{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 908,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.9013997074961662,
      "epoch": 0.027559597629874606,
      "grad_norm": 0.04541015625,
      "learning_rate": 0.00017142857142857143,
      "loss": 2.106939239501953,
      "mean_token_accuracy": 0.5557776354253292,
      "num_tokens": 140963.0,
      "step": 25
    },
    {
      "entropy": 1.7775965774059295,
      "epoch": 0.05511919525974921,
      "grad_norm": 0.06103515625,
      "learning_rate": 0.0001997191078685512,
      "loss": 1.7464599609375,
      "mean_token_accuracy": 0.6121900254487991,
      "num_tokens": 284944.0,
      "step": 50
    },
    {
      "entropy": 1.6929514452815055,
      "epoch": 0.0826787928896238,
      "grad_norm": 0.091796875,
      "learning_rate": 0.00019865462348631944,
      "loss": 1.6511944580078124,
      "mean_token_accuracy": 0.6266605603694916,
      "num_tokens": 425136.0,
      "step": 75
    },
    {
      "entropy": 1.5028782638907432,
      "epoch": 0.11023839051949842,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.00019680482470453816,
      "loss": 1.4662600708007814,
      "mean_token_accuracy": 0.659810485392809,
      "num_tokens": 567966.0,
      "step": 100
    },
    {
      "entropy": 1.4264089419692754,
      "epoch": 0.13779798814937302,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.00019418443636395248,
      "loss": 1.4101972961425782,
      "mean_token_accuracy": 0.6815375398099422,
      "num_tokens": 706204.0,
      "step": 125
    },
    {
      "entropy": 1.1659171215072275,
      "epoch": 0.1653575857792476,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.00019081431738250814,
      "loss": 1.1057415008544922,
      "mean_token_accuracy": 0.7461765056848526,
      "num_tokens": 851807.0,
      "step": 150
    },
    {
      "entropy": 1.1612900688499213,
      "epoch": 0.19291718340912223,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.00018672129471337568,
      "loss": 1.1313108062744142,
      "mean_token_accuracy": 0.7447272512316704,
      "num_tokens": 998421.0,
      "step": 175
    },
    {
      "entropy": 1.0649858928844333,
      "epoch": 0.22047678103899684,
      "grad_norm": 0.11328125,
      "learning_rate": 0.00018193794979598656,
      "loss": 1.0393009185791016,
      "mean_token_accuracy": 0.763860610127449,
      "num_tokens": 1141761.0,
      "step": 200
    },
    {
      "entropy": 0.8765474074520171,
      "epoch": 0.24803637866887143,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.00017650235919998232,
      "loss": 0.8135224914550782,
      "mean_token_accuracy": 0.807485072016716,
      "num_tokens": 1286611.0,
      "step": 225
    },
    {
      "entropy": 0.8536889077164233,
      "epoch": 0.27559597629874605,
      "grad_norm": 0.083984375,
      "learning_rate": 0.00017045779152661297,
      "loss": 0.8201860809326171,
      "mean_token_accuracy": 0.8146287898719311,
      "num_tokens": 1428015.0,
      "step": 250
    },
    {
      "entropy": 0.8703145318105817,
      "epoch": 0.30315557392862064,
      "grad_norm": 0.09765625,
      "learning_rate": 0.00016385236298032183,
      "loss": 0.815809326171875,
      "mean_token_accuracy": 0.8103299827873707,
      "num_tokens": 1574815.0,
      "step": 275
    },
    {
      "entropy": 0.8991702864505351,
      "epoch": 0.3307151715584952,
      "grad_norm": 0.091796875,
      "learning_rate": 0.00015673865435224962,
      "loss": 0.8489224243164063,
      "mean_token_accuracy": 0.8056490634381771,
      "num_tokens": 1717330.0,
      "step": 300
    },
    {
