{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 20001,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 4.875006249687516e-05,
      "loss": 5.8972,
      "step": 500
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.750012499375032e-05,
      "loss": 0.9391,
      "step": 1000
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.625018749062547e-05,
      "loss": 0.6129,
      "step": 1500
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.5000249987500626e-05,
      "loss": 0.4694,
      "step": 2000
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.3750312484375784e-05,
      "loss": 0.4011,
      "step": 2500
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.250037498125094e-05,
      "loss": 0.369,
      "step": 3000
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.12504374781261e-05,
      "loss": 0.334,
      "step": 3500
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.000049997500126e-05,
      "loss": 0.3033,
      "step": 4000
    },
    {
      "epoch": 0.67,
      "learning_rate": 3.875056247187641e-05,
      "loss": 0.3123,
      "step": 4500
    },
    {
      "epoch": 0.75,
      "learning_rate": 3.7500624968751565e-05,
      "loss": 0.2711,
      "step": 5000
    },
    {
      "epoch": 0.82,
      "learning_rate": 3.625068746562672e-05,
      "loss": 0.264,
      "step": 5500
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.500074996250188e-05,
      "loss": 0.255,
      "step": 6000
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.375081245937703e-05,
      "loss": 0.2434,
      "step": 6500
    },
    {
      "epoch": 1.05,
      "learning_rate": 3.250087495625219e-05,
      "loss": 0.2266,
      "step": 7000
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.125093745312735e-05,
      "loss": 0.2457,
      "step": 7500
    },
    {
      "epoch": 1.2,
      "learning_rate": 3.00009999500025e-05,
      "loss": 0.2245,
      "step": 8000
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.875106244687766e-05,
      "loss": 0.223,
      "step": 8500
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.7501124943752816e-05,
      "loss": 0.2106,
      "step": 9000
    },
    {
      "epoch": 1.42,
      "learning_rate": 2.6251187440627974e-05,
      "loss": 0.2072,
      "step": 9500
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.5001249937503125e-05,
      "loss": 0.2031,
      "step": 10000
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.3751312434378282e-05,
      "loss": 0.2007,
      "step": 10500
    },
    {
      "epoch": 1.65,
      "learning_rate": 2.250137493125344e-05,
      "loss": 0.1991,
      "step": 11000
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.1251437428128594e-05,
      "loss": 0.1937,
      "step": 11500
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.000149992500375e-05,
      "loss": 0.1927,
      "step": 12000
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.8751562421878906e-05,
      "loss": 0.195,
      "step": 12500
    },
    {
      "epoch": 1.95,
      "learning_rate": 1.7501624918754064e-05,
      "loss": 0.1935,
      "step": 13000
    },
    {
      "epoch": 2.02,
      "learning_rate": 1.6251687415629218e-05,
      "loss": 0.1933,
      "step": 13500
    },
    {
      "epoch": 2.1,
      "learning_rate": 1.5001749912504376e-05,
      "loss": 0.1853,
      "step": 14000
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.375181240937953e-05,
      "loss": 0.1742,
      "step": 14500
    },
    {
      "epoch": 2.25,
      "learning_rate": 1.2501874906254687e-05,
      "loss": 0.1842,
      "step": 15000
    },
    {
      "epoch": 2.32,
      "learning_rate": 1.1251937403129845e-05,
      "loss": 0.1778,
      "step": 15500
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.0001999900005001e-05,
      "loss": 0.1802,
      "step": 16000
    },
    {
      "epoch": 2.47,
      "learning_rate": 8.752062396880157e-06,
      "loss": 0.174,
      "step": 16500
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.502124893755312e-06,
      "loss": 0.1869,
      "step": 17000
    },
    {
      "epoch": 2.62,
      "learning_rate": 6.252187390630469e-06,
      "loss": 0.1786,
      "step": 17500
    },
    {
      "epoch": 2.7,
      "learning_rate": 5.002249887505625e-06,
      "loss": 0.1794,
      "step": 18000
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.7523123843807808e-06,
      "loss": 0.18,
      "step": 18500
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.502374881255937e-06,
      "loss": 0.1791,
      "step": 19000
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.2524373781310935e-06,
      "loss": 0.1684,
      "step": 19500
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.4998750062496878e-09,
      "loss": 0.168,
      "step": 20000
    },
    {
      "epoch": 3.0,
      "step": 20001,
      "total_flos": 2890815009177600.0,
      "train_loss": 0.3973977528091541,
      "train_runtime": 3972.7609,
      "train_samples_per_second": 30.206,
      "train_steps_per_second": 5.035
    }
  ],
  "logging_steps": 500,
  "max_steps": 20001,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 5000,
  "total_flos": 2890815009177600.0,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}