{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 10,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 22.5,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 1.9102,
      "step": 10
    },
    {
      "epoch": 0.02,
      "eval_loss": 1.9285269975662231,
      "eval_runtime": 2.9533,
      "eval_samples_per_second": 52.822,
      "eval_steps_per_second": 2.709,
      "step": 10
    },
    {
      "epoch": 0.04,
      "grad_norm": 14.0,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.8734,
      "step": 20
    },
    {
      "epoch": 0.04,
      "eval_loss": 1.907572627067566,
      "eval_runtime": 2.9586,
      "eval_samples_per_second": 52.728,
      "eval_steps_per_second": 2.704,
      "step": 20
    },
    {
      "epoch": 0.06,
      "grad_norm": 9.625,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.77,
      "step": 30
    },
    {
      "epoch": 0.06,
      "eval_loss": 1.8689841032028198,
      "eval_runtime": 2.9565,
      "eval_samples_per_second": 52.765,
      "eval_steps_per_second": 2.706,
      "step": 30
    },
    {
      "epoch": 0.08,
      "grad_norm": 12.75,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.8568,
      "step": 40
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.812107801437378,
      "eval_runtime": 2.9619,
      "eval_samples_per_second": 52.669,
      "eval_steps_per_second": 2.701,
      "step": 40
    },
    {
      "epoch": 0.1,
      "grad_norm": 12.875,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.7562,
      "step": 50
    },
    {
      "epoch": 0.1,
      "eval_loss": 1.7258977890014648,
      "eval_runtime": 2.9652,
      "eval_samples_per_second": 52.61,
      "eval_steps_per_second": 2.698,
      "step": 50
    },
    {
      "epoch": 0.12,
      "grad_norm": 7.8125,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.6208,
      "step": 60
    },
    {
      "epoch": 0.12,
      "eval_loss": 1.5986799001693726,
      "eval_runtime": 2.9727,
      "eval_samples_per_second": 52.478,
      "eval_steps_per_second": 2.691,
      "step": 60
    },
    {
      "epoch": 0.14,
      "grad_norm": 8.25,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 1.4408,
      "step": 70
    },
    {
      "epoch": 0.14,
      "eval_loss": 1.3833414316177368,
      "eval_runtime": 2.9613,
      "eval_samples_per_second": 52.679,
      "eval_steps_per_second": 2.701,
      "step": 70
    },
    {
      "epoch": 0.16,
      "grad_norm": 5.65625,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 1.1332,
      "step": 80
    },
    {
      "epoch": 0.16,
      "eval_loss": 1.1267226934432983,
      "eval_runtime": 2.97,
      "eval_samples_per_second": 52.525,
      "eval_steps_per_second": 2.694,
      "step": 80
    },
    {
      "epoch": 0.18,
      "grad_norm": 7.28125,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 1.0659,
      "step": 90
    },
    {
      "epoch": 0.18,
      "eval_loss": 1.0400179624557495,
      "eval_runtime": 2.9774,
      "eval_samples_per_second": 52.395,
      "eval_steps_per_second": 2.687,
      "step": 90
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.9375,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.9587,
      "step": 100
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.9716721773147583,
      "eval_runtime": 2.9692,
      "eval_samples_per_second": 52.54,
      "eval_steps_per_second": 2.694,
      "step": 100
    },
    {
      "epoch": 0.22,
      "grad_norm": 4.78125,
      "learning_rate": 4.4e-06,
      "loss": 0.9241,
      "step": 110
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.9269797205924988,
      "eval_runtime": 2.9883,
      "eval_samples_per_second": 52.204,
      "eval_steps_per_second": 2.677,
      "step": 110
    },
    {
      "epoch": 0.24,
      "grad_norm": 8.5,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.8071,
      "step": 120
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.8944205045700073,
      "eval_runtime": 2.9805,
      "eval_samples_per_second": 52.341,
      "eval_steps_per_second": 2.684,
      "step": 120
    },
    {
      "epoch": 0.26,
      "grad_norm": 3.34375,
      "learning_rate": 5.2e-06,
      "loss": 0.7872,
      "step": 130
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.874414324760437,
