{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 500.0,
  "global_step": 22000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.55,
      "eval_loss": 1.7275390625,
      "eval_runtime": 2.8452,
      "eval_samples_per_second": 113.875,
      "eval_steps_per_second": 14.41,
      "step": 200
    },
    {
      "epoch": 9.09,
      "eval_loss": 1.052734375,
      "eval_runtime": 2.8537,
      "eval_samples_per_second": 113.538,
      "eval_steps_per_second": 14.367,
      "step": 400
    },
    {
      "epoch": 11.36,
      "learning_rate": 5e-05,
      "loss": 1.9751,
      "step": 500
    },
    {
      "epoch": 13.64,
      "eval_loss": 0.64697265625,
      "eval_runtime": 2.8455,
      "eval_samples_per_second": 113.862,
      "eval_steps_per_second": 14.409,
      "step": 600
    },
    {
      "epoch": 18.18,
      "eval_loss": 0.3955078125,
      "eval_runtime": 2.8586,
      "eval_samples_per_second": 113.342,
      "eval_steps_per_second": 14.343,
      "step": 800
    },
    {
      "epoch": 22.73,
      "learning_rate": 5e-05,
      "loss": 0.7515,
      "step": 1000
    },
    {
      "epoch": 22.73,
      "eval_loss": 0.25244140625,
      "eval_runtime": 2.8476,
      "eval_samples_per_second": 113.78,
      "eval_steps_per_second": 14.398,
      "step": 1000
    },
    {
      "epoch": 27.27,
      "eval_loss": 0.167236328125,
      "eval_runtime": 2.8473,
      "eval_samples_per_second": 113.792,
      "eval_steps_per_second": 14.4,
      "step": 1200
    },
    {
      "epoch": 31.82,
      "eval_loss": 0.1171875,
      "eval_runtime": 2.8502,
      "eval_samples_per_second": 113.678,
      "eval_steps_per_second": 14.385,
      "step": 1400
    },
    {
      "epoch": 34.09,
      "learning_rate": 5e-05,
      "loss": 0.3181,
      "step": 1500
    },
    {
      "epoch": 36.36,
      "eval_loss": 0.089599609375,
      "eval_runtime": 2.853,
      "eval_samples_per_second": 113.564,
      "eval_steps_per_second": 14.371,
      "step": 1600
    },
    {
      "epoch": 40.91,
      "eval_loss": 0.07159423828125,
      "eval_runtime": 2.8542,
      "eval_samples_per_second": 113.519,
      "eval_steps_per_second": 14.365,
      "step": 1800
    },
    {
      "epoch": 45.45,
      "learning_rate": 5e-05,
      "loss": 0.1654,
      "step": 2000
    },
    {
      "epoch": 45.45,
      "eval_loss": 0.06207275390625,
      "eval_runtime": 2.8559,
      "eval_samples_per_second": 113.449,
      "eval_steps_per_second": 14.356,
      "step": 2000
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.053802490234375,
      "eval_runtime": 2.8461,
      "eval_samples_per_second": 113.841,
      "eval_steps_per_second": 14.406,
      "step": 2200
    },
    {
      "epoch": 54.55,
      "eval_loss": 0.049713134765625,
      "eval_runtime": 2.8533,
      "eval_samples_per_second": 113.551,
      "eval_steps_per_second": 14.369,
      "step": 2400
    },
    {
      "epoch": 56.82,
      "learning_rate": 5e-05,
      "loss": 0.1059,
      "step": 2500
    },
    {
      "epoch": 59.09,
      "eval_loss": 0.047698974609375,
      "eval_runtime": 2.8541,
      "eval_samples_per_second": 113.523,
      "eval_steps_per_second": 14.366,
      "step": 2600
    },
    {
      "epoch": 63.64,
      "eval_loss": 0.04449462890625,
      "eval_runtime": 2.8515,
      "eval_samples_per_second": 113.625,
      "eval_steps_per_second": 14.379,
      "step": 2800
    },
    {
