{
  "best_metric": 0.9479625225067139,
  "best_model_checkpoint": "models/checkpoints/checkpoint-5500",
  "epoch": 0.9017871782259387,
  "global_step": 5500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1.9991801934743402e-05,
      "loss": 1.7109,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9983603869486802e-05,
      "loss": 1.2983,
      "step": 200
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9975405804230203e-05,
      "loss": 1.2563,
      "step": 300
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9967207738973603e-05,
      "loss": 1.2285,
      "step": 400
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9959009673717004e-05,
      "loss": 1.169,
      "step": 500
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.0381147861480713,
      "eval_runtime": 117.4251,
      "eval_samples_per_second": 583.453,
      "eval_steps_per_second": 4.565,
      "step": 500
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9950811608460404e-05,
      "loss": 1.1833,
      "step": 600
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9942613543203805e-05,
      "loss": 1.1407,
      "step": 700
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9934415477947205e-05,
      "loss": 1.1273,
      "step": 800
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.992621741269061e-05,
      "loss": 1.1193,
      "step": 900
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.991801934743401e-05,
      "loss": 1.0891,
      "step": 1000
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.9975836873054504,
      "eval_runtime": 117.3162,
      "eval_samples_per_second": 583.994,
      "eval_steps_per_second": 4.569,
      "step": 1000
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.990982128217741e-05,
      "loss": 1.1055,
      "step": 1100
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.990162321692081e-05,
      "loss": 1.1072,
      "step": 1200
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.989342515166421e-05,
      "loss": 1.0769,
      "step": 1300
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.988522708640761e-05,
      "loss": 1.0971,
      "step": 1400
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.987702902115101e-05,
      "loss": 1.0677,
      "step": 1500
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.9721714854240417,
      "eval_runtime": 117.381,
      "eval_samples_per_second": 583.672,
      "eval_steps_per_second": 4.566,
      "step": 1500
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.9868830955894412e-05,
      "loss": 1.0891,
      "step": 1600
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.9860632890637812e-05,
      "loss": 1.0386,
      "step": 1700
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.9852434825381213e-05,
      "loss": 1.073,
      "step": 1800
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.9844236760124613e-05,
      "loss": 1.0175,
      "step": 1900
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.983603869486801e-05,
      "loss": 1.0347,
      "step": 2000
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.975879430770874,
      "eval_runtime": 117.4289,
      "eval_samples_per_second": 583.434,
      "eval_steps_per_second": 4.564,
      "step": 2000
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.9827840629611414e-05,
      "loss": 1.0358,
      "step": 2100
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.9819642564354814e-05,
      "loss": 1.0148,
      "step": 2200
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.9811444499098215e-05,
      "loss": 1.0321,
      "step": 2300
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.9803246433841615e-05,
      "loss": 1.0123,
      "step": 2400
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.9795048368585015e-05,
      "loss": 1.0258,
      "step": 2500
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.9733173847198486,
      "eval_runtime": 117.3629,
      "eval_samples_per_second": 583.762,
      "eval_steps_per_second": 4.567,
      "step": 2500
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.9786850303328416e-05,
      "loss": 0.9905,
      "step": 2600
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.9778652238071816e-05,
      "loss": 1.0195,
      "step": 2700
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.9770454172815217e-05,
      "loss": 1.0054,
      "step": 2800
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.9762256107558617e-05,
      "loss": 0.9876,
      "step": 2900
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.9754058042302018e-05,
      "loss": 1.0075,
      "step": 3000
    },
    {
      "epoch": 0.49,
      "eval_loss": 0.9672998785972595,
      "eval_runtime": 117.3979,
      "eval_samples_per_second": 583.588,
      "eval_steps_per_second": 4.566,
      "step": 3000
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.9745859977045418e-05,
      "loss": 0.9962,
      "step": 3100
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.973766191178882e-05,
      "loss": 0.9725,
      "step": 3200
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.972946384653222e-05,
      "loss": 0.9977,
      "step": 3300
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.972126578127562e-05,
      "loss": 0.9587,
      "step": 3400
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.971306771601902e-05,
      "loss": 0.9964,
      "step": 3500
    },
    {
      "epoch": 0.57,
      "eval_loss": 0.9751307368278503,
      "eval_runtime": 117.3602,
      "eval_samples_per_second": 583.776,
      "eval_steps_per_second": 4.567,
      "step": 3500
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.970486965076242e-05,
      "loss": 0.9683,
      "step": 3600
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.969667158550582e-05,
      "loss": 1.0085,
      "step": 3700
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9688473520249224e-05,
      "loss": 0.9604,
      "step": 3800
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.9680275454992625e-05,
      "loss": 0.9559,
      "step": 3900
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.9672077389736025e-05,
      "loss": 0.9755,
      "step": 4000
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.9584365487098694,
      "eval_runtime": 117.3947,
      "eval_samples_per_second": 583.604,
      "eval_steps_per_second": 4.566,
      "step": 4000
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.9663879324479426e-05,
      "loss": 0.9629,
      "step": 4100
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.9655681259222826e-05,
      "loss": 0.9398,
      "step": 4200
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.9647483193966226e-05,
      "loss": 0.9452,
      "step": 4300
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.9639285128709627e-05,
      "loss": 0.9631,
      "step": 4400
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.9631087063453027e-05,
      "loss": 0.9577,
      "step": 4500
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.9555509686470032,
      "eval_runtime": 117.2815,
      "eval_samples_per_second": 584.167,
      "eval_steps_per_second": 4.57,
      "step": 4500
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.9622888998196428e-05,
      "loss": 0.96,
      "step": 4600
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.9614690932939828e-05,
      "loss": 0.9625,
      "step": 4700
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.960649286768323e-05,
      "loss": 0.9574,
      "step": 4800
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.959829480242663e-05,
      "loss": 0.966,
      "step": 4900
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.959009673717003e-05,
      "loss": 0.9536,
      "step": 5000
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.9556442499160767,
      "eval_runtime": 117.4066,
      "eval_samples_per_second": 583.545,
      "eval_steps_per_second": 4.565,
      "step": 5000
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.958189867191343e-05,
      "loss": 0.9486,
      "step": 5100
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.957370060665683e-05,
      "loss": 0.9332,
      "step": 5200
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.956550254140023e-05,
      "loss": 0.9682,
      "step": 5300
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.955730447614363e-05,
      "loss": 0.9481,
      "step": 5400
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.9549106410887035e-05,
      "loss": 0.9395,
      "step": 5500
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.9479625225067139,
      "eval_runtime": 117.3951,
      "eval_samples_per_second": 583.602,
      "eval_steps_per_second": 4.566,
      "step": 5500
    }
  ],
  "max_steps": 243960,
  "num_train_epochs": 40,
  "total_flos": 3.844584285696e+16,
  "trial_name": null,
  "trial_params": null
}