      "eval_runtime": 2.9734,
      "eval_samples_per_second": 52.465,
      "eval_steps_per_second": 2.691,
      "step": 130
    },
    {
      "epoch": 0.28,
      "grad_norm": 4.0625,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.7182,
      "step": 140
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.8589532375335693,
      "eval_runtime": 2.9762,
      "eval_samples_per_second": 52.415,
      "eval_steps_per_second": 2.688,
      "step": 140
    },
    {
      "epoch": 0.3,
      "grad_norm": 4.8125,
      "learning_rate": 6e-06,
      "loss": 0.7139,
      "step": 150
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.8482792973518372,
      "eval_runtime": 2.974,
      "eval_samples_per_second": 52.455,
      "eval_steps_per_second": 2.69,
      "step": 150
    },
    {
      "epoch": 0.32,
      "grad_norm": 4.4375,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.7533,
      "step": 160
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.8366221785545349,
      "eval_runtime": 2.9775,
      "eval_samples_per_second": 52.393,
      "eval_steps_per_second": 2.687,
      "step": 160
    },
    {
      "epoch": 0.34,
      "grad_norm": 3.625,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.808,
      "step": 170
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.8267260789871216,
      "eval_runtime": 2.9805,
      "eval_samples_per_second": 52.341,
      "eval_steps_per_second": 2.684,
      "step": 170
    },
    {
      "epoch": 0.36,
      "grad_norm": 4.9375,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.5991,
      "step": 180
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.8209083676338196,
      "eval_runtime": 2.9739,
      "eval_samples_per_second": 52.456,
      "eval_steps_per_second": 2.69,
      "step": 180
    },
    {
      "epoch": 0.38,
      "grad_norm": 3.671875,
      "learning_rate": 7.600000000000001e-06,
      "loss": 0.6681,
      "step": 190
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.8131274580955505,
      "eval_runtime": 2.9722,
      "eval_samples_per_second": 52.486,
      "eval_steps_per_second": 2.692,
      "step": 190
    },
    {
      "epoch": 0.4,
      "grad_norm": 3.578125,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.6707,
      "step": 200
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.8074705600738525,
      "eval_runtime": 2.9825,
      "eval_samples_per_second": 52.305,
      "eval_steps_per_second": 2.682,
      "step": 200
    },
    {
      "epoch": 0.42,
      "grad_norm": 3.71875,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.7301,
      "step": 210
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.8011156916618347,
      "eval_runtime": 2.9798,
      "eval_samples_per_second": 52.352,
      "eval_steps_per_second": 2.685,
      "step": 210
    },
    {
      "epoch": 0.44,
      "grad_norm": 3.734375,
      "learning_rate": 8.8e-06,
      "loss": 0.6486,
      "step": 220
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.7965356111526489,
      "eval_runtime": 2.9734,
      "eval_samples_per_second": 52.465,
      "eval_steps_per_second": 2.69,
      "step": 220
    },
    {
      "epoch": 0.46,
      "grad_norm": 3.34375,
      "learning_rate": 9.200000000000002e-06,
      "loss": 0.6744,
      "step": 230
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.7928165197372437,
      "eval_runtime": 2.9833,
      "eval_samples_per_second": 52.29,
      "eval_steps_per_second": 2.682,
      "step": 230
    },
    {
      "epoch": 0.48,
      "grad_norm": 3.8125,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.7228,
      "step": 240
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.7894927859306335,
      "eval_runtime": 2.9769,
      "eval_samples_per_second": 52.403,
      "eval_steps_per_second": 2.687,
      "step": 240
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.5,
      "learning_rate": 1e-05,
      "loss": 0.7532,
      "step": 250
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.7849324345588684,
      "eval_runtime": 2.9712,
      "eval_samples_per_second": 52.504,
      "eval_steps_per_second": 2.693,
      "step": 250
    },
    {
      "epoch": 0.52,
      "grad_norm": 5.8125,
      "learning_rate": 1.04e-05,
      "loss": 0.7019,
