{
  "best_metric": 0.876120619396903,
  "best_model_checkpoint": "/kaggle/working/out/checkpoint-259",
  "epoch": 9.942196531791907,
  "eval_steps": 500,
  "global_step": 430,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 10.3069,
      "step": 10
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 6.3401,
      "step": 20
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.488372093023256e-05,
      "loss": 1.3487,
      "step": 30
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.651162790697675e-05,
      "loss": 0.6237,
      "step": 40
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.7302363488182559,
      "eval_loss": 0.55576092004776,
      "eval_runtime": 23.793,
      "eval_samples_per_second": 51.57,
      "eval_steps_per_second": 0.841,
      "step": 43
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.9095607235142123e-05,
      "loss": 0.5497,
      "step": 50
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.780361757105943e-05,
      "loss": 0.5057,
      "step": 60
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.651162790697675e-05,
      "loss": 0.5098,
      "step": 70
    },
    {
      "epoch": 1.85,
      "learning_rate": 4.521963824289406e-05,
      "loss": 0.462,
      "step": 80
    },
    {
      "epoch": 1.99,
      "eval_accuracy": 0.8247758761206194,
      "eval_loss": 0.4138629734516144,
      "eval_runtime": 15.5428,
      "eval_samples_per_second": 78.943,
      "eval_steps_per_second": 1.287,
      "step": 86
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.392764857881137e-05,
      "loss": 0.4337,
      "step": 90
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.263565891472868e-05,
      "loss": 0.4433,
      "step": 100
    },
    {
      "epoch": 2.54,
      "learning_rate": 4.1343669250646e-05,
      "loss": 0.4448,
      "step": 110
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.005167958656331e-05,
      "loss": 0.4174,
      "step": 120
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.8467807660961695,
      "eval_loss": 0.37035566568374634,
      "eval_runtime": 15.694,
      "eval_samples_per_second": 78.183,
      "eval_steps_per_second": 1.274,
      "step": 129
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.875968992248062e-05,
      "loss": 0.4126,
      "step": 130
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.746770025839794e-05,
      "loss": 0.3779,
      "step": 140
    },
    {
      "epoch": 3.47,
      "learning_rate": 3.617571059431525e-05,
      "loss": 0.4097,
      "step": 150
    },
    {
      "epoch": 3.7,
      "learning_rate": 3.488372093023256e-05,
      "loss": 0.3719,
      "step": 160
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.359173126614987e-05,
      "loss": 0.4089,
      "step": 170
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8231458842705787,
      "eval_loss": 0.41381287574768066,
      "eval_runtime": 15.3608,
      "eval_samples_per_second": 79.878,
      "eval_steps_per_second": 1.302,
      "step": 173
    },
    {
      "epoch": 4.16,
      "learning_rate": 3.229974160206719e-05,
      "loss": 0.3954,
      "step": 180
    },
    {
      "epoch": 4.39,
      "learning_rate": 3.1007751937984497e-05,
      "loss": 0.3791,
      "step": 190
    },
    {
      "epoch": 4.62,
      "learning_rate": 2.971576227390181e-05,
      "loss": 0.3473,
      "step": 200
    },
    {
      "epoch": 4.86,
      "learning_rate": 2.842377260981912e-05,
      "loss": 0.3656,
      "step": 210
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.8687856560717196,
      "eval_loss": 0.3321658670902252,
      "eval_runtime": 15.6831,
      "eval_samples_per_second": 78.237,
      "eval_steps_per_second": 1.275,
      "step": 216
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.7131782945736434e-05,
      "loss": 0.3504,
      "step": 220
    },
    {
      "epoch": 5.32,
      "learning_rate": 2.5839793281653746e-05,
      "loss": 0.3416,
      "step": 230
    },
