{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 35.2112676056338,
  "eval_steps": 500,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.20246478873239437,
      "eval_loss": 1.9784369468688965,
      "eval_runtime": 23.712,
      "eval_samples_per_second": 23.954,
      "eval_steps_per_second": 1.518,
      "step": 142
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.35563380281690143,
      "eval_loss": 1.7192655801773071,
      "eval_runtime": 23.7419,
      "eval_samples_per_second": 23.924,
      "eval_steps_per_second": 1.516,
      "step": 284
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.49295774647887325,
      "eval_loss": 1.470097303390503,
      "eval_runtime": 23.9735,
      "eval_samples_per_second": 23.693,
      "eval_steps_per_second": 1.502,
      "step": 426
    },
    {
      "epoch": 3.52112676056338,
      "grad_norm": 13.082801818847656,
      "learning_rate": 1.6478873239436623e-05,
      "loss": 1.7619,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6320422535211268,
      "eval_loss": 1.233864188194275,
      "eval_runtime": 23.6712,
      "eval_samples_per_second": 23.995,
      "eval_steps_per_second": 1.521,
      "step": 568
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6091549295774648,
      "eval_loss": 1.0945910215377808,
      "eval_runtime": 23.687,
      "eval_samples_per_second": 23.979,
      "eval_steps_per_second": 1.52,
      "step": 710
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7429577464788732,
      "eval_loss": 0.8935854434967041,
      "eval_runtime": 24.0091,
      "eval_samples_per_second": 23.658,
      "eval_steps_per_second": 1.499,
      "step": 852
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7535211267605634,
      "eval_loss": 0.8685159683227539,
      "eval_runtime": 23.6401,
      "eval_samples_per_second": 24.027,
      "eval_steps_per_second": 1.523,
      "step": 994
    },
    {
      "epoch": 7.042253521126761,
      "grad_norm": 7.731432914733887,
      "learning_rate": 1.2957746478873242e-05,
      "loss": 1.056,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7816901408450704,
      "eval_loss": 0.7374693155288696,
      "eval_runtime": 23.6523,
      "eval_samples_per_second": 24.015,
      "eval_steps_per_second": 1.522,
      "step": 1136
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7834507042253521,
      "eval_loss": 0.6636029481887817,
      "eval_runtime": 23.6027,
      "eval_samples_per_second": 24.065,
      "eval_steps_per_second": 1.525,
      "step": 1278
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.823943661971831,
      "eval_loss": 0.5628218650817871,
      "eval_runtime": 23.5285,
      "eval_samples_per_second": 24.141,
      "eval_steps_per_second": 1.53,
      "step": 1420
    },
    {
      "epoch": 10.56338028169014,
      "grad_norm": 4.4197211265563965,
      "learning_rate": 9.43661971830986e-06,
      "loss": 0.6356,
      "step": 1500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8116197183098591,
      "eval_loss": 0.5596714019775391,
      "eval_runtime": 23.7086,
      "eval_samples_per_second": 23.958,
      "eval_steps_per_second": 1.518,
      "step": 1562
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7764084507042254,
      "eval_loss": 0.6482901573181152,
      "eval_runtime": 23.7013,
      "eval_samples_per_second": 23.965,
      "eval_steps_per_second": 1.519,
      "step": 1704
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.823943661971831,
      "eval_loss": 0.5288650989532471,
      "eval_runtime": 23.642,
      "eval_samples_per_second": 24.025,
      "eval_steps_per_second": 1.523,
      "step": 1846
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8397887323943662,
      "eval_loss": 0.49578651785850525,
      "eval_runtime": 23.6946,
      "eval_samples_per_second": 23.972,
      "eval_steps_per_second": 1.519,
      "step": 1988
    },
    {
      "epoch": 14.084507042253522,
      "grad_norm": 9.863320350646973,
      "learning_rate": 5.915492957746479e-06,
      "loss": 0.434,
      "step": 2000
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.8221830985915493,
      "eval_loss": 0.5109295845031738,
      "eval_runtime": 23.7207,
      "eval_samples_per_second": 23.945,
      "eval_steps_per_second": 1.518,
      "step": 2130
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8257042253521126,
      "eval_loss": 0.46026095747947693,
      "eval_runtime": 23.6246,
      "eval_samples_per_second": 24.043,
      "eval_steps_per_second": 1.524,
      "step": 2272
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.8362676056338029,
      "eval_loss": 0.48158177733421326,
      "eval_runtime": 24.012,
      "eval_samples_per_second": 23.655,
      "eval_steps_per_second": 1.499,
      "step": 2414
    },
    {
      "epoch": 17.6056338028169,
      "grad_norm": 13.80960464477539,
      "learning_rate": 2.3943661971830984e-06,
      "loss": 0.3225,
      "step": 2500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9049295774647887,
      "eval_loss": 0.26005274057388306,
      "eval_runtime": 23.6863,
      "eval_samples_per_second": 23.98,
      "eval_steps_per_second": 1.52,