      "epoch": 68.18,
      "learning_rate": 5e-05,
      "loss": 0.0771,
      "step": 3000
    },
    {
      "epoch": 68.18,
      "eval_loss": 0.04400634765625,
      "eval_runtime": 2.853,
      "eval_samples_per_second": 113.565,
      "eval_steps_per_second": 14.371,
      "step": 3000
    },
    {
      "epoch": 72.73,
      "eval_loss": 0.03985595703125,
      "eval_runtime": 2.8483,
      "eval_samples_per_second": 113.753,
      "eval_steps_per_second": 14.395,
      "step": 3200
    },
    {
      "epoch": 77.27,
      "eval_loss": 0.039398193359375,
      "eval_runtime": 2.849,
      "eval_samples_per_second": 113.725,
      "eval_steps_per_second": 14.391,
      "step": 3400
    },
    {
      "epoch": 79.55,
      "learning_rate": 5e-05,
      "loss": 0.0604,
      "step": 3500
    },
    {
      "epoch": 81.82,
      "eval_loss": 0.03973388671875,
      "eval_runtime": 2.8542,
      "eval_samples_per_second": 113.518,
      "eval_steps_per_second": 14.365,
      "step": 3600
    },
    {
      "epoch": 86.36,
      "eval_loss": 0.038177490234375,
      "eval_runtime": 2.85,
      "eval_samples_per_second": 113.682,
      "eval_steps_per_second": 14.386,
      "step": 3800
    },
    {
      "epoch": 90.91,
      "learning_rate": 5e-05,
      "loss": 0.05,
      "step": 4000
    },
    {
      "epoch": 90.91,
      "eval_loss": 0.037384033203125,
      "eval_runtime": 2.8526,
      "eval_samples_per_second": 113.581,
      "eval_steps_per_second": 14.373,
      "step": 4000
    },
    {
      "epoch": 95.45,
      "eval_loss": 0.037322998046875,
      "eval_runtime": 2.8554,
      "eval_samples_per_second": 113.468,
      "eval_steps_per_second": 14.359,
      "step": 4200
    },
    {
      "epoch": 100.0,
      "eval_loss": 0.035858154296875,
      "eval_runtime": 2.8533,
      "eval_samples_per_second": 113.555,
      "eval_steps_per_second": 14.37,
      "step": 4400
    },
    {
      "epoch": 102.27,
      "learning_rate": 5e-05,
      "loss": 0.0423,
      "step": 4500
    },
    {
      "epoch": 104.55,
      "eval_loss": 0.036163330078125,
      "eval_runtime": 2.8573,
      "eval_samples_per_second": 113.392,
      "eval_steps_per_second": 14.349,
      "step": 4600
    },
    {
      "epoch": 109.09,
      "eval_loss": 0.0374755859375,
      "eval_runtime": 2.8521,
      "eval_samples_per_second": 113.599,
      "eval_steps_per_second": 14.375,
      "step": 4800
    },
    {
      "epoch": 113.64,
      "learning_rate": 5e-05,
      "loss": 0.0364,
      "step": 5000
    },
    {
      "epoch": 113.64,
      "eval_loss": 0.035125732421875,
      "eval_runtime": 2.8541,
      "eval_samples_per_second": 113.521,
      "eval_steps_per_second": 14.365,
      "step": 5000
    },
    {
      "epoch": 118.18,
      "eval_loss": 0.034515380859375,
      "eval_runtime": 2.8551,
      "eval_samples_per_second": 113.483,
      "eval_steps_per_second": 14.361,
      "step": 5200
    },
    {
      "epoch": 122.73,
      "eval_loss": 0.036224365234375,
      "eval_runtime": 2.8548,
      "eval_samples_per_second": 113.491,
      "eval_steps_per_second": 14.362,
      "step": 5400
    },
    {
      "epoch": 125.0,
      "learning_rate": 5e-05,
      "loss": 0.0321,
      "step": 5500
    },
    {
      "epoch": 127.27,
      "eval_loss": 0.0335693359375,
      "eval_runtime": 2.8567,
      "eval_samples_per_second": 113.418,
      "eval_steps_per_second": 14.352,
