{
  "best_metric": 1.0850836038589478,
  "best_model_checkpoint": "./outputs/checkpoint-4100",
  "epoch": 2.9879781420765026,
  "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": 1.7654,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 1.6328892707824707,
      "eval_runtime": 419.2151,
      "eval_samples_per_second": 14.966,
      "eval_steps_per_second": 1.873,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.6081,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.5872766971588135,
      "eval_runtime": 418.8615,
      "eval_samples_per_second": 14.979,
      "eval_steps_per_second": 1.874,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.5756,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.556390643119812,
      "eval_runtime": 418.8265,
      "eval_samples_per_second": 14.98,
      "eval_steps_per_second": 1.874,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.5444,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.5300668478012085,
      "eval_runtime": 418.8771,
      "eval_samples_per_second": 14.978,
      "eval_steps_per_second": 1.874,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.5114,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.5074748992919922,
      "eval_runtime": 418.9574,
      "eval_samples_per_second": 14.975,
      "eval_steps_per_second": 1.874,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.4948,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.4872732162475586,
      "eval_runtime": 419.0224,
      "eval_samples_per_second": 14.973,
      "eval_steps_per_second": 1.873,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.4723,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.4686368703842163,
      "eval_runtime": 418.747,
      "eval_samples_per_second": 14.983,
      "eval_steps_per_second": 1.875,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.4628,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.450691819190979,
      "eval_runtime": 418.912,
      "eval_samples_per_second": 14.977,
      "eval_steps_per_second": 1.874,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.4332,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.432775616645813,
      "eval_runtime": 418.8955,
      "eval_samples_per_second": 14.977,
      "eval_steps_per_second": 1.874,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.4271,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.4167886972427368,
      "eval_runtime": 418.6904,
      "eval_samples_per_second": 14.985,
      "eval_steps_per_second": 1.875,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.4129,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.402354121208191,
      "eval_runtime": 418.7796,
      "eval_samples_per_second": 14.982,
      "eval_steps_per_second": 1.874,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.3853,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.3865565061569214,
      "eval_runtime": 418.6288,
      "eval_samples_per_second": 14.987,
      "eval_steps_per_second": 1.875,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.394,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.372268795967102,
      "eval_runtime": 418.7001,
      "eval_samples_per_second": 14.984,
      "eval_steps_per_second": 1.875,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.356,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.3581469058990479,
      "eval_runtime": 418.565,
      "eval_samples_per_second": 14.989,
      "eval_steps_per_second": 1.875,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.3321,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.3450849056243896,
      "eval_runtime": 418.8093,
      "eval_samples_per_second": 14.981,
      "eval_steps_per_second": 1.874,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.3214,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.3320348262786865,
      "eval_runtime": 418.8397,
      "eval_samples_per_second": 14.979,
      "eval_steps_per_second": 1.874,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.3215,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.3184651136398315,
      "eval_runtime": 418.7321,
      "eval_samples_per_second": 14.983,
      "eval_steps_per_second": 1.875,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.2973,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.306998610496521,
      "eval_runtime": 418.7252,
      "eval_samples_per_second": 14.984,
      "eval_steps_per_second": 1.875,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.2893,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.293902039527893,
      "eval_runtime": 449.0285,
      "eval_samples_per_second": 13.972,
      "eval_steps_per_second": 1.748,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.2763,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.2827019691467285,
      "eval_runtime": 418.4577,
      "eval_samples_per_second": 14.993,
      "eval_steps_per_second": 1.876,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.2665,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.2712739706039429,
      "eval_runtime": 418.4388,
      "eval_samples_per_second": 14.994,
      "eval_steps_per_second": 1.876,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.2452,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.2593406438827515,
      "eval_runtime": 418.3773,
      "eval_samples_per_second": 14.996,
      "eval_steps_per_second": 1.876,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.2438,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.2501240968704224,
      "eval_runtime": 418.3937,
      "eval_samples_per_second": 14.995,
      "eval_steps_per_second": 1.876,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.2416,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.237477421760559,
      "eval_runtime": 418.5428,
      "eval_samples_per_second": 14.99,
      "eval_steps_per_second": 1.876,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.2224,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.2273696660995483,
      "eval_runtime": 418.5797,
      "eval_samples_per_second": 14.989,
      "eval_steps_per_second": 1.875,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.2106,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.2160065174102783,
      "eval_runtime": 418.6873,
      "eval_samples_per_second": 14.985,
      "eval_steps_per_second": 1.875,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.2036,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.2056827545166016,
      "eval_runtime": 418.5394,
      "eval_samples_per_second": 14.99,
      "eval_steps_per_second": 1.876,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.1786,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.196009635925293,
      "eval_runtime": 418.5266,
      "eval_samples_per_second": 14.991,
      "eval_steps_per_second": 1.876,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.1541,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.1857470273971558,
      "eval_runtime": 418.5437,
      "eval_samples_per_second": 14.99,
      "eval_steps_per_second": 1.876,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.1565,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.175557017326355,
      "eval_runtime": 418.72,
      "eval_samples_per_second": 14.984,
      "eval_steps_per_second": 1.875,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.1799,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.1915525197982788,
      "eval_runtime": 341.6989,
      "eval_samples_per_second": 18.361,
      "eval_steps_per_second": 2.297,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.1452,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.1769670248031616,
      "eval_runtime": 341.7617,
      "eval_samples_per_second": 18.358,
      "eval_steps_per_second": 2.297,
      "step": 3200
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0002,
      "loss": 1.1401,
      "step": 3300
    },
    {
      "epoch": 2.41,
      "eval_loss": 1.1650751829147339,
      "eval_runtime": 342.0204,
      "eval_samples_per_second": 18.344,
      "eval_steps_per_second": 2.295,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.1249,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.1538294553756714,
      "eval_runtime": 341.9176,
      "eval_samples_per_second": 18.349,
      "eval_steps_per_second": 2.296,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.1227,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.1433058977127075,
      "eval_runtime": 342.1437,
      "eval_samples_per_second": 18.337,
      "eval_steps_per_second": 2.294,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.1161,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.1336331367492676,
      "eval_runtime": 342.2427,
      "eval_samples_per_second": 18.332,
      "eval_steps_per_second": 2.294,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.1109,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.123780369758606,
      "eval_runtime": 342.8733,
      "eval_samples_per_second": 18.298,
      "eval_steps_per_second": 2.289,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.0954,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.114593267440796,
      "eval_runtime": 342.239,
      "eval_samples_per_second": 18.332,
      "eval_steps_per_second": 2.294,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.0901,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.1041266918182373,
      "eval_runtime": 342.4571,
      "eval_samples_per_second": 18.321,
      "eval_steps_per_second": 2.292,
      "step": 3900
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0002,
      "loss": 1.0798,
      "step": 4000
    },
    {
      "epoch": 2.92,
      "eval_loss": 1.0935524702072144,
      "eval_runtime": 342.5842,
      "eval_samples_per_second": 18.314,
      "eval_steps_per_second": 2.291,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.0688,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.0850836038589478,
      "eval_runtime": 396.9254,
      "eval_samples_per_second": 15.806,
      "eval_steps_per_second": 1.978,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.3292919651387863e+18,
  "trial_name": null,
  "trial_params": null
}