      "step": 260
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.7828314304351807,
      "eval_runtime": 2.979,
      "eval_samples_per_second": 52.366,
      "eval_steps_per_second": 2.685,
      "step": 260
    },
    {
      "epoch": 0.54,
      "grad_norm": 3.75,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 0.7255,
      "step": 270
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.7806087136268616,
      "eval_runtime": 2.9708,
      "eval_samples_per_second": 52.512,
      "eval_steps_per_second": 2.693,
      "step": 270
    },
    {
      "epoch": 0.56,
      "grad_norm": 3.75,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.6541,
      "step": 280
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.7776769995689392,
      "eval_runtime": 2.972,
      "eval_samples_per_second": 52.489,
      "eval_steps_per_second": 2.692,
      "step": 280
    },
    {
      "epoch": 0.58,
      "grad_norm": 4.09375,
      "learning_rate": 1.16e-05,
      "loss": 0.6201,
      "step": 290
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.7737365961074829,
      "eval_runtime": 2.9739,
      "eval_samples_per_second": 52.456,
      "eval_steps_per_second": 2.69,
      "step": 290
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.390625,
      "learning_rate": 1.2e-05,
      "loss": 0.6149,
      "step": 300
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.7721969485282898,
      "eval_runtime": 2.9719,
      "eval_samples_per_second": 52.492,
      "eval_steps_per_second": 2.692,
      "step": 300
    },
    {
      "epoch": 0.62,
      "grad_norm": 4.59375,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 0.6927,
      "step": 310
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.7704524397850037,
      "eval_runtime": 2.9722,
      "eval_samples_per_second": 52.487,
      "eval_steps_per_second": 2.692,
      "step": 310
    },
    {
      "epoch": 0.64,
      "grad_norm": 2.953125,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.6573,
      "step": 320
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.7695561051368713,
      "eval_runtime": 2.9675,
      "eval_samples_per_second": 52.57,
      "eval_steps_per_second": 2.696,
      "step": 320
    },
    {
      "epoch": 0.66,
      "grad_norm": 4.65625,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 0.6786,
      "step": 330
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.7669094800949097,
      "eval_runtime": 2.9692,
      "eval_samples_per_second": 52.54,
      "eval_steps_per_second": 2.694,
      "step": 330
    },
    {
      "epoch": 0.68,
      "grad_norm": 3.359375,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.7484,
      "step": 340
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.7643001079559326,
      "eval_runtime": 2.9737,
      "eval_samples_per_second": 52.46,
      "eval_steps_per_second": 2.69,
      "step": 340
    },
    {
      "epoch": 0.7,
      "grad_norm": 4.4375,
      "learning_rate": 1.4e-05,
      "loss": 0.5991,
      "step": 350
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.7622258067131042,
      "eval_runtime": 2.9746,
      "eval_samples_per_second": 52.444,
      "eval_steps_per_second": 2.689,
      "step": 350
    },
    {
      "epoch": 0.72,
      "grad_norm": 3.359375,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.6621,
      "step": 360
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.761843204498291,
      "eval_runtime": 2.9802,
      "eval_samples_per_second": 52.345,
      "eval_steps_per_second": 2.684,
      "step": 360
    },
    {
      "epoch": 0.74,
      "grad_norm": 3.890625,
      "learning_rate": 1.48e-05,
      "loss": 0.697,
      "step": 370
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.7606173753738403,
      "eval_runtime": 2.9702,
      "eval_samples_per_second": 52.522,
      "eval_steps_per_second": 2.693,
      "step": 370
    },
    {
      "epoch": 0.76,
      "grad_norm": 3.15625,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.6213,
      "step": 380
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.7589120268821716,
      "eval_runtime": 2.9752,
      "eval_samples_per_second": 52.434,
      "eval_steps_per_second": 2.689,
      "step": 380
    },
    {
      "epoch": 0.78,