      "step": 5600
    },
    {
      "epoch": 131.82,
      "eval_loss": 0.03485107421875,
      "eval_runtime": 2.853,
      "eval_samples_per_second": 113.566,
      "eval_steps_per_second": 14.371,
      "step": 5800
    },
    {
      "epoch": 136.36,
      "learning_rate": 5e-05,
      "loss": 0.0285,
      "step": 6000
    },
    {
      "epoch": 136.36,
      "eval_loss": 0.03436279296875,
      "eval_runtime": 2.8569,
      "eval_samples_per_second": 113.41,
      "eval_steps_per_second": 14.351,
      "step": 6000
    },
    {
      "epoch": 140.91,
      "eval_loss": 0.03497314453125,
      "eval_runtime": 2.8533,
      "eval_samples_per_second": 113.554,
      "eval_steps_per_second": 14.37,
      "step": 6200
    },
    {
      "epoch": 145.45,
      "eval_loss": 0.034088134765625,
      "eval_runtime": 2.8565,
      "eval_samples_per_second": 113.424,
      "eval_steps_per_second": 14.353,
      "step": 6400
    },
    {
      "epoch": 147.73,
      "learning_rate": 5e-05,
      "loss": 0.0257,
      "step": 6500
    },
    {
      "epoch": 150.0,
      "eval_loss": 0.0341796875,
      "eval_runtime": 2.8575,
      "eval_samples_per_second": 113.385,
      "eval_steps_per_second": 14.348,
      "step": 6600
    },
    {
      "epoch": 154.55,
      "eval_loss": 0.035675048828125,
      "eval_runtime": 2.8571,
      "eval_samples_per_second": 113.402,
      "eval_steps_per_second": 14.35,
      "step": 6800
    },
    {
      "epoch": 159.09,
      "learning_rate": 5e-05,
      "loss": 0.0235,
      "step": 7000
    },
    {
      "epoch": 159.09,
      "eval_loss": 0.0352783203125,
      "eval_runtime": 2.854,
      "eval_samples_per_second": 113.526,
      "eval_steps_per_second": 14.366,
      "step": 7000
    },
    {
      "epoch": 163.64,
      "eval_loss": 0.034881591796875,
      "eval_runtime": 2.8546,
      "eval_samples_per_second": 113.502,
      "eval_steps_per_second": 14.363,
      "step": 7200
    },
    {
      "epoch": 168.18,
      "eval_loss": 0.035003662109375,
      "eval_runtime": 2.8564,
      "eval_samples_per_second": 113.428,
      "eval_steps_per_second": 14.354,
      "step": 7400
    },
    {
      "epoch": 170.45,
      "learning_rate": 5e-05,
      "loss": 0.0216,
      "step": 7500
    },
    {
      "epoch": 172.73,
      "eval_loss": 0.034149169921875,
      "eval_runtime": 2.8551,
      "eval_samples_per_second": 113.483,
      "eval_steps_per_second": 14.36,
      "step": 7600
    },
    {
      "epoch": 177.27,
      "eval_loss": 0.035186767578125,
      "eval_runtime": 2.8549,
      "eval_samples_per_second": 113.49,
      "eval_steps_per_second": 14.361,
      "step": 7800
    },
    {
      "epoch": 181.82,
      "learning_rate": 5e-05,
      "loss": 0.0199,
      "step": 8000
    },
    {
      "epoch": 181.82,
      "eval_loss": 0.034149169921875,
      "eval_runtime": 2.8566,
      "eval_samples_per_second": 113.422,
      "eval_steps_per_second": 14.353,
      "step": 8000
    },
    {
      "epoch": 186.36,
      "eval_loss": 0.035308837890625,
      "eval_runtime": 2.8525,
      "eval_samples_per_second": 113.584,
      "eval_steps_per_second": 14.373,
      "step": 8200
    },
    {
      "epoch": 190.91,
      "eval_loss": 0.03472900390625,
      "eval_runtime": 2.8548,
      "eval_samples_per_second": 113.494,
      "eval_steps_per_second": 14.362,
      "step": 8400
    },
    {
      "epoch": 193.18,
