{
  "best_metric": 1.0940738916397095,
  "best_model_checkpoint": "/kaggle/output/checkpoint-12000",
  "epoch": 0.48891786179921776,
  "eval_steps": 1000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.7777777777777777e-11,
      "loss": 1.1078,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.7750000000000004e-08,
      "loss": 1.128,
      "step": 1000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.330938123752495,
      "eval_loss": 1.1083118915557861,
      "eval_runtime": 12.3979,
      "eval_samples_per_second": 404.1,
      "eval_steps_per_second": 50.573,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.5527777777777784e-08,
      "loss": 1.1164,
      "step": 2000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.3323353293413174,
      "eval_loss": 1.102672815322876,
      "eval_runtime": 12.2861,
      "eval_samples_per_second": 407.778,
      "eval_steps_per_second": 51.033,
      "step": 2000
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.327777777777778e-08,
      "loss": 1.1158,
      "step": 3000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.34211576846307384,
      "eval_loss": 1.1024774312973022,
      "eval_runtime": 12.3746,
      "eval_samples_per_second": 404.863,
      "eval_steps_per_second": 50.668,
      "step": 3000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.1105555555555557e-07,
      "loss": 1.1115,
      "step": 4000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.34890219560878244,
      "eval_loss": 1.1012382507324219,
      "eval_runtime": 12.5405,
      "eval_samples_per_second": 399.505,
      "eval_steps_per_second": 49.998,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.3880555555555558e-07,
      "loss": 1.1141,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.3534930139720559,
      "eval_loss": 1.1002192497253418,
      "eval_runtime": 12.3001,
      "eval_samples_per_second": 407.314,
      "eval_steps_per_second": 50.975,
      "step": 5000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6658333333333335e-07,
      "loss": 1.112,
      "step": 6000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.3524950099800399,
      "eval_loss": 1.100056767463684,
      "eval_runtime": 12.4062,
      "eval_samples_per_second": 403.831,
      "eval_steps_per_second": 50.539,
      "step": 6000
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9433333333333334e-07,
      "loss": 1.1115,
      "step": 7000
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.34331337325349304,
      "eval_loss": 1.098986029624939,
      "eval_runtime": 12.4456,
      "eval_samples_per_second": 402.551,
      "eval_steps_per_second": 50.379,
      "step": 7000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.2211111111111114e-07,
      "loss": 1.109,
      "step": 8000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.3347305389221557,
      "eval_loss": 1.098678469657898,
      "eval_runtime": 12.5978,
      "eval_samples_per_second": 397.688,
      "eval_steps_per_second": 49.771,
      "step": 8000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.4986111111111113e-07,
      "loss": 1.1117,
      "step": 9000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.34311377245508984,
      "eval_loss": 1.0984419584274292,
      "eval_runtime": 12.313,
      "eval_samples_per_second": 406.886,
      "eval_steps_per_second": 50.922,
      "step": 9000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.776388888888889e-07,
      "loss": 1.1081,
      "step": 10000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.35429141716566864,
      "eval_loss": 1.0968743562698364,
      "eval_runtime": 12.4088,
      "eval_samples_per_second": 403.744,
      "eval_steps_per_second": 50.528,
      "step": 10000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.053888888888889e-07,
      "loss": 1.107,
      "step": 11000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.3722554890219561,
      "eval_loss": 1.0950665473937988,
      "eval_runtime": 12.5775,
      "eval_samples_per_second": 398.33,
      "eval_steps_per_second": 49.851,
      "step": 11000
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.331666666666667e-07,
      "loss": 1.1057,
      "step": 12000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.37544910179640717,
      "eval_loss": 1.0940738916397095,
      "eval_runtime": 12.5186,
      "eval_samples_per_second": 400.203,
      "eval_steps_per_second": 50.085,
      "step": 12000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000000,
  "num_train_epochs": 408,
  "save_steps": 1000,
  "total_flos": 4181276491776000.0,
  "trial_name": null,
  "trial_params": null
}