{
  "best_metric": 1.5946060419082642,
  "best_model_checkpoint": "./outputs/checkpoint-1900",
  "epoch": 1.3846994535519126,
  "eval_steps": 100,
  "global_step": 1900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.315,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.160651922225952,
      "eval_runtime": 144.1197,
      "eval_samples_per_second": 43.533,
      "eval_steps_per_second": 5.447,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.1091,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.0636215209960938,
      "eval_runtime": 143.9553,
      "eval_samples_per_second": 43.583,
      "eval_steps_per_second": 5.453,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.0344,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.0027897357940674,
      "eval_runtime": 144.1134,
      "eval_samples_per_second": 43.535,
      "eval_steps_per_second": 5.447,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.9806,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.9537049531936646,
      "eval_runtime": 144.1171,
      "eval_samples_per_second": 43.534,
      "eval_steps_per_second": 5.447,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.9262,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.9101046323776245,
      "eval_runtime": 143.9719,
      "eval_samples_per_second": 43.578,
      "eval_steps_per_second": 5.452,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.8971,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.8801190853118896,
      "eval_runtime": 143.9079,
      "eval_samples_per_second": 43.597,
      "eval_steps_per_second": 5.455,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.8565,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.8463259935379028,
      "eval_runtime": 143.8032,
      "eval_samples_per_second": 43.629,
      "eval_steps_per_second": 5.459,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.8327,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.8173062801361084,
      "eval_runtime": 143.9045,
      "eval_samples_per_second": 43.598,
      "eval_steps_per_second": 5.455,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.7952,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.7927803993225098,
      "eval_runtime": 143.8647,
      "eval_samples_per_second": 43.61,
      "eval_steps_per_second": 5.457,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.7792,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.7664344310760498,
      "eval_runtime": 143.816,
      "eval_samples_per_second": 43.625,
      "eval_steps_per_second": 5.458,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.7588,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.7422090768814087,
      "eval_runtime": 143.8688,
      "eval_samples_per_second": 43.609,
      "eval_steps_per_second": 5.456,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.7204,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.72101891040802,
      "eval_runtime": 143.8632,
      "eval_samples_per_second": 43.611,
      "eval_steps_per_second": 5.457,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.7282,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.6993989944458008,
      "eval_runtime": 143.9159,
      "eval_samples_per_second": 43.595,
      "eval_steps_per_second": 5.455,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.7077,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.6939107179641724,
      "eval_runtime": 132.967,
      "eval_samples_per_second": 47.185,
      "eval_steps_per_second": 5.904,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.6649,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.6719050407409668,
      "eval_runtime": 132.9346,
      "eval_samples_per_second": 47.196,
      "eval_steps_per_second": 5.905,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.6428,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.649336576461792,
      "eval_runtime": 132.9323,
      "eval_samples_per_second": 47.197,
      "eval_steps_per_second": 5.905,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.6362,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.629374623298645,
      "eval_runtime": 132.9442,
      "eval_samples_per_second": 47.193,
      "eval_steps_per_second": 5.905,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.6043,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.6119710206985474,
      "eval_runtime": 132.947,
      "eval_samples_per_second": 47.192,
      "eval_steps_per_second": 5.905,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.57,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.5946060419082642,
      "eval_runtime": 133.2291,
      "eval_samples_per_second": 47.092,
      "eval_steps_per_second": 5.892,
      "step": 1900
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.131308551955497e+17,
  "trial_name": null,
  "trial_params": null
}