      "learning_rate": 5e-05,
      "loss": 0.0184,
      "step": 8500
    },
    {
      "epoch": 195.45,
      "eval_loss": 0.034210205078125,
      "eval_runtime": 2.8577,
      "eval_samples_per_second": 113.377,
      "eval_steps_per_second": 14.347,
      "step": 8600
    },
    {
      "epoch": 200.0,
      "eval_loss": 0.03448486328125,
      "eval_runtime": 2.8551,
      "eval_samples_per_second": 113.48,
      "eval_steps_per_second": 14.36,
      "step": 8800
    },
    {
      "epoch": 204.55,
      "learning_rate": 5e-05,
      "loss": 0.0173,
      "step": 9000
    },
    {
      "epoch": 204.55,
      "eval_loss": 0.034454345703125,
      "eval_runtime": 2.8538,
      "eval_samples_per_second": 113.533,
      "eval_steps_per_second": 14.367,
      "step": 9000
    },
    {
      "epoch": 209.09,
      "eval_loss": 0.0343017578125,
      "eval_runtime": 2.8484,
      "eval_samples_per_second": 113.747,
      "eval_steps_per_second": 14.394,
      "step": 9200
    },
    {
      "epoch": 213.64,
      "eval_loss": 0.0355224609375,
      "eval_runtime": 2.852,
      "eval_samples_per_second": 113.606,
      "eval_steps_per_second": 14.376,
      "step": 9400
    },
    {
      "epoch": 215.91,
      "learning_rate": 5e-05,
      "loss": 0.0162,
      "step": 9500
    },
    {
      "epoch": 218.18,
      "eval_loss": 0.0350341796875,
      "eval_runtime": 2.8527,
      "eval_samples_per_second": 113.575,
      "eval_steps_per_second": 14.372,
      "step": 9600
    },
    {
      "epoch": 222.73,
      "eval_loss": 0.035247802734375,
      "eval_runtime": 2.8538,
      "eval_samples_per_second": 113.532,
      "eval_steps_per_second": 14.367,
      "step": 9800
    },
    {
      "epoch": 227.27,
      "learning_rate": 5e-05,
      "loss": 0.0152,
      "step": 10000
    },
    {
      "epoch": 227.27,
      "eval_loss": 0.0350341796875,
      "eval_runtime": 2.856,
      "eval_samples_per_second": 113.444,
      "eval_steps_per_second": 14.356,
      "step": 10000
    },
    {
      "epoch": 231.82,
      "eval_loss": 0.0350341796875,
      "eval_runtime": 2.8472,
      "eval_samples_per_second": 113.795,
      "eval_steps_per_second": 14.4,
      "step": 10200
    },
    {
      "epoch": 236.36,
      "eval_loss": 0.035400390625,
      "eval_runtime": 2.8527,
      "eval_samples_per_second": 113.575,
      "eval_steps_per_second": 14.372,
      "step": 10400
    },
    {
      "epoch": 238.64,
      "learning_rate": 5e-05,
      "loss": 0.0145,
      "step": 10500
    },
    {
      "epoch": 240.91,
      "eval_loss": 0.035125732421875,
      "eval_runtime": 2.8574,
      "eval_samples_per_second": 113.392,
      "eval_steps_per_second": 14.349,
      "step": 10600
    },
    {
      "epoch": 245.45,
      "eval_loss": 0.03546142578125,
      "eval_runtime": 2.8478,
      "eval_samples_per_second": 113.773,
      "eval_steps_per_second": 14.397,
      "step": 10800
    },
    {
      "epoch": 250.0,
      "learning_rate": 5e-05,
      "loss": 0.0137,
      "step": 11000
    },
    {
      "epoch": 250.0,
      "eval_loss": 0.035552978515625,
      "eval_runtime": 2.8538,
      "eval_samples_per_second": 113.533,
      "eval_steps_per_second": 14.367,
      "step": 11000
    },
    {
      "epoch": 254.55,
      "eval_loss": 0.035736083984375,
      "eval_runtime": 2.848,
      "eval_samples_per_second": 113.764,
      "eval_steps_per_second": 14.396,
      "step": 11200
