{
  "best_metric": 1.0667964220046997,
  "best_model_checkpoint": "/kaggle/output/checkpoint-19000",
  "epoch": 0.7741199478487614,
  "eval_steps": 1000,
  "global_step": 19000,
  "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": 1.2076,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.7750000000000004e-08,
      "loss": 1.1301,
      "step": 1000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.3321357285429142,
      "eval_loss": 1.1047232151031494,
      "eval_runtime": 50.8344,
      "eval_samples_per_second": 98.555,
      "eval_steps_per_second": 12.334,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.5527777777777784e-08,
      "loss": 1.1286,
      "step": 2000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.3377245508982036,
      "eval_loss": 1.1038333177566528,
      "eval_runtime": 50.8114,
      "eval_samples_per_second": 98.6,
      "eval_steps_per_second": 12.34,
      "step": 2000
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.327777777777778e-08,
      "loss": 1.1214,
      "step": 3000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.3409181636726547,
      "eval_loss": 1.103529930114746,
      "eval_runtime": 50.8258,
      "eval_samples_per_second": 98.572,
      "eval_steps_per_second": 12.336,
      "step": 3000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.1105555555555557e-07,
      "loss": 1.1249,
      "step": 4000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.3469061876247505,
      "eval_loss": 1.1000434160232544,
      "eval_runtime": 50.6414,
      "eval_samples_per_second": 98.931,
      "eval_steps_per_second": 12.381,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.3880555555555558e-07,
      "loss": 1.1206,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.34171656686626745,
      "eval_loss": 1.0996097326278687,
      "eval_runtime": 50.6956,
      "eval_samples_per_second": 98.825,
      "eval_steps_per_second": 12.368,
      "step": 5000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6658333333333335e-07,
      "loss": 1.1161,
      "step": 6000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.3590818363273453,
      "eval_loss": 1.096977710723877,
      "eval_runtime": 50.5826,
      "eval_samples_per_second": 99.046,
      "eval_steps_per_second": 12.396,
      "step": 6000
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9433333333333334e-07,
      "loss": 1.1138,
      "step": 7000
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.3532934131736527,
      "eval_loss": 1.0949815511703491,
      "eval_runtime": 50.5604,
      "eval_samples_per_second": 99.089,
      "eval_steps_per_second": 12.401,
      "step": 7000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.2211111111111114e-07,
      "loss": 1.1135,
      "step": 8000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.36327345309381237,
      "eval_loss": 1.093558430671692,
      "eval_runtime": 50.608,
      "eval_samples_per_second": 98.996,
      "eval_steps_per_second": 12.389,
      "step": 8000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.4986111111111113e-07,
      "loss": 1.1063,
      "step": 9000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.35548902195608784,
      "eval_loss": 1.0992107391357422,
      "eval_runtime": 50.5434,
      "eval_samples_per_second": 99.123,
      "eval_steps_per_second": 12.405,
      "step": 9000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.776388888888889e-07,
      "loss": 1.1083,
      "step": 10000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.3836327345309381,
      "eval_loss": 1.0920206308364868,
      "eval_runtime": 50.5559,
      "eval_samples_per_second": 99.098,
      "eval_steps_per_second": 12.402,
      "step": 10000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.053888888888889e-07,
      "loss": 1.1044,
      "step": 11000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.38762475049900197,
      "eval_loss": 1.0904459953308105,
      "eval_runtime": 50.9725,
      "eval_samples_per_second": 98.288,
      "eval_steps_per_second": 12.301,
      "step": 11000
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.331666666666667e-07,
      "loss": 1.1019,
      "step": 12000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.3912175648702595,
      "eval_loss": 1.088543176651001,
      "eval_runtime": 50.8297,
      "eval_samples_per_second": 98.564,
      "eval_steps_per_second": 12.335,
      "step": 12000
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.609166666666667e-07,
      "loss": 1.1032,
      "step": 13000
    },
    {
      "epoch": 0.53,
      "eval_accuracy": 0.3435129740518962,
      "eval_loss": 1.0940682888031006,
      "eval_runtime": 51.0031,
      "eval_samples_per_second": 98.229,
      "eval_steps_per_second": 12.293,
      "step": 13000
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.886944444444445e-07,
      "loss": 1.1006,
      "step": 14000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.36826347305389223,
      "eval_loss": 1.0924910306930542,
      "eval_runtime": 51.2753,
      "eval_samples_per_second": 97.708,
      "eval_steps_per_second": 12.228,
      "step": 14000
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.164444444444445e-07,
      "loss": 1.1,
      "step": 15000
    },
    {
      "epoch": 0.61,
      "eval_accuracy": 0.38323353293413176,
      "eval_loss": 1.0879395008087158,
      "eval_runtime": 51.3783,
      "eval_samples_per_second": 97.512,
      "eval_steps_per_second": 12.204,
      "step": 15000
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.442222222222223e-07,
      "loss": 1.0982,
      "step": 16000
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.3964071856287425,
      "eval_loss": 1.0822893381118774,
      "eval_runtime": 51.2346,
      "eval_samples_per_second": 97.786,
      "eval_steps_per_second": 12.238,
      "step": 16000
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.7197222222222224e-07,
      "loss": 1.0953,
      "step": 17000
    },
    {
      "epoch": 0.69,
      "eval_accuracy": 0.4121756487025948,
      "eval_loss": 1.0795228481292725,
      "eval_runtime": 51.31,
      "eval_samples_per_second": 97.642,
      "eval_steps_per_second": 12.22,
      "step": 17000
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.997500000000001e-07,
      "loss": 1.0943,
      "step": 18000
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.42015968063872255,
      "eval_loss": 1.079458236694336,
      "eval_runtime": 51.3255,
      "eval_samples_per_second": 97.612,
      "eval_steps_per_second": 12.216,
      "step": 18000
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.275e-07,
      "loss": 1.0885,
      "step": 19000
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.437125748502994,
      "eval_loss": 1.0667964220046997,
      "eval_runtime": 51.1737,
      "eval_samples_per_second": 97.902,
      "eval_steps_per_second": 12.252,
      "step": 19000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000000,
  "num_train_epochs": 408,
  "save_steps": 1000,
  "total_flos": 3.9716747476992e+16,
  "trial_name": null,
  "trial_params": null
}