{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 11.0,
  "eval_steps": 500,
  "global_step": 352,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15625,
      "grad_norm": 1.4347870349884033,
      "learning_rate": 0.0004,
      "loss": 2.5907,
      "step": 5
    },
    {
      "epoch": 0.3125,
      "grad_norm": 1.8193588256835938,
      "learning_rate": 0.0009000000000000001,
      "loss": 2.0401,
      "step": 10
    },
    {
      "epoch": 0.46875,
      "grad_norm": 3.8659496307373047,
      "learning_rate": 0.0009936507936507937,
      "loss": 1.7812,
      "step": 15
    },
    {
      "epoch": 0.625,
      "grad_norm": 1.8431330919265747,
      "learning_rate": 0.0009857142857142857,
      "loss": 1.7989,
      "step": 20
    },
    {
      "epoch": 0.78125,
      "grad_norm": 1.5119725465774536,
      "learning_rate": 0.0009777777777777777,
      "loss": 1.8825,
      "step": 25
    },
    {
      "epoch": 0.9375,
      "grad_norm": 2.630089044570923,
      "learning_rate": 0.0009698412698412698,
      "loss": 1.8248,
      "step": 30
    },
    {
      "epoch": 1.09375,
      "grad_norm": 1.824265956878662,
      "learning_rate": 0.0009619047619047619,
      "loss": 1.4467,
      "step": 35
    },
    {
      "epoch": 1.25,
      "grad_norm": 2.9590811729431152,
      "learning_rate": 0.000953968253968254,
      "loss": 1.2601,
      "step": 40
    },
    {
      "epoch": 1.40625,
      "grad_norm": 2.812793731689453,
      "learning_rate": 0.000946031746031746,
      "loss": 1.2122,
      "step": 45
    },
    {
      "epoch": 1.5625,
      "grad_norm": 3.0169219970703125,
      "learning_rate": 0.0009380952380952381,
      "loss": 1.2601,
      "step": 50
    },
    {
      "epoch": 1.71875,
      "grad_norm": 3.0815634727478027,
      "learning_rate": 0.0009301587301587302,
      "loss": 1.2289,
      "step": 55
    },
    {
      "epoch": 1.875,
      "grad_norm": 2.0147032737731934,
      "learning_rate": 0.0009222222222222223,
      "loss": 1.2768,
      "step": 60
    },
    {
      "epoch": 2.03125,
      "grad_norm": 1.5649322271347046,
      "learning_rate": 0.0009142857142857143,
      "loss": 1.0438,
      "step": 65
    },
    {
      "epoch": 2.1875,
      "grad_norm": 1.8557994365692139,
      "learning_rate": 0.0009063492063492064,
      "loss": 0.5735,
      "step": 70
    },
    {
      "epoch": 2.34375,
      "grad_norm": 2.2182095050811768,
      "learning_rate": 0.0008984126984126985,
      "loss": 0.6358,
      "step": 75
    },
    {
      "epoch": 2.5,
      "grad_norm": 3.01589298248291,
      "learning_rate": 0.0008904761904761904,
      "loss": 0.5855,
      "step": 80
    },
    {
      "epoch": 2.65625,
      "grad_norm": 5.799967288970947,
      "learning_rate": 0.0008825396825396825,
      "loss": 0.6252,
      "step": 85
    },
    {
      "epoch": 2.8125,
      "grad_norm": 2.3106298446655273,
      "learning_rate": 0.0008746031746031746,
      "loss": 0.5751,
      "step": 90
    },
    {
      "epoch": 2.96875,
      "grad_norm": 5.094876766204834,
      "learning_rate": 0.0008666666666666667,
      "loss": 0.5947,
      "step": 95
    },
    {
      "epoch": 3.125,
      "grad_norm": 2.225015878677368,
      "learning_rate": 0.0008587301587301587,
      "loss": 0.3564,
      "step": 100
    },
    {
      "epoch": 3.28125,
      "grad_norm": 2.587599515914917,
      "learning_rate": 0.0008507936507936508,
      "loss": 0.2237,
      "step": 105
    },
    {
      "epoch": 3.4375,
      "grad_norm": 1.6697884798049927,
      "learning_rate": 0.0008428571428571429,
      "loss": 0.3015,
      "step": 110
    },
    {
      "epoch": 3.59375,
      "grad_norm": 1.5882270336151123,
      "learning_rate": 0.000834920634920635,
      "loss": 0.3354,
      "step": 115
    },
    {
      "epoch": 3.75,