    },
    {
      "epoch": 259.09,
      "eval_loss": 0.03582763671875,
      "eval_runtime": 2.8521,
      "eval_samples_per_second": 113.601,
      "eval_steps_per_second": 14.375,
      "step": 11400
    },
    {
      "epoch": 261.36,
      "learning_rate": 5e-05,
      "loss": 0.013,
      "step": 11500
    },
    {
      "epoch": 263.64,
      "eval_loss": 0.03546142578125,
      "eval_runtime": 2.8549,
      "eval_samples_per_second": 113.489,
      "eval_steps_per_second": 14.361,
      "step": 11600
    },
    {
      "epoch": 268.18,
      "eval_loss": 0.035919189453125,
      "eval_runtime": 2.8533,
      "eval_samples_per_second": 113.553,
      "eval_steps_per_second": 14.369,
      "step": 11800
    },
    {
      "epoch": 272.73,
      "learning_rate": 5e-05,
      "loss": 0.0127,
      "step": 12000
    },
    {
      "epoch": 272.73,
      "eval_loss": 0.0361328125,
      "eval_runtime": 2.8539,
      "eval_samples_per_second": 113.529,
      "eval_steps_per_second": 14.366,
      "step": 12000
    },
    {
      "epoch": 277.27,
      "eval_loss": 0.03619384765625,
      "eval_runtime": 2.8463,
      "eval_samples_per_second": 113.832,
      "eval_steps_per_second": 14.405,
      "step": 12200
    },
    {
      "epoch": 281.82,
      "eval_loss": 0.0361328125,
      "eval_runtime": 2.8549,
      "eval_samples_per_second": 113.491,
      "eval_steps_per_second": 14.362,
      "step": 12400
    },
    {
      "epoch": 284.09,
      "learning_rate": 5e-05,
      "loss": 0.0118,
      "step": 12500
    },
    {
      "epoch": 286.36,
      "eval_loss": 0.035736083984375,
      "eval_runtime": 2.8573,
      "eval_samples_per_second": 113.395,
      "eval_steps_per_second": 14.349,
      "step": 12600
    },
    {
      "epoch": 290.91,
      "eval_loss": 0.036529541015625,
      "eval_runtime": 2.8571,
      "eval_samples_per_second": 113.402,
      "eval_steps_per_second": 14.35,
      "step": 12800
    },
    {
      "epoch": 295.45,
      "learning_rate": 5e-05,
      "loss": 0.0114,
      "step": 13000
    },
    {
      "epoch": 295.45,
      "eval_loss": 0.0372314453125,
      "eval_runtime": 2.8553,
      "eval_samples_per_second": 113.473,
      "eval_steps_per_second": 14.359,
      "step": 13000
    },
    {
      "epoch": 300.0,
      "eval_loss": 0.037384033203125,
      "eval_runtime": 2.8451,
      "eval_samples_per_second": 113.881,
      "eval_steps_per_second": 14.411,
      "step": 13200
    },
    {
      "epoch": 304.55,
      "eval_loss": 0.0396728515625,
      "eval_runtime": 2.8561,
      "eval_samples_per_second": 113.44,
      "eval_steps_per_second": 14.355,
      "step": 13400
    },
    {
      "epoch": 306.82,
      "learning_rate": 5e-05,
      "loss": 0.0106,
      "step": 13500
    },
    {
      "epoch": 309.09,
      "eval_loss": 0.036529541015625,
      "eval_runtime": 2.8552,
      "eval_samples_per_second": 113.478,
      "eval_steps_per_second": 14.36,
      "step": 13600
    },
    {
      "epoch": 313.64,
      "eval_loss": 0.036346435546875,
      "eval_runtime": 2.8561,
      "eval_samples_per_second": 113.442,
      "eval_steps_per_second": 14.355,
      "step": 13800
    },
    {
      "epoch": 318.18,
      "learning_rate": 5e-05,
      "loss": 0.01,
      "step": 14000
    },
    {
      "epoch": 318.18,
      "eval_loss": 0.037811279296875,
      "eval_runtime": 2.8552,
      "eval_samples_per_second": 113.477,
