435,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13600,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.6835569143295288,
      "learning_rate": 1.6414012738853503e-05,
      "loss": 0.8255,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13610,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.5470395088195801,
      "learning_rate": 1.635031847133758e-05,
      "loss": 0.8411,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13620,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.65526282787323,
      "learning_rate": 1.6286624203821657e-05,
      "loss": 0.8264,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13630,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.6258267760276794,
      "learning_rate": 1.6222929936305733e-05,
      "loss": 0.8449,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13640,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.5335571765899658,
      "learning_rate": 1.615923566878981e-05,
      "loss": 0.8419,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13650,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.5841179490089417,
      "learning_rate": 1.6095541401273887e-05,
      "loss": 0.8447,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13660,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.5307532548904419,
      "learning_rate": 1.6031847133757964e-05,
      "loss": 0.8448,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13670,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.550396203994751,
      "learning_rate": 1.5968152866242038e-05,
      "loss": 0.8213,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13680,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.5528126955032349,
      "learning_rate": 1.5904458598726115e-05,
      "loss": 0.8329,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13690,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.5814225673675537,
      "learning_rate": 1.584076433121019e-05,
      "loss": 0.8762,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13700,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.5877110958099365,
      "learning_rate": 1.577707006369427e-05,
      "loss": 0.7988,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13710,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.5498624444007874,
      "learning_rate": 1.5713375796178345e-05,
      "loss": 0.839,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13720,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.5788300633430481,
      "learning_rate": 1.5649681528662422e-05,
      "loss": 0.8461,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13730,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.552385687828064,
      "learning_rate": 1.55859872611465e-05,
      "loss": 0.8353,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13740,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.612408459186554,
      "learning_rate": 1.5522292993630573e-05,
      "loss": 0.8423,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13750,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.5637169480323792,
      "learning_rate": 1.545859872611465e-05,
      "loss": 0.8301,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13760,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.5684060454368591,
      "learning_rate": 1.5394904458598727e-05,
      "loss": 0.8462,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13770,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.5924950242042542,
      "learning_rate": 1.5331210191082803e-05,
      "loss": 0.8544,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13780,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.6607167720794678,
      "learning_rate": 1.526751592356688e-05,
      "loss": 0.849,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13790,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.6007324457168579,
      "learning_rate": 1.5203821656050957e-05,
      "loss": 0.8168,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13800,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.592782199382782,
      "learning_rate": 1.5140127388535033e-05,
      "loss": 0.8577,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13810,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.559945821762085,
      "learning_rate": 1.507643312101911e-05,
      "loss": 0.8375,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13820,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.5329104661941528,
      "learning_rate": 1.5012738853503186e-05,
      "loss": 0.8273,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13830,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.5675476789474487,
      "learning_rate": 1.4949044585987263e-05,
      "loss": 0.8517,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13840,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.7505592703819275,
      "learning_rate": 1.4885350318471338e-05,
      "loss": 0.8623,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13850,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.6333470940589905,
      "learning_rate": 1.4821656050955415e-05,
      "loss": 0.8293,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13860,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.5732998251914978,
      "learning_rate": 1.4757961783439492e-05,
      "loss": 0.8065,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13870,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.762664794921875,
      "learning_rate": 1.4694267515923568e-05,
      "loss": 0.8442,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13880,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.6154226660728455,
      "learning_rate": 1.4630573248407644e-05,
      "loss": 0.8316,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13890,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.6072494387626648,
      "learning_rate": 1.4566878980891721e-05,
      "loss": 0.8282,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13900,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.5880212783813477,
      "learning_rate": 1.4503184713375798e-05,
      "loss": 0.8255,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13910,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.690006673336029,
      "learning_rate": 1.4439490445859873e-05,
      "loss": 0.8143,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13920,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.7075156569480896,
      "learning_rate": 1.437579617834395e-05,
      "loss": 0.8531,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13930,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.6131463646888733,
      "learning_rate": 1.4312101910828027e-05,
      "loss": 0.8441,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13940,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.5739683508872986,
      "learning_rate": 1.4248407643312103e-05,
      "loss": 0.8214,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13950,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.6377345323562622,
      "learning_rate": 1.418471337579618e-05,
      "loss": 0.8444,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13960,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.6518220901489258,
      "learning_rate": 1.4121019108280256e-05,
      "loss": 0.7968,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13970,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.607457160949707,
      "learning_rate": 1.4057324840764333e-05,
      "loss": 0.8544,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13980,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.6501799821853638,
      "learning_rate": 1.3993630573248408e-05,
      "loss": 0.816,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 13990,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.622092604637146,
      "learning_rate": 1.3929936305732485e-05,
      "loss": 0.821,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14000,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.6852797269821167,
      "learning_rate": 1.3866242038216562e-05,
      "loss": 0.8772,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14010,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.5740447640419006,
      "learning_rate": 1.380254777070064e-05,
      "loss": 0.8134,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14020,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.599518895149231,
      "learning_rate": 1.3738853503184714e-05,
      "loss": 0.8318,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14030,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.6162179708480835,
      "learning_rate": 1.3675159235668791e-05,
      "loss": 0.8333,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14040,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.6470344066619873,