      "entropy": 0.7706374152004719,
      "epoch": 0.35827476918836987,
      "grad_norm": 0.0732421875,
      "learning_rate": 0.0001491732924645604,
      "loss": 0.7274211883544922,
      "mean_token_accuracy": 0.8297103045880795,
      "num_tokens": 1859716.0,
      "step": 325
    },
    {
      "entropy": 0.7342993662878871,
      "epoch": 0.38583436681824446,
      "grad_norm": 0.0791015625,
      "learning_rate": 0.00014121649940739301,
      "loss": 0.7029119110107422,
      "mean_token_accuracy": 0.8427872517704964,
      "num_tokens": 2005201.0,
      "step": 350
    },
    {
      "entropy": 0.7452906335704028,
      "epoch": 0.41339396444811904,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.00013293161315661931,
      "loss": 0.7123774719238282,
      "mean_token_accuracy": 0.8400052092969418,
      "num_tokens": 2143419.0,
      "step": 375
    },
    {
      "entropy": 0.6938006362970919,
      "epoch": 0.4409535620779937,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.00012438458338840706,
      "loss": 0.6549488830566407,
      "mean_token_accuracy": 0.8493785254657269,
      "num_tokens": 2288125.0,
      "step": 400
    },
    {
      "entropy": 0.7443565137218684,
      "epoch": 0.4685131597078683,
      "grad_norm": 0.08203125,
      "learning_rate": 0.0001156434465040231,
      "loss": 0.6924254608154297,
      "mean_token_accuracy": 0.8412553237378597,
      "num_tokens": 2431404.0,
      "step": 425
    },
    {
      "entropy": 0.645295873908326,
      "epoch": 0.49607275733774286,
      "grad_norm": 0.055419921875,
      "learning_rate": 0.00010677778404380643,
      "loss": 0.6191695022583008,
      "mean_token_accuracy": 0.8578527195751667,
      "num_tokens": 2571370.0,
      "step": 450
    },
    {
      "entropy": 0.6608448463864625,
      "epoch": 0.5236323549676175,
      "grad_norm": 0.053955078125,
      "learning_rate": 9.78581688014642e-05,
      "loss": 0.6262045669555664,
      "mean_token_accuracy": 0.856816074848175,
      "num_tokens": 2714312.0,
      "step": 475
    },
    {
      "entropy": 0.5949629322066903,
      "epoch": 0.5511919525974921,
      "grad_norm": 0.08251953125,
      "learning_rate": 8.895560304775387e-05,
      "loss": 0.54037109375,
      "mean_token_accuracy": 0.8706019401550293,
      "num_tokens": 2850910.0,
      "step": 500
    },
    {
      "entropy": 0.7224912867229432,
      "epoch": 0.5787515502273667,
      "grad_norm": 0.053466796875,
      "learning_rate": 8.014095333542548e-05,
      "loss": 0.7142529296875,
      "mean_token_accuracy": 0.8413439232110977,
      "num_tokens": 2999359.0,
      "step": 525
    },
    {
      "entropy": 0.7125124373659492,
      "epoch": 0.6063111478572413,
      "grad_norm": 0.08740234375,
      "learning_rate": 7.148438638451144e-05,
      "loss": 0.6760162353515625,
      "mean_token_accuracy": 0.8475173614919186,
      "num_tokens": 3139525.0,
      "step": 550
    },
    {
      "entropy": 0.6205505744647235,
      "epoch": 0.6338707454871159,
      "grad_norm": 0.11669921875,
      "learning_rate": 6.305481053845261e-05,
      "loss": 0.6042692947387696,
      "mean_token_accuracy": 0.8661120194196701,
      "num_tokens": 3283709.0,
      "step": 575
    },
    {
      "entropy": 0.6542045682761818,
      "epoch": 0.6614303431169904,
      "grad_norm": 0.055419921875,
      "learning_rate": 5.4919327237202546e-05,
      "loss": 0.6398629379272461,
      "mean_token_accuracy": 0.8612794525921345,
      "num_tokens": 3427270.0,
      "step": 600
    },
    {
      "entropy": 0.5916300803422928,
      "epoch": 0.6889899407468651,
      "grad_norm": 0.09716796875,