      "eval_steps_per_second": 14.36,
      "step": 14000
    },
    {
      "epoch": 322.73,
      "eval_loss": 0.036865234375,
      "eval_runtime": 2.8537,
      "eval_samples_per_second": 113.537,
      "eval_steps_per_second": 14.367,
      "step": 14200
    },
    {
      "epoch": 327.27,
      "eval_loss": 0.037261962890625,
      "eval_runtime": 2.8546,
      "eval_samples_per_second": 113.503,
      "eval_steps_per_second": 14.363,
      "step": 14400
    },
    {
      "epoch": 329.55,
      "learning_rate": 5e-05,
      "loss": 0.0096,
      "step": 14500
    },
    {
      "epoch": 331.82,
      "eval_loss": 0.036590576171875,
      "eval_runtime": 2.8583,
      "eval_samples_per_second": 113.355,
      "eval_steps_per_second": 14.344,
      "step": 14600
    },
    {
      "epoch": 336.36,
      "eval_loss": 0.037109375,
      "eval_runtime": 2.8515,
      "eval_samples_per_second": 113.625,
      "eval_steps_per_second": 14.378,
      "step": 14800
    },
    {
      "epoch": 340.91,
      "learning_rate": 5e-05,
      "loss": 0.0093,
      "step": 15000
    },
    {
      "epoch": 340.91,
      "eval_loss": 0.037567138671875,
      "eval_runtime": 2.8557,
      "eval_samples_per_second": 113.459,
      "eval_steps_per_second": 14.357,
      "step": 15000
    },
    {
      "epoch": 345.45,
      "eval_loss": 0.03717041015625,
      "eval_runtime": 2.8575,
      "eval_samples_per_second": 113.384,
      "eval_steps_per_second": 14.348,
      "step": 15200
    },
    {
      "epoch": 350.0,
      "eval_loss": 0.03729248046875,
      "eval_runtime": 2.8457,
      "eval_samples_per_second": 113.856,
      "eval_steps_per_second": 14.408,
      "step": 15400
    },
    {
      "epoch": 352.27,
      "learning_rate": 5e-05,
      "loss": 0.0087,
      "step": 15500
    },
    {
      "epoch": 354.55,
      "eval_loss": 0.037506103515625,
      "eval_runtime": 2.8601,
      "eval_samples_per_second": 113.284,
      "eval_steps_per_second": 14.335,
      "step": 15600
    },
    {
      "epoch": 359.09,
      "eval_loss": 0.037445068359375,
      "eval_runtime": 2.8482,
      "eval_samples_per_second": 113.755,
      "eval_steps_per_second": 14.395,
      "step": 15800
    },
    {
      "epoch": 363.64,
      "learning_rate": 5e-05,
      "loss": 0.0086,
      "step": 16000
    },
    {
      "epoch": 363.64,
      "eval_loss": 0.037841796875,
      "eval_runtime": 2.8578,
      "eval_samples_per_second": 113.373,
      "eval_steps_per_second": 14.347,
      "step": 16000
    },
    {
      "epoch": 368.18,
      "eval_loss": 0.03704833984375,
      "eval_runtime": 2.8552,
      "eval_samples_per_second": 113.477,
      "eval_steps_per_second": 14.36,
      "step": 16200
    },
    {
      "epoch": 372.73,
      "eval_loss": 0.0380859375,
      "eval_runtime": 2.851,
      "eval_samples_per_second": 113.644,
      "eval_steps_per_second": 14.381,
      "step": 16400
    },
    {
      "epoch": 375.0,
      "learning_rate": 5e-05,
      "loss": 0.0084,
      "step": 16500
    },
    {
      "epoch": 377.27,
      "eval_loss": 0.037200927734375,
      "eval_runtime": 2.851,
      "eval_samples_per_second": 113.644,
      "eval_steps_per_second": 14.381,
      "step": 16600
    },
    {
      "epoch": 381.82,
      "eval_loss": 0.03680419921875,
      "eval_runtime": 2.8529,
      "eval_samples_per_second": 113.567,
      "eval_steps_per_second": 14.371,
