{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.909090909090909,
  "eval_steps": 2,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 0.9523,
      "step": 1
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.953429730181653e-05,
      "loss": 0.9586,
      "step": 2
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.8882376551628113,
      "eval_runtime": 199.3091,
      "eval_samples_per_second": 1.962,
      "eval_steps_per_second": 1.962,
      "step": 2
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.814586436738998e-05,
      "loss": 0.9663,
      "step": 3
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.586056507527266e-05,
      "loss": 0.9093,
      "step": 4
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.8382185697555542,
      "eval_runtime": 197.5958,
      "eval_samples_per_second": 1.979,
      "eval_steps_per_second": 1.979,
      "step": 4
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.272097022732443e-05,
      "loss": 0.8951,
      "step": 5
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.8785564535221e-05,
      "loss": 0.802,
      "step": 6
    },
    {
      "epoch": 0.73,
      "eval_loss": 0.8089605569839478,
      "eval_runtime": 197.4382,
      "eval_samples_per_second": 1.98,
      "eval_steps_per_second": 1.98,
      "step": 6
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.412765716093272e-05,
      "loss": 0.839,
      "step": 7
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.883401610574336e-05,
      "loss": 0.839,
      "step": 8
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.778209924697876,
      "eval_runtime": 197.4012,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 8
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.300325188655761e-05,
      "loss": 0.7963,
      "step": 9
    },
    {
      "epoch": 1.21,
      "learning_rate": 6.674398060854931e-05,
      "loss": 0.7654,
      "step": 10
    },
    {
      "epoch": 1.21,
      "eval_loss": 0.7569797039031982,
      "eval_runtime": 198.9934,
      "eval_samples_per_second": 1.965,
      "eval_steps_per_second": 1.965,
      "step": 10
    },
    {
      "epoch": 1.33,
      "learning_rate": 6.01728006526317e-05,
      "loss": 0.8633,
      "step": 11
    },
    {
      "epoch": 1.45,
      "learning_rate": 5.341212066823355e-05,
      "loss": 0.7767,
      "step": 12
    },
    {
      "epoch": 1.45,
      "eval_loss": 0.7437096238136292,
      "eval_runtime": 198.5519,
      "eval_samples_per_second": 1.969,
      "eval_steps_per_second": 1.969,
      "step": 12
    },
    {
      "epoch": 1.58,
      "learning_rate": 4.658787933176646e-05,
      "loss": 0.773,
      "step": 13
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.982719934736832e-05,
      "loss": 0.7392,
      "step": 14
    },
    {
      "epoch": 1.7,
      "eval_loss": 0.7332723140716553,
      "eval_runtime": 197.6582,
      "eval_samples_per_second": 1.978,
      "eval_steps_per_second": 1.978,
      "step": 14
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.325601939145069e-05,
      "loss": 0.8132,
      "step": 15
    },
    {
      "epoch": 1.94,
      "learning_rate": 2.6996748113442394e-05,
      "loss": 0.7252,
      "step": 16
    },
    {
      "epoch": 1.94,
      "eval_loss": 0.7241747975349426,
      "eval_runtime": 197.3839,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 16
    },
    {
      "epoch": 2.06,
      "learning_rate": 2.1165983894256647e-05,
      "loss": 0.7055,
      "step": 17
    },
    {
      "epoch": 2.18,
      "learning_rate": 1.5872342839067306e-05,
      "loss": 0.7293,
      "step": 18
    },
    {
      "epoch": 2.18,
      "eval_loss": 0.7176232933998108,
      "eval_runtime": 196.8785,
      "eval_samples_per_second": 1.986,
      "eval_steps_per_second": 1.986,
      "step": 18
    },
    {
      "epoch": 2.3,
      "learning_rate": 1.1214435464779006e-05,
      "loss": 0.7933,
      "step": 19
    },
    {
      "epoch": 2.42,
      "learning_rate": 7.2790297726755716e-06,
      "loss": 0.7216,
      "step": 20
    },
    {
      "epoch": 2.42,
      "eval_loss": 0.7138686180114746,
      "eval_runtime": 197.4955,
      "eval_samples_per_second": 1.98,
      "eval_steps_per_second": 1.98,
      "step": 20
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.139434924727359e-06,
      "loss": 0.7214,
      "step": 21
    },
    {
      "epoch": 2.67,
      "learning_rate": 1.8541356326100433e-06,
      "loss": 0.8073,
      "step": 22
    },
    {
      "epoch": 2.67,
      "eval_loss": 0.7127791047096252,
      "eval_runtime": 197.4244,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 22
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.6570269818346224e-07,
      "loss": 0.6509,
      "step": 23
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0,
      "loss": 0.7432,
      "step": 24
    },
    {
      "epoch": 2.91,
      "eval_loss": 0.7129585146903992,
      "eval_runtime": 198.2692,
      "eval_samples_per_second": 1.972,
      "eval_steps_per_second": 1.972,
      "step": 24
    },
    {
      "epoch": 2.91,
      "step": 24,
      "total_flos": 6.417806722282291e+16,
      "train_loss": 0.8036013916134834,
      "train_runtime": 3665.0764,
      "train_samples_per_second": 0.32,
      "train_steps_per_second": 0.007
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 24,
  "num_train_epochs": 3,
  "save_steps": 3,
  "total_flos": 6.417806722282291e+16,
  "trial_name": null,
  "trial_params": null
}