      "learning_rate": 1.3611464968152868e-05,
      "loss": 0.8437,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14050,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.6147076487541199,
      "learning_rate": 1.3547770700636944e-05,
      "loss": 0.8394,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14060,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.5687283873558044,
      "learning_rate": 1.348407643312102e-05,
      "loss": 0.8219,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14070,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.5249031186103821,
      "learning_rate": 1.3420382165605097e-05,
      "loss": 0.8563,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14080,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.5840549468994141,
      "learning_rate": 1.3356687898089174e-05,
      "loss": 0.8249,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14090,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.6300859451293945,
      "learning_rate": 1.329299363057325e-05,
      "loss": 0.8453,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.6822474002838135,
      "learning_rate": 1.3229299363057326e-05,
      "loss": 0.8445,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.5768283605575562,
      "learning_rate": 1.3165605095541403e-05,
      "loss": 0.8067,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6939039826393127,
      "learning_rate": 1.3101910828025479e-05,
      "loss": 0.8396,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14130,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.5618141889572144,
      "learning_rate": 1.3038216560509555e-05,
      "loss": 0.8161,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14140,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6244920492172241,
      "learning_rate": 1.2974522292993632e-05,
      "loss": 0.8396,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14150,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.5430182814598083,
      "learning_rate": 1.291082802547771e-05,
      "loss": 0.8311,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6174887418746948,
      "learning_rate": 1.2847133757961784e-05,
      "loss": 0.833,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14170,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.5842854976654053,
      "learning_rate": 1.2783439490445861e-05,
      "loss": 0.8638,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14180,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6127358675003052,
      "learning_rate": 1.2719745222929938e-05,
      "loss": 0.8324,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14190,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.7250066995620728,
      "learning_rate": 1.2656050955414015e-05,
      "loss": 0.8343,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.5589357614517212,
      "learning_rate": 1.259235668789809e-05,
      "loss": 0.8547,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14210,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.5304135084152222,
      "learning_rate": 1.2528662420382167e-05,
      "loss": 0.8325,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14220,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.560059666633606,
      "learning_rate": 1.2464968152866243e-05,
      "loss": 0.8474,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14230,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.5708820223808289,
      "learning_rate": 1.240127388535032e-05,
      "loss": 0.8501,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.6215733885765076,
      "learning_rate": 1.2337579617834395e-05,
      "loss": 0.819,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14250,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.6077948212623596,
      "learning_rate": 1.2273885350318472e-05,
      "loss": 0.8235,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14260,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.5893474817276001,
      "learning_rate": 1.2210191082802549e-05,
      "loss": 0.8284,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14270,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.5367608070373535,
      "learning_rate": 1.2146496815286624e-05,
      "loss": 0.815,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.5464059114456177,
      "learning_rate": 1.20828025477707e-05,
      "loss": 0.8038,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14290,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.6341402530670166,
      "learning_rate": 1.2019108280254778e-05,
      "loss": 0.8415,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14300,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.6049284934997559,
      "learning_rate": 1.1955414012738854e-05,
      "loss": 0.8316,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14310,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.6299666166305542,
      "learning_rate": 1.189171974522293e-05,
      "loss": 0.8407,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.6380517482757568,
      "learning_rate": 1.1828025477707007e-05,
      "loss": 0.8387,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14330,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.6077808737754822,
      "learning_rate": 1.1764331210191084e-05,
      "loss": 0.8465,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14340,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.6068784594535828,
      "learning_rate": 1.170063694267516e-05,
      "loss": 0.8281,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14350,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.5653278827667236,
      "learning_rate": 1.1636942675159236e-05,
      "loss": 0.8244,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.6187404990196228,
      "learning_rate": 1.1573248407643313e-05,
      "loss": 0.8249,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14370,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.5997905731201172,
      "learning_rate": 1.150955414012739e-05,
      "loss": 0.8258,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14380,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.571714460849762,
      "learning_rate": 1.1445859872611465e-05,
      "loss": 0.8185,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14390,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.6664509177207947,
      "learning_rate": 1.1382165605095542e-05,
      "loss": 0.8445,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14400,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.5972470641136169,
      "learning_rate": 1.1318471337579619e-05,
      "loss": 0.8292,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14410,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.6396761536598206,
      "learning_rate": 1.1254777070063695e-05,
      "loss": 0.8392,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14420,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.6399985551834106,
      "learning_rate": 1.119108280254777e-05,
      "loss": 0.8365,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14430,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.7280391454696655,
      "learning_rate": 1.1127388535031848e-05,
      "loss": 0.862,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14440,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.542850136756897,
      "learning_rate": 1.1063694267515924e-05,
      "loss": 0.8261,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14450,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.5785331726074219,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.8261,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14460,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.5825355052947998,
      "learning_rate": 1.0936305732484077e-05,
      "loss": 0.8389,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14470,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.5818856954574585,
      "learning_rate": 1.0872611464968154e-05,
      "loss": 0.82,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14480,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.5814858078956604,
      "learning_rate": 1.080891719745223e-05,
      "loss": 0.8477,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14490,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.6301888823509216,
      "learning_rate": 1.0745222929936306e-05,
      "loss": 0.8416,
      "max_memory_allocated (GB)": 94.6,
      "memory_allocated (GB)": 14.97,
      "step": 14500,
      "total_memory_available (GB)": 94.62
    }
  ],
  "logging_steps": 10,
  "max_steps": 16186,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 4.062009042959073e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}