      "step": 16800
    },
    {
      "epoch": 386.36,
      "learning_rate": 5e-05,
      "loss": 0.0078,
      "step": 17000
    },
    {
      "epoch": 386.36,
      "eval_loss": 0.036956787109375,
      "eval_runtime": 2.8565,
      "eval_samples_per_second": 113.425,
      "eval_steps_per_second": 14.353,
      "step": 17000
    },
    {
      "epoch": 390.91,
      "eval_loss": 0.03759765625,
      "eval_runtime": 2.8479,
      "eval_samples_per_second": 113.768,
      "eval_steps_per_second": 14.397,
      "step": 17200
    },
    {
      "epoch": 395.45,
      "eval_loss": 0.038238525390625,
      "eval_runtime": 2.8538,
      "eval_samples_per_second": 113.534,
      "eval_steps_per_second": 14.367,
      "step": 17400
    },
    {
      "epoch": 397.73,
      "learning_rate": 5e-05,
      "loss": 0.0076,
      "step": 17500
    },
    {
      "epoch": 400.0,
      "eval_loss": 0.03778076171875,
      "eval_runtime": 2.8549,
      "eval_samples_per_second": 113.488,
      "eval_steps_per_second": 14.361,
      "step": 17600
    },
    {
      "epoch": 404.55,
      "eval_loss": 0.038482666015625,
      "eval_runtime": 2.8533,
      "eval_samples_per_second": 113.554,
      "eval_steps_per_second": 14.37,
      "step": 17800
    },
    {
      "epoch": 409.09,
      "learning_rate": 5e-05,
      "loss": 0.0073,
      "step": 18000
    },
    {
      "epoch": 409.09,
      "eval_loss": 0.037872314453125,
      "eval_runtime": 2.8518,
      "eval_samples_per_second": 113.614,
      "eval_steps_per_second": 14.377,
      "step": 18000
    },
    {
      "epoch": 413.64,
      "eval_loss": 0.038787841796875,
      "eval_runtime": 2.8524,
      "eval_samples_per_second": 113.588,
      "eval_steps_per_second": 14.374,
      "step": 18200
    },
    {
      "epoch": 418.18,
      "eval_loss": 0.0379638671875,
      "eval_runtime": 2.848,
      "eval_samples_per_second": 113.766,
      "eval_steps_per_second": 14.396,
      "step": 18400
    },
    {
      "epoch": 420.45,
      "learning_rate": 5e-05,
      "loss": 0.0072,
      "step": 18500
    },
    {
      "epoch": 422.73,
      "eval_loss": 0.0380859375,
      "eval_runtime": 2.8569,
      "eval_samples_per_second": 113.409,
      "eval_steps_per_second": 14.351,
      "step": 18600
    },
    {
      "epoch": 427.27,
      "eval_loss": 0.037506103515625,
      "eval_runtime": 2.8521,
      "eval_samples_per_second": 113.599,
      "eval_steps_per_second": 14.375,
      "step": 18800
    },
    {
      "epoch": 431.82,
      "learning_rate": 5e-05,
      "loss": 0.0072,
      "step": 19000
    },
    {
      "epoch": 431.82,
      "eval_loss": 0.038818359375,
      "eval_runtime": 2.8529,
      "eval_samples_per_second": 113.569,
      "eval_steps_per_second": 14.371,
      "step": 19000
    },
    {
      "epoch": 436.36,
      "eval_loss": 0.0380859375,
      "eval_runtime": 2.8546,
      "eval_samples_per_second": 113.501,
      "eval_steps_per_second": 14.363,
      "step": 19200
    },
    {
      "epoch": 440.91,
      "eval_loss": 0.037322998046875,
      "eval_runtime": 2.8547,
      "eval_samples_per_second": 113.496,
      "eval_steps_per_second": 14.362,
      "step": 19400
    },
    {
      "epoch": 443.18,
      "learning_rate": 5e-05,
      "loss": 0.0066,
      "step": 19500
    },
    {
      "epoch": 445.45,
      "eval_loss": 0.037567138671875,
      "eval_runtime": 2.8553,
      "eval_samples_per_second": 113.474,
