{
  "best_metric": 1.2128502130508423,
  "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": 2.1918,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.024646282196045,
      "eval_runtime": 143.5031,
      "eval_samples_per_second": 43.72,
      "eval_steps_per_second": 5.47,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.9739,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.9310301542282104,
      "eval_runtime": 143.4865,
      "eval_samples_per_second": 43.725,
      "eval_steps_per_second": 5.471,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.8997,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.869322419166565,
      "eval_runtime": 143.5199,
      "eval_samples_per_second": 43.715,
      "eval_steps_per_second": 5.47,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.8476,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.8241251707077026,
      "eval_runtime": 143.5226,
      "eval_samples_per_second": 43.714,
      "eval_steps_per_second": 5.47,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.7962,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.7821052074432373,
      "eval_runtime": 143.5402,
      "eval_samples_per_second": 43.709,
      "eval_steps_per_second": 5.469,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.7679,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.7543551921844482,
      "eval_runtime": 143.714,
      "eval_samples_per_second": 43.656,
      "eval_steps_per_second": 5.462,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.7293,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.7226487398147583,
      "eval_runtime": 143.4831,
      "eval_samples_per_second": 43.726,
      "eval_steps_per_second": 5.471,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.7081,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.6933537721633911,
      "eval_runtime": 143.5765,
      "eval_samples_per_second": 43.698,
      "eval_steps_per_second": 5.467,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.6726,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.6686474084854126,
      "eval_runtime": 143.516,
      "eval_samples_per_second": 43.716,
      "eval_steps_per_second": 5.47,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.6577,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.6451665163040161,
      "eval_runtime": 146.451,
      "eval_samples_per_second": 42.84,
      "eval_steps_per_second": 5.36,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.6377,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.623993992805481,
      "eval_runtime": 143.5112,
      "eval_samples_per_second": 43.718,
      "eval_steps_per_second": 5.47,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.6028,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.6054009199142456,
      "eval_runtime": 143.561,
      "eval_samples_per_second": 43.703,
      "eval_steps_per_second": 5.468,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.6101,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.5822986364364624,
      "eval_runtime": 143.5287,
      "eval_samples_per_second": 43.713,
      "eval_steps_per_second": 5.469,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.5635,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.5618135929107666,
      "eval_runtime": 143.5801,
      "eval_samples_per_second": 43.697,
      "eval_steps_per_second": 5.467,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.5353,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.546177864074707,
      "eval_runtime": 143.5634,
      "eval_samples_per_second": 43.702,
      "eval_steps_per_second": 5.468,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.5167,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.5285366773605347,
      "eval_runtime": 143.5631,
      "eval_samples_per_second": 43.702,
      "eval_steps_per_second": 5.468,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.5143,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.511216640472412,
      "eval_runtime": 143.4634,
      "eval_samples_per_second": 43.732,
      "eval_steps_per_second": 5.472,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.4878,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.496885061264038,
      "eval_runtime": 143.4809,
      "eval_samples_per_second": 43.727,
      "eval_steps_per_second": 5.471,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.4773,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.4805316925048828,
      "eval_runtime": 143.5169,
      "eval_samples_per_second": 43.716,
      "eval_steps_per_second": 5.47,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.4603,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.4675525426864624,
      "eval_runtime": 143.6108,
      "eval_samples_per_second": 43.688,
      "eval_steps_per_second": 5.466,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.446,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.4523862600326538,
      "eval_runtime": 143.5584,
      "eval_samples_per_second": 43.703,
      "eval_steps_per_second": 5.468,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.4205,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.4386054277420044,
      "eval_runtime": 143.497,
      "eval_samples_per_second": 43.722,
      "eval_steps_per_second": 5.47,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.4198,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.4249650239944458,
      "eval_runtime": 143.5274,
      "eval_samples_per_second": 43.713,
      "eval_steps_per_second": 5.469,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4191,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.4121719598770142,
      "eval_runtime": 143.5251,
      "eval_samples_per_second": 43.714,
      "eval_steps_per_second": 5.469,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3902,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.3990552425384521,
      "eval_runtime": 143.4243,
      "eval_samples_per_second": 43.744,
      "eval_steps_per_second": 5.473,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3802,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.3864269256591797,
      "eval_runtime": 144.1474,
      "eval_samples_per_second": 43.525,
      "eval_steps_per_second": 5.446,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3683,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.3723174333572388,
      "eval_runtime": 143.4406,
      "eval_samples_per_second": 43.739,
      "eval_steps_per_second": 5.473,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.34,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3610557317733765,
      "eval_runtime": 143.5173,
      "eval_samples_per_second": 43.716,
      "eval_steps_per_second": 5.47,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.3145,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.347936749458313,
      "eval_runtime": 143.5505,
      "eval_samples_per_second": 43.706,
      "eval_steps_per_second": 5.468,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.3152,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.3374146223068237,
      "eval_runtime": 143.4898,
      "eval_samples_per_second": 43.724,
      "eval_steps_per_second": 5.471,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.2956,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.325629472732544,
      "eval_runtime": 143.5978,
      "eval_samples_per_second": 43.691,
      "eval_steps_per_second": 5.467,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.2991,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.3144720792770386,
      "eval_runtime": 143.5018,
      "eval_samples_per_second": 43.721,
      "eval_steps_per_second": 5.47,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 1.2803,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.3016444444656372,
      "eval_runtime": 143.5185,
      "eval_samples_per_second": 43.716,
      "eval_steps_per_second": 5.47,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.2618,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.2920358180999756,
      "eval_runtime": 143.6048,
      "eval_samples_per_second": 43.689,
      "eval_steps_per_second": 5.466,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.2626,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.2805367708206177,
      "eval_runtime": 143.5377,
      "eval_samples_per_second": 43.71,
      "eval_steps_per_second": 5.469,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.2507,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.267604947090149,
      "eval_runtime": 143.5375,
      "eval_samples_per_second": 43.71,
      "eval_steps_per_second": 5.469,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.2342,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.2559330463409424,
      "eval_runtime": 143.5745,
      "eval_samples_per_second": 43.699,
      "eval_steps_per_second": 5.468,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.2114,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.2479283809661865,
      "eval_runtime": 143.5861,
      "eval_samples_per_second": 43.695,
      "eval_steps_per_second": 5.467,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.2207,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.234686255455017,
      "eval_runtime": 143.5291,
      "eval_samples_per_second": 43.712,
      "eval_steps_per_second": 5.469,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 1.2032,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 1.2260726690292358,
      "eval_runtime": 143.5138,
      "eval_samples_per_second": 43.717,
      "eval_steps_per_second": 5.47,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.2012,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.2128502130508423,
      "eval_runtime": 143.5068,
      "eval_samples_per_second": 43.719,
      "eval_steps_per_second": 5.47,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 2.4438234279579648e+17,
  "trial_name": null,
  "trial_params": null
}