      "grad_norm": 2.361110210418701,
      "learning_rate": 0.000826984126984127,
      "loss": 0.3349,
      "step": 120
    },
    {
      "epoch": 3.90625,
      "grad_norm": 2.789064884185791,
      "learning_rate": 0.0008190476190476191,
      "loss": 0.3435,
      "step": 125
    },
    {
      "epoch": 4.0625,
      "grad_norm": 1.9216350317001343,
      "learning_rate": 0.0008111111111111111,
      "loss": 0.3043,
      "step": 130
    },
    {
      "epoch": 4.21875,
      "grad_norm": 2.5817618370056152,
      "learning_rate": 0.0008031746031746032,
      "loss": 0.1865,
      "step": 135
    },
    {
      "epoch": 4.375,
      "grad_norm": 1.6777920722961426,
      "learning_rate": 0.0007952380952380952,
      "loss": 0.1989,
      "step": 140
    },
    {
      "epoch": 4.53125,
      "grad_norm": 1.9233051538467407,
      "learning_rate": 0.0007873015873015873,
      "loss": 0.1912,
      "step": 145
    },
    {
      "epoch": 4.6875,
      "grad_norm": 1.441566824913025,
      "learning_rate": 0.0007793650793650794,
      "loss": 0.1941,
      "step": 150
    },
    {
      "epoch": 4.84375,
      "grad_norm": 1.7641961574554443,
      "learning_rate": 0.0007714285714285715,
      "loss": 0.1927,
      "step": 155
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.937122344970703,
      "learning_rate": 0.0007634920634920634,
      "loss": 0.2052,
      "step": 160
    },
    {
      "epoch": 5.15625,
      "grad_norm": 2.509890079498291,
      "learning_rate": 0.0007555555555555555,
      "loss": 0.1141,
      "step": 165
    },
    {
      "epoch": 5.3125,
      "grad_norm": 2.4012136459350586,
      "learning_rate": 0.0007476190476190476,
      "loss": 0.1059,
      "step": 170
    },
    {
      "epoch": 5.46875,
      "grad_norm": 1.8514283895492554,
      "learning_rate": 0.0007396825396825397,
      "loss": 0.1349,
      "step": 175
    },
    {
      "epoch": 5.625,
      "grad_norm": 3.880389451980591,
      "learning_rate": 0.0007317460317460317,
      "loss": 0.1334,
      "step": 180
    },
    {
      "epoch": 5.78125,
      "grad_norm": 1.4267611503601074,
      "learning_rate": 0.0007238095238095238,
      "loss": 0.1403,
      "step": 185
    },
    {
      "epoch": 5.9375,
      "grad_norm": 2.336212635040283,
      "learning_rate": 0.0007158730158730159,
      "loss": 0.1702,
      "step": 190
    },
    {
      "epoch": 6.09375,
      "grad_norm": 1.6925745010375977,
      "learning_rate": 0.000707936507936508,
      "loss": 0.1224,
      "step": 195
    },
    {
      "epoch": 6.25,
      "grad_norm": 2.5146541595458984,
      "learning_rate": 0.0007,
      "loss": 0.0903,
      "step": 200
    },
    {
      "epoch": 6.40625,
      "grad_norm": 1.3388704061508179,
      "learning_rate": 0.0006920634920634921,
      "loss": 0.0846,
      "step": 205
    },
    {
      "epoch": 6.5625,
      "grad_norm": 1.7252851724624634,
      "learning_rate": 0.0006841269841269842,
      "loss": 0.1077,
      "step": 210
    },
    {
      "epoch": 6.71875,
      "grad_norm": 1.524242639541626,
      "learning_rate": 0.0006761904761904763,
      "loss": 0.07,
      "step": 215
    },
    {
      "epoch": 6.875,
      "grad_norm": 2.5351924896240234,
      "learning_rate": 0.0006682539682539683,
      "loss": 0.1065,
      "step": 220
    },
    {
      "epoch": 7.03125,
      "grad_norm": 0.7822954654693604,
      "learning_rate": 0.0006603174603174604,
      "loss": 0.097,
      "step": 225
    },
    {
      "epoch": 7.1875,
      "grad_norm": 0.8932663202285767,
      "learning_rate": 0.0006523809523809525,
      "loss": 0.0564,
      "step": 230
    },
    {
      "epoch": 7.34375,
      "grad_norm": 2.202021360397339,
      "learning_rate": 0.0006444444444444444,
      "loss": 0.0899,
      "step": 235
    },
    {
      "epoch": 7.5,