      "eval_steps_per_second": 14.359,
      "step": 19600
    },
    {
      "epoch": 450.0,
      "eval_loss": 0.038421630859375,
      "eval_runtime": 2.8536,
      "eval_samples_per_second": 113.543,
      "eval_steps_per_second": 14.368,
      "step": 19800
    },
    {
      "epoch": 454.55,
      "learning_rate": 5e-05,
      "loss": 0.0066,
      "step": 20000
    },
    {
      "epoch": 454.55,
      "eval_loss": 0.0384521484375,
      "eval_runtime": 2.8565,
      "eval_samples_per_second": 113.424,
      "eval_steps_per_second": 14.353,
      "step": 20000
    },
    {
      "epoch": 459.09,
      "eval_loss": 0.039764404296875,
      "eval_runtime": 2.858,
      "eval_samples_per_second": 113.365,
      "eval_steps_per_second": 14.346,
      "step": 20200
    },
    {
      "epoch": 463.64,
      "eval_loss": 0.0384521484375,
      "eval_runtime": 2.8472,
      "eval_samples_per_second": 113.796,
      "eval_steps_per_second": 14.4,
      "step": 20400
    },
    {
      "epoch": 465.91,
      "learning_rate": 5e-05,
      "loss": 0.0064,
      "step": 20500
    },
    {
      "epoch": 468.18,
      "eval_loss": 0.038665771484375,
      "eval_runtime": 2.8586,
      "eval_samples_per_second": 113.343,
      "eval_steps_per_second": 14.343,
      "step": 20600
    },
    {
      "epoch": 472.73,
      "eval_loss": 0.038970947265625,
      "eval_runtime": 2.8462,
      "eval_samples_per_second": 113.837,
      "eval_steps_per_second": 14.405,
      "step": 20800
    },
    {
      "epoch": 477.27,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 21000
    },
    {
      "epoch": 477.27,
      "eval_loss": 0.037750244140625,
      "eval_runtime": 2.8576,
      "eval_samples_per_second": 113.383,
      "eval_steps_per_second": 14.348,
      "step": 21000
    },
    {
      "epoch": 481.82,
      "eval_loss": 0.039154052734375,
      "eval_runtime": 2.8571,
      "eval_samples_per_second": 113.4,
      "eval_steps_per_second": 14.35,
      "step": 21200
    },
    {
      "epoch": 486.36,
      "eval_loss": 0.037994384765625,
      "eval_runtime": 2.8502,
      "eval_samples_per_second": 113.674,
      "eval_steps_per_second": 14.385,
      "step": 21400
    },
    {
      "epoch": 488.64,
      "learning_rate": 5e-05,
      "loss": 0.006,
      "step": 21500
    },
    {
      "epoch": 490.91,
      "eval_loss": 0.0377197265625,
      "eval_runtime": 2.8583,
      "eval_samples_per_second": 113.356,
      "eval_steps_per_second": 14.344,
      "step": 21600
    },
    {
      "epoch": 495.45,
      "eval_loss": 0.03857421875,
      "eval_runtime": 2.8515,
      "eval_samples_per_second": 113.623,
      "eval_steps_per_second": 14.378,
      "step": 21800
    },
    {
      "epoch": 500.0,
      "learning_rate": 5e-05,
      "loss": 0.006,
      "step": 22000
    },
    {
      "epoch": 500.0,
      "eval_loss": 0.038116455078125,
      "eval_runtime": 2.8565,
      "eval_samples_per_second": 113.426,
      "eval_steps_per_second": 14.353,
      "step": 22000
    },
    {
      "epoch": 500.0,
      "step": 22000,
      "total_flos": 9.1976456208384e+16,
      "train_loss": 0.09131514180790294,
      "train_runtime": 19622.0994,
      "train_samples_per_second": 17.939,
      "train_steps_per_second": 1.121
    }
  ],
  "max_steps": 22000,
  "num_train_epochs": 500,
  "total_flos": 9.1976456208384e+16,
  "trial_name": null,
  "trial_params": null
}