{
  "best_metric": 2.0112404823303223,
  "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": 2.7401,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.6427574157714844,
      "eval_runtime": 205.0145,
      "eval_samples_per_second": 30.603,
      "eval_steps_per_second": 3.829,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.6057,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.591986656188965,
      "eval_runtime": 205.1897,
      "eval_samples_per_second": 30.577,
      "eval_steps_per_second": 3.826,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.5624,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.5554449558258057,
      "eval_runtime": 204.8733,
      "eval_samples_per_second": 30.624,
      "eval_steps_per_second": 3.832,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.5354,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.523040533065796,
      "eval_runtime": 205.0355,
      "eval_samples_per_second": 30.6,
      "eval_steps_per_second": 3.829,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.4938,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.497437000274658,
      "eval_runtime": 204.8026,
      "eval_samples_per_second": 30.634,
      "eval_steps_per_second": 3.833,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 2.4668,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 2.4717156887054443,
      "eval_runtime": 204.8144,
      "eval_samples_per_second": 30.633,
      "eval_steps_per_second": 3.833,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 2.4508,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 2.449354410171509,
      "eval_runtime": 204.5211,
      "eval_samples_per_second": 30.677,
      "eval_steps_per_second": 3.838,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 2.4278,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 2.4278199672698975,
      "eval_runtime": 204.6341,
      "eval_samples_per_second": 30.66,
      "eval_steps_per_second": 3.836,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 2.3928,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 2.4054911136627197,
      "eval_runtime": 204.5808,
      "eval_samples_per_second": 30.668,
      "eval_steps_per_second": 3.837,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 2.389,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 2.3888440132141113,
      "eval_runtime": 204.475,
      "eval_samples_per_second": 30.683,
      "eval_steps_per_second": 3.839,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 2.3831,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 2.366816997528076,
      "eval_runtime": 204.7174,
      "eval_samples_per_second": 30.647,
      "eval_steps_per_second": 3.835,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 2.338,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 2.348148822784424,
      "eval_runtime": 204.9159,
      "eval_samples_per_second": 30.617,
      "eval_steps_per_second": 3.831,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 2.3495,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 2.331716775894165,
      "eval_runtime": 204.6122,
      "eval_samples_per_second": 30.663,
      "eval_steps_per_second": 3.837,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 2.3005,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 2.3074026107788086,
      "eval_runtime": 204.9029,
      "eval_samples_per_second": 30.619,
      "eval_steps_per_second": 3.831,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 2.2609,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 2.2931342124938965,
      "eval_runtime": 205.4253,
      "eval_samples_per_second": 30.542,
      "eval_steps_per_second": 3.821,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 2.2532,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 2.2731645107269287,
      "eval_runtime": 205.0407,
      "eval_samples_per_second": 30.599,
      "eval_steps_per_second": 3.829,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 2.2527,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 2.25679087638855,
      "eval_runtime": 205.2907,
      "eval_samples_per_second": 30.562,
      "eval_steps_per_second": 3.824,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 2.2141,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 2.2405381202697754,
      "eval_runtime": 205.0141,
      "eval_samples_per_second": 30.603,
      "eval_steps_per_second": 3.829,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 2.2,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 2.2261803150177,
      "eval_runtime": 205.38,
      "eval_samples_per_second": 30.548,
      "eval_steps_per_second": 3.822,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 2.1926,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 2.211982011795044,
      "eval_runtime": 205.3797,
      "eval_samples_per_second": 30.548,
      "eval_steps_per_second": 3.822,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 2.177,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 2.1988422870635986,
      "eval_runtime": 205.5358,
      "eval_samples_per_second": 30.525,
      "eval_steps_per_second": 3.819,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 2.1604,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 2.1834826469421387,
      "eval_runtime": 205.7043,
      "eval_samples_per_second": 30.5,
      "eval_steps_per_second": 3.816,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 2.1481,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 2.1711061000823975,
      "eval_runtime": 205.9233,
      "eval_samples_per_second": 30.468,
      "eval_steps_per_second": 3.812,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 2.1557,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 2.1574361324310303,
      "eval_runtime": 205.5846,
      "eval_samples_per_second": 30.518,
      "eval_steps_per_second": 3.818,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 2.1231,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 2.1463205814361572,
      "eval_runtime": 205.712,
      "eval_samples_per_second": 30.499,
      "eval_steps_per_second": 3.816,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 2.1232,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 2.1318325996398926,
      "eval_runtime": 205.4314,
      "eval_samples_per_second": 30.541,
      "eval_steps_per_second": 3.821,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 2.1077,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 2.1178267002105713,
      "eval_runtime": 205.6052,
      "eval_samples_per_second": 30.515,
      "eval_steps_per_second": 3.818,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 2.064,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 2.110224485397339,
      "eval_runtime": 206.36,
      "eval_samples_per_second": 30.403,
      "eval_steps_per_second": 3.804,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 2.037,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 2.0975825786590576,
      "eval_runtime": 206.0587,
      "eval_samples_per_second": 30.448,
      "eval_steps_per_second": 3.81,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 2.0582,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 2.089369297027588,
      "eval_runtime": 206.2047,
      "eval_samples_per_second": 30.426,
      "eval_steps_per_second": 3.807,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 2.0245,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 2.076017379760742,
      "eval_runtime": 206.0223,
      "eval_samples_per_second": 30.453,
      "eval_steps_per_second": 3.81,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 2.0405,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 2.0672309398651123,
      "eval_runtime": 206.3429,
      "eval_samples_per_second": 30.406,
      "eval_steps_per_second": 3.804,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 2.0093,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 2.05438232421875,
      "eval_runtime": 205.9025,
      "eval_samples_per_second": 30.471,
      "eval_steps_per_second": 3.812,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 2.0032,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 2.0523083209991455,
      "eval_runtime": 205.5591,
      "eval_samples_per_second": 30.522,
      "eval_steps_per_second": 3.819,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 2.001,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 2.0382425785064697,
      "eval_runtime": 205.9189,
      "eval_samples_per_second": 30.468,
      "eval_steps_per_second": 3.812,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.9958,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 2.029693365097046,
      "eval_runtime": 205.9884,
      "eval_samples_per_second": 30.458,
      "eval_steps_per_second": 3.811,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.9759,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 2.0174672603607178,
      "eval_runtime": 206.1463,
      "eval_samples_per_second": 30.435,
      "eval_steps_per_second": 3.808,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.9684,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 2.0112404823303223,
      "eval_runtime": 205.7298,
      "eval_samples_per_second": 30.496,
      "eval_steps_per_second": 3.816,
      "step": 3800
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.1085385650167808e+17,
  "trial_name": null,
  "trial_params": null
}