{
  "best_metric": 2.0496439933776855,
  "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": 2.8179,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.723731517791748,
      "eval_runtime": 208.4619,
      "eval_samples_per_second": 30.097,
      "eval_steps_per_second": 3.766,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.6952,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.6728808879852295,
      "eval_runtime": 204.3784,
      "eval_samples_per_second": 30.698,
      "eval_steps_per_second": 3.841,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.6508,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.6387226581573486,
      "eval_runtime": 205.5464,
      "eval_samples_per_second": 30.524,
      "eval_steps_per_second": 3.819,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.6247,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.6044600009918213,
      "eval_runtime": 204.5115,
      "eval_samples_per_second": 30.678,
      "eval_steps_per_second": 3.838,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.5814,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.5784292221069336,
      "eval_runtime": 204.5265,
      "eval_samples_per_second": 30.676,
      "eval_steps_per_second": 3.838,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 2.5538,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 2.5523250102996826,
      "eval_runtime": 205.1078,
      "eval_samples_per_second": 30.589,
      "eval_steps_per_second": 3.827,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 2.5382,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 2.5274579524993896,
      "eval_runtime": 204.8352,
      "eval_samples_per_second": 30.63,
      "eval_steps_per_second": 3.832,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 2.5141,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 2.506831645965576,
      "eval_runtime": 204.4886,
      "eval_samples_per_second": 30.681,
      "eval_steps_per_second": 3.839,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 2.4765,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 2.484386920928955,
      "eval_runtime": 204.734,
      "eval_samples_per_second": 30.645,
      "eval_steps_per_second": 3.834,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 2.4729,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 2.4648232460021973,
      "eval_runtime": 204.5514,
      "eval_samples_per_second": 30.672,
      "eval_steps_per_second": 3.838,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 2.468,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 2.442662477493286,
      "eval_runtime": 204.6558,
      "eval_samples_per_second": 30.656,
      "eval_steps_per_second": 3.836,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 2.4198,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 2.4222779273986816,
      "eval_runtime": 204.5507,
      "eval_samples_per_second": 30.672,
      "eval_steps_per_second": 3.838,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 2.4326,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 2.405611753463745,
      "eval_runtime": 204.6008,
      "eval_samples_per_second": 30.665,
      "eval_steps_per_second": 3.837,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 2.3782,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 2.381305694580078,
      "eval_runtime": 204.4744,
      "eval_samples_per_second": 30.684,
      "eval_steps_per_second": 3.839,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 2.3396,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 2.3646771907806396,
      "eval_runtime": 204.6348,
      "eval_samples_per_second": 30.659,
      "eval_steps_per_second": 3.836,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 2.3327,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 2.3456737995147705,
      "eval_runtime": 204.4826,
      "eval_samples_per_second": 30.682,
      "eval_steps_per_second": 3.839,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 2.3308,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 2.329716920852661,
      "eval_runtime": 204.6937,
      "eval_samples_per_second": 30.651,
      "eval_steps_per_second": 3.835,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 2.2938,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 2.313258409500122,
      "eval_runtime": 204.5658,
      "eval_samples_per_second": 30.67,
      "eval_steps_per_second": 3.837,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 2.2756,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 2.298874855041504,
      "eval_runtime": 204.84,
      "eval_samples_per_second": 30.629,
      "eval_steps_per_second": 3.832,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 2.2724,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 2.2830755710601807,
      "eval_runtime": 204.6478,
      "eval_samples_per_second": 30.658,
      "eval_steps_per_second": 3.836,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 2.252,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 2.2688376903533936,
      "eval_runtime": 204.8236,
      "eval_samples_per_second": 30.631,
      "eval_steps_per_second": 3.833,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 2.2357,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 2.2520592212677,
      "eval_runtime": 204.8048,
      "eval_samples_per_second": 30.634,
      "eval_steps_per_second": 3.833,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 2.2266,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 2.240290880203247,
      "eval_runtime": 204.7035,
      "eval_samples_per_second": 30.649,
      "eval_steps_per_second": 3.835,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 2.2316,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 2.22774338722229,
      "eval_runtime": 204.8482,
      "eval_samples_per_second": 30.628,
      "eval_steps_per_second": 3.832,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 2.1995,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 2.2125885486602783,
      "eval_runtime": 204.9544,
      "eval_samples_per_second": 30.612,
      "eval_steps_per_second": 3.83,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 2.1997,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 2.201737880706787,
      "eval_runtime": 205.102,
      "eval_samples_per_second": 30.59,
      "eval_steps_per_second": 3.827,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 2.1825,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 2.185546636581421,
      "eval_runtime": 204.9463,
      "eval_samples_per_second": 30.613,
      "eval_steps_per_second": 3.83,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 2.1384,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 2.1790666580200195,
      "eval_runtime": 204.8473,
      "eval_samples_per_second": 30.628,
      "eval_steps_per_second": 3.832,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 2.1105,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 2.167008638381958,
      "eval_runtime": 204.8809,
      "eval_samples_per_second": 30.623,
      "eval_steps_per_second": 3.831,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 2.1284,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 2.1564178466796875,
      "eval_runtime": 205.2562,
      "eval_samples_per_second": 30.567,
      "eval_steps_per_second": 3.824,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 2.0972,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 2.146613359451294,
      "eval_runtime": 205.098,
      "eval_samples_per_second": 30.59,
      "eval_steps_per_second": 3.827,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 2.1167,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 2.134009599685669,
      "eval_runtime": 205.3378,
      "eval_samples_per_second": 30.555,
      "eval_steps_per_second": 3.823,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 2.0774,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 2.125035285949707,
      "eval_runtime": 204.8829,
      "eval_samples_per_second": 30.622,
      "eval_steps_per_second": 3.831,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 2.0749,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 2.1194825172424316,
      "eval_runtime": 208.2795,
      "eval_samples_per_second": 30.123,
      "eval_steps_per_second": 3.769,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 2.072,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 2.106606960296631,
      "eval_runtime": 205.5389,
      "eval_samples_per_second": 30.525,
      "eval_steps_per_second": 3.819,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 2.0712,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 2.096245765686035,
      "eval_runtime": 205.0401,
      "eval_samples_per_second": 30.599,
      "eval_steps_per_second": 3.829,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 2.0508,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 2.086164951324463,
      "eval_runtime": 206.1713,
      "eval_samples_per_second": 30.431,
      "eval_steps_per_second": 3.808,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 2.0398,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 2.0767734050750732,
      "eval_runtime": 205.3418,
      "eval_samples_per_second": 30.554,
      "eval_steps_per_second": 3.823,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 2.0276,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 2.0676474571228027,
      "eval_runtime": 205.4014,
      "eval_samples_per_second": 30.545,
      "eval_steps_per_second": 3.822,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 2.0277,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 2.0607223510742188,
      "eval_runtime": 205.2677,
      "eval_samples_per_second": 30.565,
      "eval_steps_per_second": 3.824,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 2.0315,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 2.0496439933776855,
      "eval_runtime": 205.7242,
      "eval_samples_per_second": 30.497,
      "eval_steps_per_second": 3.816,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.1671768877819904e+17,
  "trial_name": null,
  "trial_params": null
}