{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9473684210526314,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14035087719298245,
      "grad_norm": 0.15580356121063232,
      "learning_rate": 4e-08,
      "loss": 1.1187,
      "step": 1
    },
    {
      "epoch": 0.2807017543859649,
      "grad_norm": 0.22189772129058838,
      "learning_rate": 8e-08,
      "loss": 1.1567,
      "step": 2
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 0.1661478728055954,
      "learning_rate": 1.2000000000000002e-07,
      "loss": 1.0823,
      "step": 3
    },
    {
      "epoch": 0.5614035087719298,
      "grad_norm": 0.20104122161865234,
      "learning_rate": 1.6e-07,
      "loss": 1.1356,
      "step": 4
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 0.16658252477645874,
      "learning_rate": 2.0000000000000002e-07,
      "loss": 1.1005,
      "step": 5
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 0.5534899830818176,
      "learning_rate": 2.4000000000000003e-07,
      "loss": 1.1723,
      "step": 6
    },
    {
      "epoch": 0.9824561403508771,
      "grad_norm": 0.16570821404457092,
      "learning_rate": 2.8e-07,
      "loss": 1.0728,
      "step": 7
    },
    {
      "epoch": 1.1228070175438596,
      "grad_norm": 0.3442396819591522,
      "learning_rate": 3.2e-07,
      "loss": 1.1199,
      "step": 8
    },
    {
      "epoch": 1.263157894736842,
      "grad_norm": 0.16980823874473572,
      "learning_rate": 3.6e-07,
      "loss": 1.1108,
      "step": 9
    },
    {
      "epoch": 1.4035087719298245,
      "grad_norm": 0.5345178246498108,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 1.06,
      "step": 10
    },
    {
      "epoch": 1.543859649122807,
      "grad_norm": 0.5070182681083679,
      "learning_rate": 4.4e-07,
      "loss": 1.095,
      "step": 11
    },
    {
      "epoch": 1.6842105263157894,
      "grad_norm": 0.13856372237205505,
      "learning_rate": 4.800000000000001e-07,
      "loss": 1.1721,
      "step": 12
    },
    {
      "epoch": 1.8245614035087718,
      "grad_norm": 0.1828210949897766,
      "learning_rate": 5.2e-07,
      "loss": 1.1585,
      "step": 13
    },
    {
      "epoch": 1.9649122807017543,
      "grad_norm": 0.19238212704658508,
      "learning_rate": 5.6e-07,
      "loss": 1.1213,
      "step": 14
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 0.17399907112121582,
      "learning_rate": 6.000000000000001e-07,
      "loss": 1.1722,
      "step": 15
    },
    {
      "epoch": 2.245614035087719,
      "grad_norm": 0.22373908758163452,
      "learning_rate": 6.4e-07,
      "loss": 1.1152,
      "step": 16
    },
    {
      "epoch": 2.3859649122807016,
      "grad_norm": 0.24265660345554352,
      "learning_rate": 6.800000000000001e-07,
      "loss": 1.1472,
      "step": 17
    },
    {
      "epoch": 2.526315789473684,
      "grad_norm": 0.1733134537935257,
      "learning_rate": 7.2e-07,
      "loss": 1.1316,
      "step": 18
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.14888471364974976,
      "learning_rate": 7.6e-07,
      "loss": 0.9929,
      "step": 19
    },
    {
      "epoch": 2.807017543859649,
      "grad_norm": 0.1817580908536911,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.2424,
      "step": 20
    },
    {
      "epoch": 2.9473684210526314,
      "grad_norm": 0.13880617916584015,
      "learning_rate": 8.400000000000001e-07,
      "loss": 1.1539,
      "step": 21
    }
  ],
  "logging_steps": 1,
  "max_steps": 21,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.9136378493952e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}