      "grad_norm": 3.96875,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 0.6767,
      "step": 390
    },
    {
      "epoch": 0.78,
      "eval_loss": 0.7586512565612793,
      "eval_runtime": 2.9725,
      "eval_samples_per_second": 52.482,
      "eval_steps_per_second": 2.691,
      "step": 390
    },
    {
      "epoch": 0.8,
      "grad_norm": 4.34375,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.8044,
      "step": 400
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.7563453316688538,
      "eval_runtime": 2.9763,
      "eval_samples_per_second": 52.414,
      "eval_steps_per_second": 2.688,
      "step": 400
    },
    {
      "epoch": 0.82,
      "grad_norm": 3.203125,
      "learning_rate": 1.64e-05,
      "loss": 0.6338,
      "step": 410
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.7531482577323914,
      "eval_runtime": 2.9746,
      "eval_samples_per_second": 52.444,
      "eval_steps_per_second": 2.689,
      "step": 410
    },
    {
      "epoch": 0.84,
      "grad_norm": 3.140625,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.6242,
      "step": 420
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.7514854669570923,
      "eval_runtime": 2.9804,
      "eval_samples_per_second": 52.342,
      "eval_steps_per_second": 2.684,
      "step": 420
    },
    {
      "epoch": 0.86,
      "grad_norm": 3.203125,
      "learning_rate": 1.72e-05,
      "loss": 0.6462,
      "step": 430
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.7528823614120483,
      "eval_runtime": 2.9744,
      "eval_samples_per_second": 52.448,
      "eval_steps_per_second": 2.69,
      "step": 430
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.9375,
      "learning_rate": 1.76e-05,
      "loss": 0.6832,
      "step": 440
    },
    {
      "epoch": 0.88,
      "eval_loss": 0.7538066506385803,
      "eval_runtime": 2.9692,
      "eval_samples_per_second": 52.54,
      "eval_steps_per_second": 2.694,
      "step": 440
    },
    {
      "epoch": 0.9,
      "grad_norm": 6.28125,
      "learning_rate": 1.8e-05,
      "loss": 0.5959,
      "step": 450
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.7551948428153992,
      "eval_runtime": 2.9739,
      "eval_samples_per_second": 52.456,
      "eval_steps_per_second": 2.69,
      "step": 450
    },
    {
      "epoch": 0.92,
      "grad_norm": 7.5625,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.6891,
      "step": 460
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.7493566870689392,
      "eval_runtime": 2.976,
      "eval_samples_per_second": 52.42,
      "eval_steps_per_second": 2.688,
      "step": 460
    },
    {
      "epoch": 0.94,
      "grad_norm": 5.75,
      "learning_rate": 1.88e-05,
      "loss": 0.6819,
      "step": 470
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.7475622296333313,
      "eval_runtime": 2.9745,
      "eval_samples_per_second": 52.445,
      "eval_steps_per_second": 2.69,
      "step": 470
    },
    {
      "epoch": 0.96,
      "grad_norm": 3.0,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.6456,
      "step": 480
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.7493394613265991,
      "eval_runtime": 2.968,
      "eval_samples_per_second": 52.56,
      "eval_steps_per_second": 2.695,
      "step": 480
    },
    {
      "epoch": 0.98,
      "grad_norm": 3.65625,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.6415,
      "step": 490
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.7468904256820679,
      "eval_runtime": 2.972,
      "eval_samples_per_second": 52.491,
      "eval_steps_per_second": 2.692,
      "step": 490
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.421875,
      "learning_rate": 2e-05,
      "loss": 0.658,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7486743330955505,
      "eval_runtime": 2.9736,
      "eval_samples_per_second": 52.462,
      "eval_steps_per_second": 2.69,
      "step": 500
    },
    {
      "epoch": 1.02,
      "grad_norm": 5.625,
      "learning_rate": 1.9999756307053947e-05,
      "loss": 0.547,
      "step": 510
    },
    {
      "epoch": 1.02,
      "eval_loss": 0.7537599205970764,
      "eval_runtime": 2.957,
      "eval_samples_per_second": 52.757,
      "eval_steps_per_second": 2.705,
