{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 1.0905472040176392,
      "learning_rate": 0.0007920000000000001,
      "loss": 7.247091674804688,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 2.1956381797790527,
      "learning_rate": 0.001592,
      "loss": 5.962919311523438,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 0.7685134410858154,
      "learning_rate": 0.002392,
      "loss": 5.536254272460938,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 0.6038330793380737,
      "learning_rate": 0.003192,
      "loss": 5.27086181640625,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.38015660643577576,
      "learning_rate": 0.003992,
      "loss": 5.034593505859375,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 0.25242117047309875,
      "learning_rate": 0.004792,
      "loss": 4.90749267578125,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 0.2624945342540741,
      "learning_rate": 0.005592000000000001,
      "loss": 4.80042236328125,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.17131584882736206,
      "learning_rate": 0.006392,
      "loss": 4.716185607910156,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 0.2284715175628662,
      "learning_rate": 0.007192,
      "loss": 4.674341430664063,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.24492475390434265,
      "learning_rate": 0.007992,
      "loss": 4.623594665527344,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.24955772994129158,
      "eval_loss": 4.598090648651123,
      "eval_runtime": 4.6622,
      "eval_samples_per_second": 214.491,
      "eval_steps_per_second": 1.716,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 0.16292935609817505,
      "learning_rate": 0.008792,
      "loss": 4.553927307128906,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.15358929336071014,
      "learning_rate": 0.009592000000000002,
      "loss": 4.5277734375,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 0.20846614241600037,
      "learning_rate": 0.010391999999999998,
      "loss": 4.486420288085937,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 0.16785487532615662,
      "learning_rate": 0.011192,
      "loss": 4.471618041992188,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.17252753674983978,
      "learning_rate": 0.011992000000000001,
      "loss": 4.429487915039062,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.42827746272087097,
      "learning_rate": 0.012792,
      "loss": 4.423707885742187,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 0.18584232032299042,
      "learning_rate": 0.013592,
      "loss": 4.405695190429688,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 0.1389407366514206,
      "learning_rate": 0.014392,
      "loss": 4.381403503417968,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 0.17971371114253998,
      "learning_rate": 0.015192,
      "loss": 4.340931091308594,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1317117065191269,
      "learning_rate": 0.015992,
      "loss": 4.357503356933594,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.27022504892367905,
      "eval_loss": 4.341521739959717,
      "eval_runtime": 4.6212,
      "eval_samples_per_second": 216.396,
      "eval_steps_per_second": 1.731,
      "step": 2000
    },
    {
      "epoch": 0.042,
      "grad_norm": 0.3252929747104645,
      "learning_rate": 0.016792,
      "loss": 4.336614685058594,
      "step": 2100
    },
    {
      "epoch": 0.044,
      "grad_norm": 0.3191538155078888,
      "learning_rate": 0.017592,
      "loss": 4.3118289184570315,
      "step": 2200
    },
    {
      "epoch": 0.046,
      "grad_norm": 0.11609218269586563,
      "learning_rate": 0.018392,
      "loss": 4.328140563964844,
      "step": 2300
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.09269429743289948,
      "learning_rate": 0.019192,
      "loss": 4.2995166015625,
      "step": 2400
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.11243760585784912,
      "learning_rate": 0.019992000000000003,
      "loss": 4.306504821777343,
      "step": 2500
    },
    {
      "epoch": 0.052,
      "grad_norm": 0.1890493929386139,
      "learning_rate": 0.019999785636239033,
      "loss": 4.267638549804688,
      "step": 2600
    },
    {
      "epoch": 0.054,
      "grad_norm": 0.09589467197656631,
      "learning_rate": 0.01999913387133122,
      "loss": 4.277190551757813,
      "step": 2700
    },
    {
      "epoch": 0.056,
      "grad_norm": 0.21187792718410492,
      "learning_rate": 0.019998044711915176,
      "loss": 4.244146118164062,
      "step": 2800
    },
    {
      "epoch": 0.058,
      "grad_norm": 0.18817682564258575,
      "learning_rate": 0.019996518205634257,
      "loss": 4.229488220214844,
      "step": 2900
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.1112157553434372,
      "learning_rate": 0.0199945544192628,
      "loss": 4.231150512695312,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.2808140900195695,
      "eval_loss": 4.229600429534912,
      "eval_runtime": 4.8217,
      "eval_samples_per_second": 207.397,
      "eval_steps_per_second": 1.659,
      "step": 3000
    },
    {
      "epoch": 0.062,
      "grad_norm": 0.3390547037124634,
      "learning_rate": 0.019992153438703173,
      "loss": 4.223145141601562,
      "step": 3100
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.7005836367607117,
      "learning_rate": 0.019989315368982054,
      "loss": 4.203560791015625,
      "step": 3200
    },
    {
      "epoch": 0.066,
      "grad_norm": 0.1793125420808792,
      "learning_rate": 0.019986040334245798,
      "loss": 4.1834033203125,
      "step": 3300
    },
    {
      "epoch": 0.068,
      "grad_norm": 0.11384831368923187,
      "learning_rate": 0.019982328477755038,
      "loss": 4.1909048461914065,
      "step": 3400
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6594609022140503,
      "learning_rate": 0.019978179961878402,
      "loss": 4.169663696289063,
      "step": 3500
    },
    {
      "epoch": 0.072,
      "grad_norm": 0.3771114647388458,
      "learning_rate": 0.01997359496808541,
      "loss": 4.174803161621094,
      "step": 3600
    },
    {
      "epoch": 0.074,
      "grad_norm": 0.31003686785697937,
      "learning_rate": 0.01996857369693855,
      "loss": 4.1531503295898435,
      "step": 3700
    },
    {
      "epoch": 0.076,
      "grad_norm": 0.11918768286705017,
      "learning_rate": 0.019963116368084486,
      "loss": 4.163074951171875,
      "step": 3800
    },
    {
      "epoch": 0.078,
      "grad_norm": 0.5589029788970947,
      "learning_rate": 0.01995722322024446,
      "loss": 4.147608032226563,
      "step": 3900
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.15616266429424286,
      "learning_rate": 0.01995089451120385,
      "loss": 4.122293701171875,
      "step": 4000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.287454011741683,
      "eval_loss": 4.152503967285156,
      "eval_runtime": 4.8042,
      "eval_samples_per_second": 208.152,
      "eval_steps_per_second": 1.665,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}