{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 59.927739863508634,
  "eval_steps": 1866,
  "global_step": 18660,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.992773986350863,
      "grad_norm": 7.747014045715332,
      "learning_rate": 9.001071811361202e-06,
      "loss": 2.0902,
      "step": 1866
    },
    {
      "epoch": 5.992773986350863,
      "eval_accuracy": 0.06874315610171554,
      "eval_loss": 3.075465679168701,
      "eval_runtime": 32.3867,
      "eval_samples_per_second": 253.777,
      "eval_steps_per_second": 12.69,
      "step": 1866
    },
    {
      "epoch": 11.985547972701726,
      "grad_norm": 9.559131622314453,
      "learning_rate": 8.001607717041802e-06,
      "loss": 1.8876,
      "step": 3732
    },
    {
      "epoch": 11.985547972701726,
      "eval_accuracy": 0.06691811655919211,
      "eval_loss": 3.2739903926849365,
      "eval_runtime": 32.3147,
      "eval_samples_per_second": 254.342,
      "eval_steps_per_second": 12.719,
      "step": 3732
    },
    {
      "epoch": 17.97832195905259,
      "grad_norm": 4.176961421966553,
      "learning_rate": 7.002143622722402e-06,
      "loss": 1.763,
      "step": 5598
    },
    {
      "epoch": 17.97832195905259,
      "eval_accuracy": 0.06805369671898447,
      "eval_loss": 3.246858596801758,
      "eval_runtime": 32.3509,
      "eval_samples_per_second": 254.057,
      "eval_steps_per_second": 12.704,
      "step": 5598
    },
    {
      "epoch": 23.97109594540345,
      "grad_norm": 5.021180152893066,
      "learning_rate": 6.002679528403002e-06,
      "loss": 1.7048,
      "step": 7464
    },
    {
      "epoch": 23.97109594540345,
      "eval_accuracy": 0.06770896702761893,
      "eval_loss": 3.4241533279418945,
      "eval_runtime": 32.3523,
      "eval_samples_per_second": 254.047,
      "eval_steps_per_second": 12.704,
      "step": 7464
    },
    {
      "epoch": 29.963869931754317,
      "grad_norm": 1.899871587753296,
      "learning_rate": 5.003215434083602e-06,
      "loss": 1.6776,
      "step": 9330
    },
    {
      "epoch": 29.963869931754317,
      "eval_accuracy": 0.06735612604939774,
      "eval_loss": 3.498687505722046,
      "eval_runtime": 32.3207,
      "eval_samples_per_second": 254.295,
      "eval_steps_per_second": 12.716,
      "step": 9330
    },
    {
      "epoch": 35.95664391810518,
      "grad_norm": 2.520411968231201,
      "learning_rate": 4.003751339764202e-06,
      "loss": 1.6518,
      "step": 11196
    },
    {
      "epoch": 35.95664391810518,
      "eval_accuracy": 0.06750618485622745,
      "eval_loss": 3.563265323638916,
      "eval_runtime": 32.3151,
      "eval_samples_per_second": 254.34,
      "eval_steps_per_second": 12.719,
      "step": 11196
    },
    {
      "epoch": 41.94941790445604,
      "grad_norm": 3.6811821460723877,
      "learning_rate": 3.0037513397642017e-06,
      "loss": 1.6471,
      "step": 13062
    },
    {
      "epoch": 41.94941790445604,
      "eval_accuracy": 0.06683120991431005,
      "eval_loss": 3.6389496326446533,
      "eval_runtime": 32.2528,
      "eval_samples_per_second": 254.831,
      "eval_steps_per_second": 12.743,
      "step": 13062
    },
    {
      "epoch": 47.9421918908069,
      "grad_norm": 2.4686813354492188,
      "learning_rate": 2.004287245444802e-06,
      "loss": 1.6319,
      "step": 14928
    },
    {
      "epoch": 47.9421918908069,
      "eval_accuracy": 0.066324978707872,
      "eval_loss": 3.684262990951538,
      "eval_runtime": 32.2881,
      "eval_samples_per_second": 254.552,
      "eval_steps_per_second": 12.729,
      "step": 14928
    },
    {
      "epoch": 53.934965877157765,
      "grad_norm": 3.0867984294891357,
      "learning_rate": 1.0048231511254019e-06,
      "loss": 1.6325,
      "step": 16794
    },
    {
      "epoch": 53.934965877157765,
      "eval_accuracy": 0.06579605521082586,
      "eval_loss": 3.7068209648132324,
      "eval_runtime": 32.2239,
      "eval_samples_per_second": 255.059,
      "eval_steps_per_second": 12.754,
      "step": 16794
    },
    {
      "epoch": 59.927739863508634,
      "grad_norm": 1.7128721475601196,
      "learning_rate": 4.82315112540193e-09,
      "loss": 1.6255,
      "step": 18660
    },
    {
      "epoch": 59.927739863508634,
      "eval_accuracy": 0.0654215841343229,
      "eval_loss": 3.788184881210327,
      "eval_runtime": 33.1166,
      "eval_samples_per_second": 248.183,
      "eval_steps_per_second": 12.411,
      "step": 18660
    },
    {
      "epoch": 59.927739863508634,
      "step": 18660,
      "total_flos": 1.4754079994623488e+18,
      "train_loss": 1.731198005635383,
      "train_runtime": 33514.7147,
      "train_samples_per_second": 111.453,
      "train_steps_per_second": 0.557
    }
  ],
  "logging_steps": 1866,
  "max_steps": 18660,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 60,
  "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": 1.4754079994623488e+18,
  "train_batch_size": 25,
  "trial_name": null,
  "trial_params": null
}