      "step": 510
    },
    {
      "epoch": 1.04,
      "grad_norm": 3.09375,
      "learning_rate": 1.9999025240093045e-05,
      "loss": 0.5551,
      "step": 520
    },
    {
      "epoch": 1.04,
      "eval_loss": 0.7580664157867432,
      "eval_runtime": 2.9622,
      "eval_samples_per_second": 52.664,
      "eval_steps_per_second": 2.701,
      "step": 520
    },
    {
      "epoch": 1.06,
      "grad_norm": 2.828125,
      "learning_rate": 1.9997806834748455e-05,
      "loss": 0.596,
      "step": 530
    },
    {
      "epoch": 1.06,
      "eval_loss": 0.7601476907730103,
      "eval_runtime": 2.9678,
      "eval_samples_per_second": 52.564,
      "eval_steps_per_second": 2.696,
      "step": 530
    },
    {
      "epoch": 1.08,
      "grad_norm": 2.3125,
      "learning_rate": 1.9996101150403543e-05,
      "loss": 0.5611,
      "step": 540
    },
    {
      "epoch": 1.08,
      "eval_loss": 0.7624161243438721,
      "eval_runtime": 2.9644,
      "eval_samples_per_second": 52.624,
      "eval_steps_per_second": 2.699,
      "step": 540
    },
    {
      "epoch": 1.1,
      "grad_norm": 2.953125,
      "learning_rate": 1.999390827019096e-05,
      "loss": 0.5514,
      "step": 550
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.7590138912200928,
      "eval_runtime": 2.9709,
      "eval_samples_per_second": 52.51,
      "eval_steps_per_second": 2.693,
      "step": 550
    },
    {
      "epoch": 1.12,
      "grad_norm": 3.59375,
      "learning_rate": 1.9991228300988586e-05,
      "loss": 0.5109,
      "step": 560
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.754431426525116,
      "eval_runtime": 2.9642,
      "eval_samples_per_second": 52.628,
      "eval_steps_per_second": 2.699,
      "step": 560
    },
    {
      "epoch": 1.14,
      "grad_norm": 3.03125,
      "learning_rate": 1.9988061373414342e-05,
      "loss": 0.5984,
      "step": 570
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.7549035549163818,
      "eval_runtime": 2.9676,
      "eval_samples_per_second": 52.567,
      "eval_steps_per_second": 2.696,
      "step": 570
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.9296875,
      "learning_rate": 1.9984407641819812e-05,
      "loss": 0.5346,
      "step": 580
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.7530075311660767,
      "eval_runtime": 2.9726,
      "eval_samples_per_second": 52.48,
      "eval_steps_per_second": 2.691,
      "step": 580
    },
    {
      "epoch": 1.18,
      "grad_norm": 3.15625,
      "learning_rate": 1.9980267284282718e-05,
      "loss": 0.6438,
      "step": 590
    },
    {
      "epoch": 1.18,
      "eval_loss": 0.7559706568717957,
      "eval_runtime": 2.9783,
      "eval_samples_per_second": 52.378,
      "eval_steps_per_second": 2.686,
      "step": 590
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.0,
      "learning_rate": 1.9975640502598243e-05,
      "loss": 0.5193,
      "step": 600
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.7533587217330933,
      "eval_runtime": 2.9733,
      "eval_samples_per_second": 52.468,
      "eval_steps_per_second": 2.691,
      "step": 600
    },
    {
      "epoch": 1.22,
      "grad_norm": 4.34375,
      "learning_rate": 1.9970527522269204e-05,
      "loss": 0.6232,
      "step": 610
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.7505587339401245,
      "eval_runtime": 2.9729,
      "eval_samples_per_second": 52.474,
      "eval_steps_per_second": 2.691,
      "step": 610
    },
    {
      "epoch": 1.24,
      "grad_norm": 2.125,
      "learning_rate": 1.9964928592495046e-05,
      "loss": 0.6118,
      "step": 620
    },
    {
      "epoch": 1.24,
      "eval_loss": 0.747382402420044,
      "eval_runtime": 2.9751,
      "eval_samples_per_second": 52.436,
      "eval_steps_per_second": 2.689,
      "step": 620
    },
    {
      "epoch": 1.26,
      "grad_norm": 3.21875,
      "learning_rate": 1.9958843986159705e-05,
      "loss": 0.6072,
      "step": 630
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.7488105893135071,
      "eval_runtime": 2.9725,
      "eval_samples_per_second": 52.481,
      "eval_steps_per_second": 2.691,
      "step": 630
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.71875,
      "learning_rate": 1.9952273999818312e-05,
