{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0081276033729554,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.8456366062164307,
      "learning_rate": 0.0001,
      "loss": 1.7067,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.7521228790283203,
      "learning_rate": 0.0002,
      "loss": 2.4493,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.7731740474700928,
      "learning_rate": 0.00018888888888888888,
      "loss": 1.7336,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.415656089782715,
      "learning_rate": 0.00017777777777777779,
      "loss": 2.0648,
      "step": 4
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.8457527756690979,
      "learning_rate": 0.0001666666666666667,
      "loss": 1.1381,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.1292682886123657,
      "learning_rate": 0.00015555555555555556,
      "loss": 1.7235,
      "step": 6
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.8633381128311157,
      "learning_rate": 0.00014444444444444444,
      "loss": 1.0572,
      "step": 7
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.8930997848510742,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.1119,
      "step": 8
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.5441715717315674,
      "learning_rate": 0.00012222222222222224,
      "loss": 1.6177,
      "step": 9
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.9879918694496155,
      "learning_rate": 0.00011111111111111112,
      "loss": 1.2242,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.7914152145385742,
      "learning_rate": 0.0001,
      "loss": 1.2375,
      "step": 11
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.8946468234062195,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.9588,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7223942279815674,
      "learning_rate": 7.777777777777778e-05,
      "loss": 0.9085,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2389971017837524,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.0339,
      "step": 14
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.125956654548645,
      "learning_rate": 5.555555555555556e-05,
      "loss": 1.4494,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.5024690628051758,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.3538,
      "step": 16
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.010714054107666,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.2944,
      "step": 17
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.901399850845337,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 1.3021,
      "step": 18
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.5073766708374023,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 1.2913,
      "step": 19
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.580161690711975,
      "learning_rate": 0.0,
      "loss": 1.3949,
      "step": 20
    },
    {
      "epoch": 0.01,
      "step": 20,
      "total_flos": 1201627119648768.0,
      "train_loss": 1.4025826781988144,
      "train_runtime": 63.619,
      "train_samples_per_second": 1.257,
      "train_steps_per_second": 0.314
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1201627119648768.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}