{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 1.1044,
      "step": 1
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019948693233918952,
      "loss": 1.0409,
      "step": 2
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019795299412524945,
      "loss": 0.9112,
      "step": 3
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00019541392564000488,
      "loss": 0.7554,
      "step": 4
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00019189578116202307,
      "loss": 0.6185,
      "step": 5
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00018743466161445823,
      "loss": 0.5743,
      "step": 6
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00018207634412072764,
      "loss": 0.5785,
      "step": 7
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001758758122692791,
      "loss": 0.436,
      "step": 8
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.00016889669190756868,
      "loss": 0.3596,
      "step": 9
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0001612105982547663,
      "loss": 0.3703,
      "step": 10
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00015289640103269625,
      "loss": 0.2485,
      "step": 11
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00014403941515576344,
      "loss": 0.2869,
      "step": 12
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00013473052528448201,
      "loss": 0.1574,
      "step": 13
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.00012506525322587207,
      "loss": 0.2161,
      "step": 14
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.00011514277775045768,
      "loss": 0.1179,
      "step": 15
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00010506491688387127,
      "loss": 0.1789,
      "step": 16
    },
    {
      "epoch": 4.25,
      "learning_rate": 9.493508311612874e-05,
      "loss": 0.0866,
      "step": 17
    },
    {
      "epoch": 4.5,
      "learning_rate": 8.485722224954237e-05,
      "loss": 0.0816,
      "step": 18
    },
    {
      "epoch": 4.75,
      "learning_rate": 7.493474677412794e-05,
      "loss": 0.0849,
      "step": 19
    },
    {
      "epoch": 5.0,
      "learning_rate": 6.526947471551798e-05,
      "loss": 0.0903,
      "step": 20
    },
    {
      "epoch": 5.25,
      "learning_rate": 5.596058484423656e-05,
      "loss": 0.0444,
      "step": 21
    },
    {
      "epoch": 5.5,
      "learning_rate": 4.710359896730379e-05,
      "loss": 0.0389,
      "step": 22
    },
    {
      "epoch": 5.75,
      "learning_rate": 3.878940174523371e-05,
      "loss": 0.0437,
      "step": 23
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.110330809243134e-05,
      "loss": 0.055,
      "step": 24
    },
    {
      "epoch": 6.25,
      "learning_rate": 2.4124187730720917e-05,
      "loss": 0.0347,
      "step": 25
    },
    {
      "epoch": 6.5,
      "learning_rate": 1.7923655879272393e-05,
      "loss": 0.0194,
      "step": 26
    },
    {
      "epoch": 6.75,
      "learning_rate": 1.2565338385541792e-05,
      "loss": 0.0299,
      "step": 27
    },
    {
      "epoch": 7.0,
      "learning_rate": 8.10421883797694e-06,
      "loss": 0.0329,
      "step": 28
    },
    {
      "epoch": 7.25,
      "learning_rate": 4.586074359995119e-06,
      "loss": 0.0189,
      "step": 29
    },
    {
      "epoch": 7.5,
      "learning_rate": 2.0470058747505516e-06,
      "loss": 0.0244,
      "step": 30
    },
    {
      "epoch": 7.75,
      "learning_rate": 5.130676608104845e-07,
      "loss": 0.0259,
      "step": 31
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0,
      "loss": 0.0235,
      "step": 32
    },
    {
      "epoch": 8.0,
      "step": 32,
      "total_flos": 1564227624960.0,
      "train_loss": 0.27154927648371086,
      "train_runtime": 170.0181,
      "train_samples_per_second": 2.823,
      "train_steps_per_second": 0.188
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 50000,
  "total_flos": 1564227624960.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}