      "learning_rate": 4.714269687371581e-05,
      "loss": 0.5418794250488281,
      "mean_token_accuracy": 0.8722360366582871,
      "num_tokens": 3568256.0,
      "step": 625
    },
    {
      "entropy": 0.5375593125354499,
      "epoch": 0.7165495383767397,
      "grad_norm": 0.10693359375,
      "learning_rate": 3.978682328573063e-05,
      "loss": 0.5057635116577148,
      "mean_token_accuracy": 0.8873542912304402,
      "num_tokens": 3709441.0,
      "step": 650
    },
    {
      "entropy": 0.6055101400800049,
      "epoch": 0.7441091360066143,
      "grad_norm": 0.0712890625,
      "learning_rate": 3.291026098641397e-05,
      "loss": 0.5837324905395508,
      "mean_token_accuracy": 0.8686715243756771,
      "num_tokens": 3858373.0,
      "step": 675
    },
    {
      "entropy": 0.509702676339075,
      "epoch": 0.7716687336364889,
      "grad_norm": 0.04150390625,
      "learning_rate": 2.6567749056431467e-05,
      "loss": 0.46534008026123047,
      "mean_token_accuracy": 0.8935040257871151,
      "num_tokens": 3998452.0,
      "step": 700
    },
    {
      "entropy": 0.6449116668663919,
      "epoch": 0.7992283312663635,
      "grad_norm": 0.08203125,
      "learning_rate": 2.0809775407772503e-05,
      "loss": 0.5881856155395507,
      "mean_token_accuracy": 0.8635844677686692,
      "num_tokens": 4139915.0,
      "step": 725
    },
    {
      "entropy": 0.575501447301358,
      "epoch": 0.8267879288962381,
      "grad_norm": 0.056396484375,
      "learning_rate": 1.5682174887895083e-05,
      "loss": 0.5590339279174805,
      "mean_token_accuracy": 0.8768933260440827,
      "num_tokens": 4282188.0,
      "step": 750
    },
    {
      "entropy": 0.5558508610818536,
      "epoch": 0.8543475265261127,
      "grad_norm": 0.07861328125,
      "learning_rate": 1.122576442337696e-05,
      "loss": 0.5243261337280274,
      "mean_token_accuracy": 0.8842477214336395,
      "num_tokens": 4427400.0,
      "step": 775
    },
    {
      "entropy": 0.688429104378447,
      "epoch": 0.8819071241559874,
      "grad_norm": 0.0810546875,
      "learning_rate": 7.476018107417371e-06,
      "loss": 0.623744010925293,
      "mean_token_accuracy": 0.8572955358028412,
      "num_tokens": 4567225.0,
      "step": 800
    },
    {
      "entropy": 0.5038102147076279,
      "epoch": 0.909466721785862,
      "grad_norm": 0.07470703125,
      "learning_rate": 4.462784817570331e-06,
      "loss": 0.47533946990966797,
      "mean_token_accuracy": 0.8936529903113842,
      "num_tokens": 4709595.0,
      "step": 825
    },
    {
      "entropy": 0.5760456078313291,
      "epoch": 0.9370263194157366,
      "grad_norm": 0.0751953125,
      "learning_rate": 2.2100506115399377e-06,
      "loss": 0.5174454879760743,
      "mean_token_accuracy": 0.8788142189383507,
      "num_tokens": 4853579.0,
      "step": 850
    },
    {
      "entropy": 0.5857030095718802,
      "epoch": 0.9645859170456111,
      "grad_norm": 0.10888671875,
      "learning_rate": 7.357477924239797e-07,
      "loss": 0.5400792694091797,
      "mean_token_accuracy": 0.8724570070207119,
      "num_tokens": 4994351.0,
      "step": 875
    },
    {
      "entropy": 0.47729114089161156,
      "epoch": 0.9921455146754857,
      "grad_norm": 0.0498046875,
      "learning_rate": 5.161216329209184e-08,
      "loss": 0.4375122833251953,
      "mean_token_accuracy": 0.9014435681700707,
      "num_tokens": 5136659.0,
      "step": 900
    }
  ],
  "logging_steps": 25,
  "max_steps": 908,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.209155120942981e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}