{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 4,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04938271604938271,
      "grad_norm": 1.6329176425933838,
      "learning_rate": 0.0,
      "loss": 1.5989837646484375,
      "step": 1
    },
    {
      "epoch": 0.09876543209876543,
      "grad_norm": 1.5437233448028564,
      "learning_rate": 5e-05,
      "loss": 1.397216796875,
      "step": 2
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 1.8224302530288696,
      "learning_rate": 4.97229876633906e-05,
      "loss": 1.8496208190917969,
      "step": 3
    },
    {
      "epoch": 0.19753086419753085,
      "grad_norm": 2.892070770263672,
      "learning_rate": 4.889877161664096e-05,
      "loss": 1.764373779296875,
      "step": 4
    },
    {
      "epoch": 0.24691358024691357,
      "grad_norm": 1.495557188987732,
      "learning_rate": 4.7547646794238276e-05,
      "loss": 1.0567550659179688,
      "step": 5
    },
    {
      "epoch": 0.24691358024691357,
      "eval_loss": 1.6196187734603882,
      "eval_runtime": 1.4051,
      "eval_samples_per_second": 2.135,
      "eval_steps_per_second": 0.712,
      "step": 5
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 1.864427924156189,
      "learning_rate": 4.570288237343632e-05,
      "loss": 1.2507476806640625,
      "step": 6
    },
    {
      "epoch": 0.345679012345679,
      "grad_norm": 1.3507896661758423,
      "learning_rate": 4.340990257669732e-05,
      "loss": 1.331756591796875,
      "step": 7
    },
    {
      "epoch": 0.3950617283950617,
      "grad_norm": 1.7334824800491333,
      "learning_rate": 4.072516817658065e-05,
      "loss": 1.53045654296875,
      "step": 8
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 1.8571280241012573,
      "learning_rate": 3.771478624413981e-05,
      "loss": 1.5605926513671875,
      "step": 9
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 1.555816650390625,
      "eval_runtime": 1.413,
      "eval_samples_per_second": 2.123,
      "eval_steps_per_second": 0.708,
      "step": 9
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 0.7206937670707703,
      "learning_rate": 3.4452882373436316e-05,
      "loss": 0.989166259765625,
      "step": 10
    },
    {
      "epoch": 0.5432098765432098,
      "grad_norm": 1.1625598669052124,
      "learning_rate": 3.1019775463405195e-05,
      "loss": 1.3414230346679688,
      "step": 11
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 0.6076969504356384,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 1.179987907409668,
      "step": 12
    },
    {
      "epoch": 0.6419753086419753,
      "grad_norm": 1.648446798324585,
      "learning_rate": 2.3980224536594803e-05,
      "loss": 1.587249755859375,
      "step": 13
    },
    {
      "epoch": 0.6419753086419753,
      "eval_loss": 1.5245157480239868,
      "eval_runtime": 1.4071,
      "eval_samples_per_second": 2.132,
      "eval_steps_per_second": 0.711,
      "step": 13
    },
    {
      "epoch": 0.691358024691358,
      "grad_norm": 0.6548490524291992,
      "learning_rate": 2.054711762656369e-05,
      "loss": 1.12933349609375,
      "step": 14
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 1.4055821895599365,
      "learning_rate": 1.72852137558602e-05,
      "loss": 1.6487960815429688,
      "step": 15
    },
    {
      "epoch": 0.7901234567901234,
      "grad_norm": 0.8826307058334351,
      "learning_rate": 1.427483182341936e-05,
      "loss": 1.3438186645507812,
      "step": 16
    },
    {
      "epoch": 0.8395061728395061,
      "grad_norm": 0.8792888522148132,
      "learning_rate": 1.1590097423302684e-05,
      "loss": 1.407318115234375,
      "step": 17
    },
    {
      "epoch": 0.8395061728395061,
      "eval_loss": 1.5126241445541382,
      "eval_runtime": 1.3931,
      "eval_samples_per_second": 2.153,
      "eval_steps_per_second": 0.718,
      "step": 17
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 1.2766690254211426,
      "learning_rate": 9.297117626563687e-06,
      "loss": 1.725189208984375,
      "step": 18
    },
    {
      "epoch": 0.9382716049382716,
      "grad_norm": 1.0004721879959106,
      "learning_rate": 7.452353205761725e-06,
      "loss": 1.5459928512573242,
      "step": 19
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 0.8124762177467346,
      "learning_rate": 6.1012283833590465e-06,
      "loss": 1.2266616821289062,
      "step": 20
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.1515922546386719,
      "learning_rate": 5.277012336609403e-06,
      "loss": 1.62939453125,
      "step": 21
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.506256103515625,
      "eval_runtime": 1.365,
      "eval_samples_per_second": 2.198,
      "eval_steps_per_second": 0.733,
      "step": 21
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 21,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.0688776401453056e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}