{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 12345,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 4.797488861887404e-05,
      "loss": 0.1829,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 4.5949777237748076e-05,
      "loss": 0.1089,
      "step": 1000
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.392466585662211e-05,
      "loss": 0.0657,
      "step": 1500
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.189955447549616e-05,
      "loss": 0.0602,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.08678339421749115,
      "eval_runtime": 151.5828,
      "eval_samples_per_second": 55.831,
      "eval_steps_per_second": 6.98,
      "step": 2469
    },
    {
      "epoch": 1.01,
      "learning_rate": 3.987444309437019e-05,
      "loss": 0.0489,
      "step": 2500
    },
    {
      "epoch": 1.22,
      "learning_rate": 3.784933171324423e-05,
      "loss": 0.0415,
      "step": 3000
    },
    {
      "epoch": 1.42,
      "learning_rate": 3.582422033211827e-05,
      "loss": 0.0553,
      "step": 3500
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.379910895099231e-05,
      "loss": 0.0416,
      "step": 4000
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.177399756986634e-05,
      "loss": 0.0436,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.02826148457825184,
      "eval_runtime": 151.5973,
      "eval_samples_per_second": 55.826,
      "eval_steps_per_second": 6.979,
      "step": 4938
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.974888618874038e-05,
      "loss": 0.0313,
      "step": 5000
    },
    {
      "epoch": 2.23,
      "learning_rate": 2.7723774807614422e-05,
      "loss": 0.0158,
      "step": 5500
    },
    {
      "epoch": 2.43,
      "learning_rate": 2.5698663426488455e-05,
      "loss": 0.0209,
      "step": 6000
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.3673552045362496e-05,
      "loss": 0.0163,
      "step": 6500
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.1648440664236532e-05,
      "loss": 0.0152,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.05320103093981743,
      "eval_runtime": 151.4765,
      "eval_samples_per_second": 55.87,
      "eval_steps_per_second": 6.985,
      "step": 7407
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.9623329283110573e-05,
      "loss": 0.0253,
      "step": 7500
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.759821790198461e-05,
      "loss": 0.0099,
      "step": 8000
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.557310652085865e-05,
      "loss": 0.0309,
      "step": 8500
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.3547995139732685e-05,
      "loss": 0.0204,
      "step": 9000
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.1522883758606724e-05,
      "loss": 0.0433,
      "step": 9500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.04505368694663048,
      "eval_runtime": 151.4226,
      "eval_samples_per_second": 55.89,
      "eval_steps_per_second": 6.987,
      "step": 9876
    },
    {
      "epoch": 4.05,
      "learning_rate": 9.497772377480762e-06,
      "loss": 0.0097,
      "step": 10000
    },
    {
      "epoch": 4.25,
      "learning_rate": 7.472660996354799e-06,
      "loss": 0.0172,
      "step": 10500
    },
    {
      "epoch": 4.46,
      "learning_rate": 5.447549615228838e-06,
      "loss": 0.0079,
      "step": 11000
    },
    {
      "epoch": 4.66,
      "learning_rate": 3.422438234102876e-06,
      "loss": 0.0134,
      "step": 11500
    },
    {
      "epoch": 4.86,
      "learning_rate": 1.3973268529769137e-06,
      "loss": 0.0107,
      "step": 12000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.03718067705631256,
      "eval_runtime": 151.4526,
      "eval_samples_per_second": 55.879,
      "eval_steps_per_second": 6.986,
      "step": 12345
    }
  ],
  "logging_steps": 500,
  "max_steps": 12345,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2.577587649663294e+16,
  "trial_name": null,
  "trial_params": null
}