{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0778911060510112,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 2e-05,
      "loss": 3.1471,
      "step": 500
    },
    {
      "epoch": 0.05,
      "learning_rate": 4e-05,
      "loss": 2.6593,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 6e-05,
      "loss": 2.6154,
      "step": 1500
    },
    {
      "epoch": 0.11,
      "learning_rate": 8e-05,
      "loss": 2.5936,
      "step": 2000
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 2.5892,
      "step": 2500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.777777777777778e-05,
      "loss": 2.5745,
      "step": 3000
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.555555555555557e-05,
      "loss": 2.5523,
      "step": 3500
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.333333333333334e-05,
      "loss": 2.5062,
      "step": 4000
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.111111111111112e-05,
      "loss": 2.5098,
      "step": 4500
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.888888888888889e-05,
      "loss": 2.4883,
      "step": 5000
    },
    {
      "epoch": 0.27,
      "eval_accuracy": 0.524201837373072,
      "eval_loss": 2.5380380153656006,
      "eval_runtime": 2042.925,
      "eval_samples_per_second": 122.374,
      "eval_steps_per_second": 7.648,
      "step": 5000
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.666666666666667e-05,
      "loss": 2.4751,
      "step": 5500
    },
    {
      "epoch": 0.32,
      "learning_rate": 8.444444444444444e-05,
      "loss": 2.4578,
      "step": 6000
    },
    {
      "epoch": 0.35,
      "learning_rate": 8.222222222222222e-05,
      "loss": 2.4484,
      "step": 6500
    },
    {
      "epoch": 0.38,
      "learning_rate": 8e-05,
      "loss": 2.4408,
      "step": 7000
    },
    {
      "epoch": 0.4,
      "learning_rate": 7.777777777777778e-05,
      "loss": 2.418,
      "step": 7500
    },
    {
      "epoch": 0.43,
      "learning_rate": 7.555555555555556e-05,
      "loss": 2.4162,
      "step": 8000
    },
    {
      "epoch": 0.46,
      "learning_rate": 7.333333333333333e-05,
      "loss": 2.4067,
      "step": 8500
    },
    {
      "epoch": 0.49,
      "learning_rate": 7.111111111111112e-05,
      "loss": 2.3932,
      "step": 9000
    },
    {
      "epoch": 0.51,
      "learning_rate": 6.88888888888889e-05,
      "loss": 2.3785,
      "step": 9500
    },
    {
      "epoch": 0.54,
      "learning_rate": 6.666666666666667e-05,
      "loss": 2.3628,
      "step": 10000
    },
    {
      "epoch": 0.54,
      "eval_accuracy": 0.530768888210296,
      "eval_loss": 2.4960896968841553,
      "eval_runtime": 2067.1205,
      "eval_samples_per_second": 120.941,
      "eval_steps_per_second": 7.559,
      "step": 10000
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.444444444444446e-05,
      "loss": 2.3703,
      "step": 10500
    },
    {
      "epoch": 0.59,
      "learning_rate": 6.222222222222222e-05,
      "loss": 2.3593,
      "step": 11000
    },
    {
      "epoch": 0.62,
      "learning_rate": 6e-05,
      "loss": 2.3603,
      "step": 11500
    },
    {
      "epoch": 0.65,
      "learning_rate": 5.7777777777777776e-05,
      "loss": 2.3447,
      "step": 12000
    },
    {
      "epoch": 0.67,
      "learning_rate": 5.555555555555556e-05,
      "loss": 2.3431,
      "step": 12500
    },
    {
      "epoch": 0.7,
      "learning_rate": 5.333333333333333e-05,
      "loss": 2.332,
      "step": 13000
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.111111111111111e-05,
      "loss": 2.3199,
      "step": 13500
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.888888888888889e-05,
      "loss": 2.3178,
      "step": 14000
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.666666666666667e-05,
      "loss": 2.3043,
      "step": 14500
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 2.3087,
      "step": 15000
    },
    {
      "epoch": 0.81,
      "eval_accuracy": 0.5384137066763606,
      "eval_loss": 2.4490230083465576,
      "eval_runtime": 2059.234,
      "eval_samples_per_second": 121.404,
      "eval_steps_per_second": 7.588,
      "step": 15000
    },
    {
      "epoch": 0.84,
      "learning_rate": 4.222222222222222e-05,
      "loss": 2.2956,
      "step": 15500
    },
    {
      "epoch": 0.86,
      "learning_rate": 4e-05,
      "loss": 2.2908,
      "step": 16000
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.777777777777778e-05,
      "loss": 2.3026,
      "step": 16500
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.555555555555556e-05,
      "loss": 2.2831,
      "step": 17000
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.2937,
      "step": 17500
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.111111111111111e-05,
      "loss": 2.2769,
      "step": 18000
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 2.2593,
      "step": 18500
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 2.245,
      "step": 19000
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 2.2561,
      "step": 19500
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 2.2413,
      "step": 20000
    },
    {
      "epoch": 1.08,
      "eval_accuracy": 0.5411358684415605,
      "eval_loss": 2.435114860534668,
      "eval_runtime": 2054.406,
      "eval_samples_per_second": 121.69,
      "eval_steps_per_second": 7.606,
      "step": 20000
    }
  ],
  "max_steps": 25000,
  "num_train_epochs": 2,
  "total_flos": 1.505927324733604e+17,
  "trial_name": null,
  "trial_params": null
}