      "loss": 0.6088,
      "step": 640
    },
    {
      "epoch": 1.28,
      "eval_loss": 0.7461674809455872,
      "eval_runtime": 2.9722,
      "eval_samples_per_second": 52.486,
      "eval_steps_per_second": 2.692,
      "step": 640
    },
    {
      "epoch": 1.3,
      "grad_norm": 2.5,
      "learning_rate": 1.9945218953682736e-05,
      "loss": 0.6276,
      "step": 650
    },
    {
      "epoch": 1.3,
      "eval_loss": 0.7464598417282104,
      "eval_runtime": 2.9781,
      "eval_samples_per_second": 52.383,
      "eval_steps_per_second": 2.686,
      "step": 650
    },
    {
      "epoch": 1.32,
      "grad_norm": 2.421875,
      "learning_rate": 1.9937679191605964e-05,
      "loss": 0.5798,
      "step": 660
    },
    {
      "epoch": 1.32,
      "eval_loss": 0.7467732429504395,
      "eval_runtime": 2.977,
      "eval_samples_per_second": 52.403,
      "eval_steps_per_second": 2.687,
      "step": 660
    },
    {
      "epoch": 1.34,
      "grad_norm": 3.4375,
      "learning_rate": 1.992965508106537e-05,
      "loss": 0.6283,
      "step": 670
    },
    {
      "epoch": 1.34,
      "eval_loss": 0.7440893054008484,
      "eval_runtime": 2.9772,
      "eval_samples_per_second": 52.398,
      "eval_steps_per_second": 2.687,
      "step": 670
    },
    {
      "epoch": 1.36,
      "grad_norm": 2.78125,
      "learning_rate": 1.9921147013144782e-05,
      "loss": 0.5477,
      "step": 680
    },
    {
      "epoch": 1.36,
      "eval_loss": 0.745173454284668,
      "eval_runtime": 2.9776,
      "eval_samples_per_second": 52.392,
      "eval_steps_per_second": 2.687,
      "step": 680
    },
    {
      "epoch": 1.38,
      "grad_norm": 2.21875,
      "learning_rate": 1.991215540251542e-05,
      "loss": 0.5479,
      "step": 690
    },
    {
      "epoch": 1.38,
      "eval_loss": 0.7425445318222046,
      "eval_runtime": 2.9811,
      "eval_samples_per_second": 52.33,
      "eval_steps_per_second": 2.684,
      "step": 690
    },
    {
      "epoch": 1.4,
      "grad_norm": 3.171875,
      "learning_rate": 1.9902680687415704e-05,
      "loss": 0.6648,
      "step": 700
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.7418652772903442,
      "eval_runtime": 2.9712,
      "eval_samples_per_second": 52.503,
      "eval_steps_per_second": 2.692,
      "step": 700
    },
    {
      "epoch": 1.42,
      "grad_norm": 3.5,
      "learning_rate": 1.9892723329629885e-05,
      "loss": 0.5988,
      "step": 710
    },
    {
      "epoch": 1.42,
      "eval_loss": 0.7394148707389832,
      "eval_runtime": 2.9689,
      "eval_samples_per_second": 52.544,
      "eval_steps_per_second": 2.695,
      "step": 710
    },
    {
      "epoch": 1.44,
      "grad_norm": 2.984375,
      "learning_rate": 1.988228381446553e-05,
      "loss": 0.5557,
      "step": 720
    },
    {
      "epoch": 1.44,
      "eval_loss": 0.7416422963142395,
      "eval_runtime": 2.9747,
      "eval_samples_per_second": 52.442,
      "eval_steps_per_second": 2.689,
      "step": 720
    },
    {
      "epoch": 1.46,
      "grad_norm": 2.421875,
      "learning_rate": 1.987136265072988e-05,
      "loss": 0.4506,
      "step": 730
    },
    {
      "epoch": 1.46,
      "eval_loss": 0.7449523210525513,
      "eval_runtime": 2.9679,
      "eval_samples_per_second": 52.563,
      "eval_steps_per_second": 2.696,
      "step": 730
    },
    {
      "epoch": 1.48,
      "grad_norm": 3.265625,
      "learning_rate": 1.985996037070505e-05,
      "loss": 0.6386,
      "step": 740
    },
    {
      "epoch": 1.48,
      "eval_loss": 0.7423770427703857,
      "eval_runtime": 2.9762,
      "eval_samples_per_second": 52.416,
      "eval_steps_per_second": 2.688,
      "step": 740
    },
    {
      "epoch": 1.5,
      "grad_norm": 4.1875,
      "learning_rate": 1.9848077530122083e-05,
      "loss": 0.5889,
      "step": 750
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.7404635548591614,
      "eval_runtime": 2.9716,
      "eval_samples_per_second": 52.497,
      "eval_steps_per_second": 2.692,
      "step": 750
    },
    {
      "epoch": 1.52,
      "grad_norm": 2.546875,
      "learning_rate": 1.983571470813386e-05,
      "loss": 0.5561,
      "step": 760
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.7413467764854431,
