{
  "best_metric": 1.2482833862304688,
  "best_model_checkpoint": "./outputs/checkpoint-3800",
  "epoch": 2.768670309653916,
  "eval_steps": 100,
  "global_step": 3800,
  "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": 2.1911,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.0239477157592773,
      "eval_runtime": 143.6069,
      "eval_samples_per_second": 43.689,
      "eval_steps_per_second": 5.466,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.9736,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.93068528175354,
      "eval_runtime": 143.6288,
      "eval_samples_per_second": 43.682,
      "eval_steps_per_second": 5.465,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.9,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.8703595399856567,
      "eval_runtime": 143.4955,
      "eval_samples_per_second": 43.723,
      "eval_steps_per_second": 5.471,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.8484,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.8250001668930054,
      "eval_runtime": 143.5707,
      "eval_samples_per_second": 43.7,
      "eval_steps_per_second": 5.468,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.7975,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.7833176851272583,
      "eval_runtime": 143.7076,
      "eval_samples_per_second": 43.658,
      "eval_steps_per_second": 5.462,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.769,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.755702018737793,
      "eval_runtime": 143.6171,
      "eval_samples_per_second": 43.686,
      "eval_steps_per_second": 5.466,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.7302,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.7235804796218872,
      "eval_runtime": 143.5461,
      "eval_samples_per_second": 43.707,
      "eval_steps_per_second": 5.469,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.709,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.694319248199463,
      "eval_runtime": 143.4905,
      "eval_samples_per_second": 43.724,
      "eval_steps_per_second": 5.471,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.6738,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.6700080633163452,
      "eval_runtime": 143.5611,
      "eval_samples_per_second": 43.703,
      "eval_steps_per_second": 5.468,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.6587,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.6456865072250366,
      "eval_runtime": 143.5526,
      "eval_samples_per_second": 43.705,
      "eval_steps_per_second": 5.468,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.6381,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.6236809492111206,
      "eval_runtime": 143.4948,
      "eval_samples_per_second": 43.723,
      "eval_steps_per_second": 5.471,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.6032,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.6063587665557861,
      "eval_runtime": 143.5371,
      "eval_samples_per_second": 43.71,
      "eval_steps_per_second": 5.469,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.6099,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.5829013586044312,
      "eval_runtime": 143.5391,
      "eval_samples_per_second": 43.709,
      "eval_steps_per_second": 5.469,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.5636,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.5621864795684814,
      "eval_runtime": 143.917,
      "eval_samples_per_second": 43.595,
      "eval_steps_per_second": 5.455,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.5352,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.5466724634170532,
      "eval_runtime": 143.4292,
      "eval_samples_per_second": 43.743,
      "eval_steps_per_second": 5.473,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.5165,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.5282047986984253,
      "eval_runtime": 152.278,
      "eval_samples_per_second": 41.201,
      "eval_steps_per_second": 5.155,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.5146,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.5113346576690674,
      "eval_runtime": 143.5758,
      "eval_samples_per_second": 43.698,
      "eval_steps_per_second": 5.467,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.4882,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.4970587491989136,
      "eval_runtime": 143.4845,
      "eval_samples_per_second": 43.726,
      "eval_steps_per_second": 5.471,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.4774,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.48170804977417,
      "eval_runtime": 143.5274,
      "eval_samples_per_second": 43.713,
      "eval_steps_per_second": 5.469,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.4619,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.4683845043182373,
      "eval_runtime": 143.5254,
      "eval_samples_per_second": 43.714,
      "eval_steps_per_second": 5.469,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.4467,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.4535853862762451,
      "eval_runtime": 143.4994,
      "eval_samples_per_second": 43.721,
      "eval_steps_per_second": 5.47,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.4209,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.438491940498352,
      "eval_runtime": 143.5083,
      "eval_samples_per_second": 43.719,
      "eval_steps_per_second": 5.47,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.4197,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.4254748821258545,
      "eval_runtime": 143.6399,
      "eval_samples_per_second": 43.679,
      "eval_steps_per_second": 5.465,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4198,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.4130115509033203,
      "eval_runtime": 143.5484,
      "eval_samples_per_second": 43.707,
      "eval_steps_per_second": 5.469,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3909,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.4006593227386475,
      "eval_runtime": 143.5037,
      "eval_samples_per_second": 43.72,
      "eval_steps_per_second": 5.47,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3813,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.3870867490768433,
      "eval_runtime": 143.5677,
      "eval_samples_per_second": 43.701,
      "eval_steps_per_second": 5.468,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3693,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.3732330799102783,
      "eval_runtime": 143.6296,
      "eval_samples_per_second": 43.682,
      "eval_steps_per_second": 5.465,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.3409,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3628050088882446,
      "eval_runtime": 143.5437,
      "eval_samples_per_second": 43.708,
      "eval_steps_per_second": 5.469,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.3156,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.3481007814407349,
      "eval_runtime": 143.5887,
      "eval_samples_per_second": 43.694,
      "eval_steps_per_second": 5.467,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.316,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.3366949558258057,
      "eval_runtime": 143.6048,
      "eval_samples_per_second": 43.689,
      "eval_steps_per_second": 5.466,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.2964,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.3265715837478638,
      "eval_runtime": 143.6034,
      "eval_samples_per_second": 43.69,
      "eval_steps_per_second": 5.466,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.3005,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.3147220611572266,
      "eval_runtime": 143.6246,
      "eval_samples_per_second": 43.683,
      "eval_steps_per_second": 5.466,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 1.2811,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.3015214204788208,
      "eval_runtime": 143.5355,
      "eval_samples_per_second": 43.71,
      "eval_steps_per_second": 5.469,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.2626,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.2932065725326538,
      "eval_runtime": 143.6069,
      "eval_samples_per_second": 43.689,
      "eval_steps_per_second": 5.466,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.2635,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.28367018699646,
      "eval_runtime": 143.5245,
      "eval_samples_per_second": 43.714,
      "eval_steps_per_second": 5.469,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.2518,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.2685626745224,
      "eval_runtime": 143.6487,
      "eval_samples_per_second": 43.676,
      "eval_steps_per_second": 5.465,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.2361,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.257932186126709,
      "eval_runtime": 143.6501,
      "eval_samples_per_second": 43.676,
      "eval_steps_per_second": 5.465,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.2134,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.2482833862304688,
      "eval_runtime": 143.6429,
      "eval_samples_per_second": 43.678,
      "eval_steps_per_second": 5.465,
      "step": 3800
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 2.2648457919379046e+17,
  "trial_name": null,
  "trial_params": null
}