      "step": 2556
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.8961267605633803,
      "eval_loss": 0.24130958318710327,
      "eval_runtime": 23.7497,
      "eval_samples_per_second": 23.916,
      "eval_steps_per_second": 1.516,
      "step": 2698
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9014084507042254,
      "eval_loss": 0.24745188653469086,
      "eval_runtime": 23.7635,
      "eval_samples_per_second": 23.902,
      "eval_steps_per_second": 1.515,
      "step": 2840
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.9084507042253521,
      "eval_loss": 0.19828735291957855,
      "eval_runtime": 23.7046,
      "eval_samples_per_second": 23.962,
      "eval_steps_per_second": 1.519,
      "step": 2982
    },
    {
      "epoch": 21.12676056338028,
      "grad_norm": 4.427192211151123,
      "learning_rate": 9.43661971830986e-06,
      "loss": 0.3404,
      "step": 3000
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.903169014084507,
      "eval_loss": 0.21990956366062164,
      "eval_runtime": 23.6099,
      "eval_samples_per_second": 24.058,
      "eval_steps_per_second": 1.525,
      "step": 3124
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.9049295774647887,
      "eval_loss": 0.22014880180358887,
      "eval_runtime": 23.6019,
      "eval_samples_per_second": 24.066,
      "eval_steps_per_second": 1.525,
      "step": 3266
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.897887323943662,
      "eval_loss": 0.2570597231388092,
      "eval_runtime": 23.7096,
      "eval_samples_per_second": 23.957,
      "eval_steps_per_second": 1.518,
      "step": 3408
    },
    {
      "epoch": 24.647887323943664,
      "grad_norm": 16.97844886779785,
      "learning_rate": 7.67605633802817e-06,
      "loss": 0.265,
      "step": 3500
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.897887323943662,
      "eval_loss": 0.24254758656024933,
      "eval_runtime": 23.4896,
      "eval_samples_per_second": 24.181,
      "eval_steps_per_second": 1.533,
      "step": 3550
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.9119718309859155,
      "eval_loss": 0.24259935319423676,
      "eval_runtime": 23.5966,
      "eval_samples_per_second": 24.071,
      "eval_steps_per_second": 1.526,
      "step": 3692
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.8908450704225352,
      "eval_loss": 0.278877854347229,
      "eval_runtime": 23.6986,
      "eval_samples_per_second": 23.968,
      "eval_steps_per_second": 1.519,
      "step": 3834
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.897887323943662,
      "eval_loss": 0.2662661075592041,
      "eval_runtime": 23.6515,
      "eval_samples_per_second": 24.015,
      "eval_steps_per_second": 1.522,
      "step": 3976
    },
    {
      "epoch": 28.169014084507044,
      "grad_norm": 10.978034973144531,
      "learning_rate": 5.915492957746479e-06,
      "loss": 0.2065,
      "step": 4000
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.9049295774647887,
      "eval_loss": 0.24609780311584473,
      "eval_runtime": 23.6764,
      "eval_samples_per_second": 23.99,
      "eval_steps_per_second": 1.52,
      "step": 4118
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.9066901408450704,
      "eval_loss": 0.21895237267017365,
      "eval_runtime": 23.7437,
      "eval_samples_per_second": 23.922,
      "eval_steps_per_second": 1.516,
      "step": 4260
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.8926056338028169,
      "eval_loss": 0.24163208901882172,
      "eval_runtime": 23.5335,
      "eval_samples_per_second": 24.136,
      "eval_steps_per_second": 1.53,
      "step": 4402
    },
    {
      "epoch": 31.690140845070424,
      "grad_norm": 0.09197159856557846,
      "learning_rate": 4.154929577464789e-06,
      "loss": 0.1812,
      "step": 4500
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8961267605633803,
      "eval_loss": 0.2705635726451874,
      "eval_runtime": 23.6938,
      "eval_samples_per_second": 23.972,
      "eval_steps_per_second": 1.519,
      "step": 4544
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.9049295774647887,
      "eval_loss": 0.225112184882164,
      "eval_runtime": 23.7202,
      "eval_samples_per_second": 23.946,
      "eval_steps_per_second": 1.518,
      "step": 4686
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.9084507042253521,
      "eval_loss": 0.22959274053573608,
      "eval_runtime": 23.6809,
      "eval_samples_per_second": 23.986,
      "eval_steps_per_second": 1.52,
      "step": 4828
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.9049295774647887,
      "eval_loss": 0.22617577016353607,
      "eval_runtime": 23.6903,
      "eval_samples_per_second": 23.976,
      "eval_steps_per_second": 1.52,
      "step": 4970
    },
    {
      "epoch": 35.2112676056338,
      "grad_norm": 0.8059929013252258,
      "learning_rate": 2.3943661971830984e-06,
      "loss": 0.1701,
      "step": 5000
    }
  ],
  "logging_steps": 500,
  "max_steps": 5680,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "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": 3.071488373035011e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}