      "eval_runtime": 2.9834,
      "eval_samples_per_second": 52.29,
      "eval_steps_per_second": 2.682,
      "step": 760
    },
    {
      "epoch": 1.54,
      "grad_norm": 2.125,
      "learning_rate": 1.982287250728689e-05,
      "loss": 0.5307,
      "step": 770
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.7405444383621216,
      "eval_runtime": 2.9797,
      "eval_samples_per_second": 52.355,
      "eval_steps_per_second": 2.685,
      "step": 770
    },
    {
      "epoch": 1.56,
      "grad_norm": 2.0625,
      "learning_rate": 1.9809551553491918e-05,
      "loss": 0.5464,
      "step": 780
    },
    {
      "epoch": 1.56,
      "eval_loss": 0.742436945438385,
      "eval_runtime": 2.973,
      "eval_samples_per_second": 52.471,
      "eval_steps_per_second": 2.691,
      "step": 780
    },
    {
      "epoch": 1.58,
      "grad_norm": 2.3125,
      "learning_rate": 1.979575249599344e-05,
      "loss": 0.5726,
      "step": 790
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.7410677671432495,
      "eval_runtime": 2.968,
      "eval_samples_per_second": 52.561,
      "eval_steps_per_second": 2.695,
      "step": 790
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.578125,
      "learning_rate": 1.9781476007338058e-05,
      "loss": 0.6039,
      "step": 800
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.7446211576461792,
      "eval_runtime": 2.9788,
      "eval_samples_per_second": 52.37,
      "eval_steps_per_second": 2.686,
      "step": 800
    },
    {
      "epoch": 1.62,
      "grad_norm": 3.5,
      "learning_rate": 1.9766722783341682e-05,
      "loss": 0.5236,
      "step": 810
    },
    {
      "epoch": 1.62,
      "eval_loss": 0.7396934628486633,
      "eval_runtime": 2.9783,
      "eval_samples_per_second": 52.378,
      "eval_steps_per_second": 2.686,
      "step": 810
    },
    {
      "epoch": 1.64,
      "grad_norm": 3.703125,
      "learning_rate": 1.9751493543055634e-05,
      "loss": 0.5176,
      "step": 820
    },
    {
      "epoch": 1.64,
      "eval_loss": 0.7380808591842651,
      "eval_runtime": 2.9661,
      "eval_samples_per_second": 52.595,
      "eval_steps_per_second": 2.697,
      "step": 820
    },
    {
      "epoch": 1.66,
      "grad_norm": 2.25,
      "learning_rate": 1.9735789028731603e-05,
      "loss": 0.5991,
      "step": 830
    },
    {
      "epoch": 1.66,
      "eval_loss": 0.7352851629257202,
      "eval_runtime": 2.9801,
      "eval_samples_per_second": 52.348,
      "eval_steps_per_second": 2.685,
      "step": 830
    },
    {
      "epoch": 1.68,
      "grad_norm": 3.71875,
      "learning_rate": 1.9719610005785466e-05,
      "loss": 0.5451,
      "step": 840
    },
    {
      "epoch": 1.68,
      "eval_loss": 0.7372366786003113,
      "eval_runtime": 2.9742,
      "eval_samples_per_second": 52.452,
      "eval_steps_per_second": 2.69,
      "step": 840
    },
    {
      "epoch": 1.7,
      "grad_norm": 2.140625,
      "learning_rate": 1.9702957262759964e-05,
      "loss": 0.4484,
      "step": 850
    },
    {
      "epoch": 1.7,
      "eval_loss": 0.7352163791656494,
      "eval_runtime": 2.9703,
      "eval_samples_per_second": 52.52,
      "eval_steps_per_second": 2.693,
      "step": 850
    },
    {
      "epoch": 1.72,
      "grad_norm": 2.125,
      "learning_rate": 1.9685831611286312e-05,
      "loss": 0.5568,
      "step": 860
    },
    {
      "epoch": 1.72,
      "eval_loss": 0.7367239594459534,
      "eval_runtime": 2.9748,
      "eval_samples_per_second": 52.441,
      "eval_steps_per_second": 2.689,
      "step": 860
    },
    {
      "epoch": 1.74,
      "grad_norm": 2.890625,
      "learning_rate": 1.9668233886044597e-05,
      "loss": 0.5643,
      "step": 870
    },
    {
      "epoch": 1.74,
      "eval_loss": 0.7365850806236267,
      "eval_runtime": 2.9727,
      "eval_samples_per_second": 52.477,
      "eval_steps_per_second": 2.691,
      "step": 870
    },
    {
      "epoch": 1.76,
      "grad_norm": 3.3125,
      "learning_rate": 1.9650164944723116e-05,
      "loss": 0.5341,
      "step": 880
    },
    {
      "epoch": 1.76,
      "eval_loss": 0.7373999357223511,
      "eval_runtime": 2.9825,
      "eval_samples_per_second": 52.305,
      "eval_steps_per_second": 2.682,
