{
  "best_metric": 1.1990991830825806,
  "best_model_checkpoint": "./outputs/checkpoint-4100",
  "epoch": 2.987249544626594,
  "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": 2.182,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.0125839710235596,
      "eval_runtime": 144.1418,
      "eval_samples_per_second": 43.527,
      "eval_steps_per_second": 5.446,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.9626,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.9195271730422974,
      "eval_runtime": 144.2253,
      "eval_samples_per_second": 43.501,
      "eval_steps_per_second": 5.443,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.8881,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.858821988105774,
      "eval_runtime": 144.0313,
      "eval_samples_per_second": 43.56,
      "eval_steps_per_second": 5.45,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.8367,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.8125503063201904,
      "eval_runtime": 143.9861,
      "eval_samples_per_second": 43.574,
      "eval_steps_per_second": 5.452,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.785,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.7720445394515991,
      "eval_runtime": 144.0061,
      "eval_samples_per_second": 43.568,
      "eval_steps_per_second": 5.451,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.7562,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.7424511909484863,
      "eval_runtime": 144.0245,
      "eval_samples_per_second": 43.562,
      "eval_steps_per_second": 5.45,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.7176,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.7112771272659302,
      "eval_runtime": 144.2204,
      "eval_samples_per_second": 43.503,
      "eval_steps_per_second": 5.443,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.6969,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.6823335886001587,
      "eval_runtime": 144.0134,
      "eval_samples_per_second": 43.565,
      "eval_steps_per_second": 5.451,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.6619,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.6567240953445435,
      "eval_runtime": 144.0655,
      "eval_samples_per_second": 43.55,
      "eval_steps_per_second": 5.449,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.6469,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.6352009773254395,
      "eval_runtime": 144.0664,
      "eval_samples_per_second": 43.549,
      "eval_steps_per_second": 5.449,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.6461,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.6240431070327759,
      "eval_runtime": 293.5989,
      "eval_samples_per_second": 21.369,
      "eval_steps_per_second": 2.674,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.5992,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.5974311828613281,
      "eval_runtime": 291.7,
      "eval_samples_per_second": 21.508,
      "eval_steps_per_second": 2.691,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.6021,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.5751127004623413,
      "eval_runtime": 289.9524,
      "eval_samples_per_second": 21.638,
      "eval_steps_per_second": 2.707,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.5538,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.5539450645446777,
      "eval_runtime": 287.8748,
      "eval_samples_per_second": 21.794,
      "eval_steps_per_second": 2.727,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.5249,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.5348094701766968,
      "eval_runtime": 287.891,
      "eval_samples_per_second": 21.793,
      "eval_steps_per_second": 2.727,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.506,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.515953540802002,
      "eval_runtime": 289.836,
      "eval_samples_per_second": 21.647,
      "eval_steps_per_second": 2.708,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.5042,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.4988901615142822,
      "eval_runtime": 291.5471,
      "eval_samples_per_second": 21.52,
      "eval_steps_per_second": 2.693,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.4762,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.4844294786453247,
      "eval_runtime": 293.6668,
      "eval_samples_per_second": 21.364,
      "eval_steps_per_second": 2.673,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.4652,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.4694663286209106,
      "eval_runtime": 295.5867,
      "eval_samples_per_second": 21.226,
      "eval_steps_per_second": 2.656,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.4481,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.4534634351730347,
      "eval_runtime": 296.7451,
      "eval_samples_per_second": 21.143,
      "eval_steps_per_second": 2.645,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.4335,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.4383305311203003,
      "eval_runtime": 294.659,
      "eval_samples_per_second": 21.292,
      "eval_steps_per_second": 2.664,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.4075,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.4232139587402344,
      "eval_runtime": 292.737,
      "eval_samples_per_second": 21.432,
      "eval_steps_per_second": 2.682,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.4059,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.411597490310669,
      "eval_runtime": 290.6759,
      "eval_samples_per_second": 21.584,
      "eval_steps_per_second": 2.701,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4055,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.3974188566207886,
      "eval_runtime": 288.6807,
      "eval_samples_per_second": 21.733,
      "eval_steps_per_second": 2.719,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3772,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.384261965751648,
      "eval_runtime": 287.1594,
      "eval_samples_per_second": 21.848,
      "eval_steps_per_second": 2.734,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3669,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.3715009689331055,
      "eval_runtime": 288.6305,
      "eval_samples_per_second": 21.737,
      "eval_steps_per_second": 2.72,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3548,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.3575737476348877,
      "eval_runtime": 290.5389,
      "eval_samples_per_second": 21.594,
      "eval_steps_per_second": 2.702,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.3266,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3451271057128906,
      "eval_runtime": 292.4987,
      "eval_samples_per_second": 21.45,
      "eval_steps_per_second": 2.684,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.3004,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.333436369895935,
      "eval_runtime": 294.3881,
      "eval_samples_per_second": 21.312,
      "eval_steps_per_second": 2.667,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.3009,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.3215913772583008,
      "eval_runtime": 296.2165,
      "eval_samples_per_second": 21.18,
      "eval_steps_per_second": 2.65,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.2822,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.311138391494751,
      "eval_runtime": 295.6634,
      "eval_samples_per_second": 21.22,
      "eval_steps_per_second": 2.655,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.2846,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.3013139963150024,
      "eval_runtime": 293.7639,
      "eval_samples_per_second": 21.357,
      "eval_steps_per_second": 2.672,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 1.2674,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.2875950336456299,
      "eval_runtime": 291.8744,
      "eval_samples_per_second": 21.496,
      "eval_steps_per_second": 2.69,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.2485,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.2778161764144897,
      "eval_runtime": 289.8138,
      "eval_samples_per_second": 21.648,
      "eval_steps_per_second": 2.709,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.2499,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.2662204504013062,
      "eval_runtime": 288.0522,
      "eval_samples_per_second": 21.781,
      "eval_steps_per_second": 2.725,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.2363,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.2541649341583252,
      "eval_runtime": 287.6646,
      "eval_samples_per_second": 21.81,
      "eval_steps_per_second": 2.729,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.22,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.2425366640090942,
      "eval_runtime": 289.5395,
      "eval_samples_per_second": 21.669,
      "eval_steps_per_second": 2.711,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.1977,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.2353510856628418,
      "eval_runtime": 291.5958,
      "eval_samples_per_second": 21.516,
      "eval_steps_per_second": 2.692,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.2087,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.2214804887771606,
      "eval_runtime": 293.3693,
      "eval_samples_per_second": 21.386,
      "eval_steps_per_second": 2.676,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 1.1912,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 1.2116272449493408,
      "eval_runtime": 295.3584,
      "eval_samples_per_second": 21.242,
      "eval_steps_per_second": 2.658,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.1882,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.1990991830825806,
      "eval_runtime": 133.6346,
      "eval_samples_per_second": 46.949,
      "eval_steps_per_second": 5.874,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 2.444085005899776e+17,
  "trial_name": null,
  "trial_params": null
}