{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 111,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13513513513513514,
      "grad_norm": 18.339345376913933,
      "learning_rate": 5e-07,
      "loss": 1.7475,
      "step": 5
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 12.203185616101074,
      "learning_rate": 1e-06,
      "loss": 1.5992,
      "step": 10
    },
    {
      "epoch": 0.40540540540540543,
      "grad_norm": 7.556141195569309,
      "learning_rate": 9.939652198703783e-07,
      "loss": 1.299,
      "step": 15
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 4.135337196427636,
      "learning_rate": 9.76006553766365e-07,
      "loss": 1.1371,
      "step": 20
    },
    {
      "epoch": 0.6756756756756757,
      "grad_norm": 3.8408629474742417,
      "learning_rate": 9.465575080933957e-07,
      "loss": 1.064,
      "step": 25
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 3.5434703501005735,
      "learning_rate": 9.063289569141251e-07,
      "loss": 1.013,
      "step": 30
    },
    {
      "epoch": 0.9459459459459459,
      "grad_norm": 3.468410387637921,
      "learning_rate": 8.562919820737535e-07,
      "loss": 0.9957,
      "step": 35
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 3.5955588251338915,
      "learning_rate": 7.97654432138333e-07,
      "loss": 0.9521,
      "step": 40
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 3.727284147136237,
      "learning_rate": 7.318317659926636e-07,
      "loss": 0.9176,
      "step": 45
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 3.94980461123267,
      "learning_rate": 6.604128849076838e-07,
      "loss": 0.8959,
      "step": 50
    },
    {
      "epoch": 1.3513513513513513,
      "eval_loss": 0.9550710320472717,
      "eval_runtime": 19.211,
      "eval_samples_per_second": 54.656,
      "eval_steps_per_second": 0.885,
      "step": 50
    },
    {
      "epoch": 1.4864864864864864,
      "grad_norm": 3.470038603538472,
      "learning_rate": 5.851217778611993e-07,
      "loss": 0.8854,
      "step": 55
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 3.6321186413064686,
      "learning_rate": 5.077759059601755e-07,
      "loss": 0.8564,
      "step": 60
    },
    {
      "epoch": 1.7567567567567568,
      "grad_norm": 3.4878718381078953,
      "learning_rate": 4.302423305280385e-07,
      "loss": 0.8855,
      "step": 65
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 3.363284911698833,
      "learning_rate": 3.54392643886374e-07,
      "loss": 0.8806,
      "step": 70
    },
    {
      "epoch": 2.027027027027027,
      "grad_norm": 3.5632610689531363,
      "learning_rate": 2.8205779076231446e-07,
      "loss": 0.8547,
      "step": 75
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 3.4329782654828573,
      "learning_rate": 2.1498387089310865e-07,
      "loss": 0.8263,
      "step": 80
    },
    {
      "epoch": 2.2972972972972974,
      "grad_norm": 3.6353570223897598,
      "learning_rate": 1.5478998971412666e-07,
      "loss": 0.8104,
      "step": 85
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 3.507820404183458,
      "learning_rate": 1.0292917457762323e-07,
      "loss": 0.8137,
      "step": 90
    },
    {
      "epoch": 2.5675675675675675,
      "grad_norm": 3.568527072541467,
      "learning_rate": 6.065329995036572e-08,
      "loss": 0.8102,
      "step": 95
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 3.423831168506516,
      "learning_rate": 2.8982868265005454e-08,
      "loss": 0.8066,
      "step": 100
    },
    {
      "epoch": 2.7027027027027026,
      "eval_loss": 0.9287886619567871,
      "eval_runtime": 18.8032,
      "eval_samples_per_second": 55.841,
      "eval_steps_per_second": 0.904,
      "step": 100
    },
    {
      "epoch": 2.8378378378378377,
      "grad_norm": 3.6606624490508564,
      "learning_rate": 8.682375888868166e-09,
      "loss": 0.8146,
      "step": 105
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 3.4585615976656388,
      "learning_rate": 2.418588540059607e-10,
      "loss": 0.7984,
      "step": 110
    },
    {
      "epoch": 3.0,
      "step": 111,
      "total_flos": 654272366641152.0,
      "train_loss": 0.9826487050400124,
      "train_runtime": 1526.2759,
      "train_samples_per_second": 18.575,
      "train_steps_per_second": 0.073
    }
  ],
  "logging_steps": 5,
  "max_steps": 111,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 654272366641152.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}