      "step": 880
    },
    {
      "epoch": 1.78,
      "grad_norm": 3.625,
      "learning_rate": 1.9631625667976584e-05,
      "loss": 0.5255,
      "step": 890
    },
    {
      "epoch": 1.78,
      "eval_loss": 0.7373459935188293,
      "eval_runtime": 2.9729,
      "eval_samples_per_second": 52.474,
      "eval_steps_per_second": 2.691,
      "step": 890
    },
    {
      "epoch": 1.8,
      "grad_norm": 3.296875,
      "learning_rate": 1.961261695938319e-05,
      "loss": 0.5351,
      "step": 900
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.7352133393287659,
      "eval_runtime": 2.9749,
      "eval_samples_per_second": 52.439,
      "eval_steps_per_second": 2.689,
      "step": 900
    },
    {
      "epoch": 1.82,
      "grad_norm": 3.15625,
      "learning_rate": 1.9593139745400575e-05,
      "loss": 0.6188,
      "step": 910
    },
    {
      "epoch": 1.82,
      "eval_loss": 0.7367925643920898,
      "eval_runtime": 2.9685,
      "eval_samples_per_second": 52.552,
      "eval_steps_per_second": 2.695,
      "step": 910
    },
    {
      "epoch": 1.84,
      "grad_norm": 2.71875,
      "learning_rate": 1.9573194975320672e-05,
      "loss": 0.4968,
      "step": 920
    },
    {
      "epoch": 1.84,
      "eval_loss": 0.7363703846931458,
      "eval_runtime": 2.9748,
      "eval_samples_per_second": 52.441,
      "eval_steps_per_second": 2.689,
      "step": 920
    },
    {
      "epoch": 1.86,
      "grad_norm": 2.5625,
      "learning_rate": 1.9552783621223437e-05,
      "loss": 0.543,
      "step": 930
    },
    {
      "epoch": 1.86,
      "eval_loss": 0.7352641224861145,
      "eval_runtime": 2.9728,
      "eval_samples_per_second": 52.476,
      "eval_steps_per_second": 2.691,
      "step": 930
    },
    {
      "epoch": 1.88,
      "grad_norm": 2.921875,
      "learning_rate": 1.9531906677929472e-05,
      "loss": 0.6299,
      "step": 940
    },
    {
      "epoch": 1.88,
      "eval_loss": 0.7344778776168823,
      "eval_runtime": 2.994,
      "eval_samples_per_second": 52.104,
      "eval_steps_per_second": 2.672,
      "step": 940
    },
    {
      "epoch": 1.9,
      "grad_norm": 2.703125,
      "learning_rate": 1.9510565162951538e-05,
      "loss": 0.6208,
      "step": 950
    },
    {
      "epoch": 1.9,
      "eval_loss": 0.7322725057601929,
      "eval_runtime": 2.9757,
      "eval_samples_per_second": 52.425,
      "eval_steps_per_second": 2.688,
      "step": 950
    },
    {
      "epoch": 1.92,
      "grad_norm": 2.453125,
      "learning_rate": 1.9488760116444966e-05,
      "loss": 0.5667,
      "step": 960
    },
    {
      "epoch": 1.92,
      "eval_loss": 0.7319491505622864,
      "eval_runtime": 2.9677,
      "eval_samples_per_second": 52.565,
      "eval_steps_per_second": 2.696,
      "step": 960
    },
    {
      "epoch": 1.94,
      "grad_norm": 3.0,
      "learning_rate": 1.9466492601156964e-05,
      "loss": 0.5686,
      "step": 970
    },
    {
      "epoch": 1.94,
      "eval_loss": 0.7332378625869751,
      "eval_runtime": 2.9718,
      "eval_samples_per_second": 52.494,
      "eval_steps_per_second": 2.692,
      "step": 970
    },
    {
      "epoch": 1.96,
      "grad_norm": 2.890625,
      "learning_rate": 1.944376370237481e-05,
      "loss": 0.5745,
      "step": 980
    },
    {
      "epoch": 1.96,
      "eval_loss": 0.7319948077201843,
      "eval_runtime": 2.9667,
      "eval_samples_per_second": 52.584,
      "eval_steps_per_second": 2.697,
      "step": 980
    },
    {
      "epoch": 1.98,
      "grad_norm": 2.71875,
      "learning_rate": 1.942057452787297e-05,
      "loss": 0.5632,
      "step": 990
    },
    {
      "epoch": 1.98,
      "eval_loss": 0.7325617074966431,
      "eval_runtime": 2.9757,
      "eval_samples_per_second": 52.425,
      "eval_steps_per_second": 2.688,
      "step": 990
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.84375,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 0.5558,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7342491149902344,
      "eval_runtime": 2.9766,
      "eval_samples_per_second": 52.41,
      "eval_steps_per_second": 2.688,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 5.34460687754199e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}