{
  "best_metric": 1.299881935119629,
  "best_model_checkpoint": "./outputs/checkpoint-4100",
  "epoch": 2.9876138433515482,
  "eval_steps": 100,
  "global_step": 4100,
  "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
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.5637,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.577666163444519,
      "eval_runtime": 133.1755,
      "eval_samples_per_second": 47.111,
      "eval_steps_per_second": 5.894,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.5403,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.563066005706787,
      "eval_runtime": 133.3045,
      "eval_samples_per_second": 47.065,
      "eval_steps_per_second": 5.889,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.5421,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.5478307008743286,
      "eval_runtime": 133.145,
      "eval_samples_per_second": 47.122,
      "eval_steps_per_second": 5.896,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.5198,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.531746506690979,
      "eval_runtime": 133.271,
      "eval_samples_per_second": 47.077,
      "eval_steps_per_second": 5.89,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.5204,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.5162949562072754,
      "eval_runtime": 133.1561,
      "eval_samples_per_second": 47.118,
      "eval_steps_per_second": 5.895,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.4905,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.5029550790786743,
      "eval_runtime": 133.1207,
      "eval_samples_per_second": 47.13,
      "eval_steps_per_second": 5.897,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.4783,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.4883637428283691,
      "eval_runtime": 133.2381,
      "eval_samples_per_second": 47.089,
      "eval_steps_per_second": 5.892,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.4709,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.4737261533737183,
      "eval_runtime": 133.3108,
      "eval_samples_per_second": 47.063,
      "eval_steps_per_second": 5.888,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.4331,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.462561011314392,
      "eval_runtime": 133.3104,
      "eval_samples_per_second": 47.063,
      "eval_steps_per_second": 5.889,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.4116,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.4483799934387207,
      "eval_runtime": 133.2589,
      "eval_samples_per_second": 47.081,
      "eval_steps_per_second": 5.891,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.4104,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.4345015287399292,
      "eval_runtime": 133.1546,
      "eval_samples_per_second": 47.118,
      "eval_steps_per_second": 5.895,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.3987,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.4243552684783936,
      "eval_runtime": 133.2552,
      "eval_samples_per_second": 47.083,
      "eval_steps_per_second": 5.891,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.3984,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.4116299152374268,
      "eval_runtime": 133.2306,
      "eval_samples_per_second": 47.091,
      "eval_steps_per_second": 5.892,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 1.3786,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.3974164724349976,
      "eval_runtime": 133.3674,
      "eval_samples_per_second": 47.043,
      "eval_steps_per_second": 5.886,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.3594,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.3871479034423828,
      "eval_runtime": 133.2189,
      "eval_samples_per_second": 47.095,
      "eval_steps_per_second": 5.893,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.3587,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.3761059045791626,
      "eval_runtime": 133.2457,
      "eval_samples_per_second": 47.086,
      "eval_steps_per_second": 5.891,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.3466,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.3624399900436401,
      "eval_runtime": 133.3002,
      "eval_samples_per_second": 47.067,
      "eval_steps_per_second": 5.889,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.3282,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.3485013246536255,
      "eval_runtime": 133.2962,
      "eval_samples_per_second": 47.068,
      "eval_steps_per_second": 5.889,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.3053,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.3402942419052124,
      "eval_runtime": 133.2508,
      "eval_samples_per_second": 47.084,
      "eval_steps_per_second": 5.891,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.312,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.3246259689331055,
      "eval_runtime": 133.2626,
      "eval_samples_per_second": 47.08,
      "eval_steps_per_second": 5.891,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 1.2977,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 1.314718246459961,
      "eval_runtime": 133.1685,
      "eval_samples_per_second": 47.113,
      "eval_steps_per_second": 5.895,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.2922,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.299881935119629,
      "eval_runtime": 133.2671,
      "eval_samples_per_second": 47.078,
      "eval_steps_per_second": 5.89,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 2.4410408101719245e+17,
  "trial_name": null,
  "trial_params": null
}