      "grad_norm": 5.009730815887451,
      "learning_rate": 0.0006365079365079364,
      "loss": 0.0894,
      "step": 240
    },
    {
      "epoch": 7.65625,
      "grad_norm": 1.9714306592941284,
      "learning_rate": 0.0006285714285714285,
      "loss": 0.1092,
      "step": 245
    },
    {
      "epoch": 7.8125,
      "grad_norm": 1.9612032175064087,
      "learning_rate": 0.0006206349206349206,
      "loss": 0.0769,
      "step": 250
    },
    {
      "epoch": 7.96875,
      "grad_norm": 2.019763231277466,
      "learning_rate": 0.0006126984126984127,
      "loss": 0.066,
      "step": 255
    },
    {
      "epoch": 8.125,
      "grad_norm": 1.9064483642578125,
      "learning_rate": 0.0006047619047619047,
      "loss": 0.0798,
      "step": 260
    },
    {
      "epoch": 8.28125,
      "grad_norm": 1.2875497341156006,
      "learning_rate": 0.0005968253968253968,
      "loss": 0.0609,
      "step": 265
    },
    {
      "epoch": 8.4375,
      "grad_norm": 0.9999807476997375,
      "learning_rate": 0.0005888888888888889,
      "loss": 0.087,
      "step": 270
    },
    {
      "epoch": 8.59375,
      "grad_norm": 1.205248236656189,
      "learning_rate": 0.000580952380952381,
      "loss": 0.0791,
      "step": 275
    },
    {
      "epoch": 8.75,
      "grad_norm": 1.612770915031433,
      "learning_rate": 0.000573015873015873,
      "loss": 0.0768,
      "step": 280
    },
    {
      "epoch": 8.90625,
      "grad_norm": 1.8061549663543701,
      "learning_rate": 0.0005650793650793651,
      "loss": 0.0714,
      "step": 285
    },
    {
      "epoch": 9.0625,
      "grad_norm": 0.8511625528335571,
      "learning_rate": 0.0005571428571428572,
      "loss": 0.066,
      "step": 290
    },
    {
      "epoch": 9.21875,
      "grad_norm": 2.199141025543213,
      "learning_rate": 0.0005492063492063493,
      "loss": 0.0404,
      "step": 295
    },
    {
      "epoch": 9.375,
      "grad_norm": 1.1211730241775513,
      "learning_rate": 0.0005412698412698413,
      "loss": 0.0645,
      "step": 300
    },
    {
      "epoch": 9.53125,
      "grad_norm": 0.6761476993560791,
      "learning_rate": 0.0005333333333333334,
      "loss": 0.0731,
      "step": 305
    },
    {
      "epoch": 9.6875,
      "grad_norm": 0.9598909020423889,
      "learning_rate": 0.0005253968253968255,
      "loss": 0.053,
      "step": 310
    },
    {
      "epoch": 9.84375,
      "grad_norm": 1.074629545211792,
      "learning_rate": 0.0005174603174603176,
      "loss": 0.0606,
      "step": 315
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.9772604703903198,
      "learning_rate": 0.0005095238095238095,
      "loss": 0.0523,
      "step": 320
    },
    {
      "epoch": 10.15625,
      "grad_norm": 0.6214720010757446,
      "learning_rate": 0.0005015873015873016,
      "loss": 0.0401,
      "step": 325
    },
    {
      "epoch": 10.3125,
      "grad_norm": 1.0092564821243286,
      "learning_rate": 0.0004936507936507937,
      "loss": 0.0343,
      "step": 330
    },
    {
      "epoch": 10.46875,
      "grad_norm": 0.7020597457885742,
      "learning_rate": 0.0004857142857142857,
      "loss": 0.0285,
      "step": 335
    },
    {
      "epoch": 10.625,
      "grad_norm": 0.5726391077041626,
      "learning_rate": 0.0004777777777777778,
      "loss": 0.042,
      "step": 340
    },
    {
      "epoch": 10.78125,
      "grad_norm": 0.4384658634662628,
      "learning_rate": 0.00046984126984126986,
      "loss": 0.043,
      "step": 345
    },
    {
      "epoch": 10.9375,
      "grad_norm": 0.40464839339256287,
      "learning_rate": 0.00046190476190476195,
      "loss": 0.0344,
      "step": 350
    }
  ],
  "logging_steps": 5,
  "max_steps": 640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.217928864934707e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}