    {
      "epoch": 5.55,
      "learning_rate": 2.4547803617571062e-05,
      "loss": 0.3339,
      "step": 240
    },
    {
      "epoch": 5.78,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 0.3282,
      "step": 250
    },
    {
      "epoch": 5.99,
      "eval_accuracy": 0.876120619396903,
      "eval_loss": 0.32511940598487854,
      "eval_runtime": 15.6099,
      "eval_samples_per_second": 78.604,
      "eval_steps_per_second": 1.281,
      "step": 259
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1963824289405686e-05,
      "loss": 0.3505,
      "step": 260
    },
    {
      "epoch": 6.24,
      "learning_rate": 2.0671834625323e-05,
      "loss": 0.3177,
      "step": 270
    },
    {
      "epoch": 6.47,
      "learning_rate": 1.937984496124031e-05,
      "loss": 0.2742,
      "step": 280
    },
    {
      "epoch": 6.71,
      "learning_rate": 1.8087855297157624e-05,
      "loss": 0.3334,
      "step": 290
    },
    {
      "epoch": 6.94,
      "learning_rate": 1.6795865633074936e-05,
      "loss": 0.3251,
      "step": 300
    },
    {
      "epoch": 6.98,
      "eval_accuracy": 0.86960065199674,
      "eval_loss": 0.3296392858028412,
      "eval_runtime": 15.5863,
      "eval_samples_per_second": 78.723,
      "eval_steps_per_second": 1.283,
      "step": 302
    },
    {
      "epoch": 7.17,
      "learning_rate": 1.5503875968992248e-05,
      "loss": 0.316,
      "step": 310
    },
    {
      "epoch": 7.4,
      "learning_rate": 1.421188630490956e-05,
      "loss": 0.3163,
      "step": 320
    },
    {
      "epoch": 7.63,
      "learning_rate": 1.2919896640826873e-05,
      "loss": 0.3176,
      "step": 330
    },
    {
      "epoch": 7.86,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 0.3025,
      "step": 340
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8720456397718012,
      "eval_loss": 0.3221036493778229,
      "eval_runtime": 15.5655,
      "eval_samples_per_second": 78.828,
      "eval_steps_per_second": 1.285,
      "step": 346
    },
    {
      "epoch": 8.09,
      "learning_rate": 1.03359173126615e-05,
      "loss": 0.3025,
      "step": 350
    },
    {
      "epoch": 8.32,
      "learning_rate": 9.043927648578812e-06,
      "loss": 0.2936,
      "step": 360
    },
    {
      "epoch": 8.55,
      "learning_rate": 7.751937984496124e-06,
      "loss": 0.2871,
      "step": 370
    },
    {
      "epoch": 8.79,
      "learning_rate": 6.4599483204134365e-06,
      "loss": 0.2937,
      "step": 380
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.8728606356968215,
      "eval_loss": 0.32183438539505005,
      "eval_runtime": 15.8056,
      "eval_samples_per_second": 77.631,
      "eval_steps_per_second": 1.265,
      "step": 389
    },
    {
      "epoch": 9.02,
      "learning_rate": 5.16795865633075e-06,
      "loss": 0.2932,
      "step": 390
    },
    {
      "epoch": 9.25,
      "learning_rate": 3.875968992248062e-06,
      "loss": 0.289,
      "step": 400
    },
    {
      "epoch": 9.48,
      "learning_rate": 2.583979328165375e-06,
      "loss": 0.2809,
      "step": 410
    },
    {
      "epoch": 9.71,
      "learning_rate": 1.2919896640826874e-06,
      "loss": 0.2906,
      "step": 420
    },
    {
      "epoch": 9.94,
      "learning_rate": 0.0,
      "loss": 0.2812,
      "step": 430
    },
    {
      "epoch": 9.94,
      "eval_accuracy": 0.8753056234718827,
      "eval_loss": 0.321910560131073,
      "eval_runtime": 18.4587,
      "eval_samples_per_second": 66.473,
      "eval_steps_per_second": 1.083,
      "step": 430
    },
    {
      "epoch": 9.94,
      "step": 430,
      "total_flos": 8.584102532322165e+18,
      "train_loss": 0.7598524143529493,
      "train_runtime": 2853.5799,
      "train_samples_per_second": 38.688,
      "train_steps_per_second": 0.151
    }
  ],
  "logging_steps": 10,
  "max_steps": 430,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 8.584102532322165e+18,
  "trial_name": null,
  "trial_params": null
}