{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.842105263157895,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 8.838129043579102,
      "learning_rate": 0.0,
      "loss": 1.6145,
      "step": 1
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 8.977352142333984,
      "learning_rate": 6.309297535714574e-06,
      "loss": 1.6585,
      "step": 2
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": 8.706807136535645,
      "learning_rate": 1e-05,
      "loss": 1.6189,
      "step": 3
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 6.9042205810546875,
      "learning_rate": 1e-05,
      "loss": 1.5618,
      "step": 4
    },
    {
      "epoch": 1.2105263157894737,
      "grad_norm": 7.266517162322998,
      "learning_rate": 1e-05,
      "loss": 1.7769,
      "step": 5
    },
    {
      "epoch": 1.4210526315789473,
      "grad_norm": 5.683848857879639,
      "learning_rate": 1e-05,
      "loss": 1.3457,
      "step": 6
    },
    {
      "epoch": 1.631578947368421,
      "grad_norm": 5.676620006561279,
      "learning_rate": 1e-05,
      "loss": 1.2714,
      "step": 7
    },
    {
      "epoch": 1.8421052631578947,
      "grad_norm": 4.595517635345459,
      "learning_rate": 1e-05,
      "loss": 1.2011,
      "step": 8
    },
    {
      "epoch": 2.2105263157894735,
      "grad_norm": 3.7834291458129883,
      "learning_rate": 1e-05,
      "loss": 1.4379,
      "step": 9
    },
    {
      "epoch": 2.4210526315789473,
      "grad_norm": 3.7361178398132324,
      "learning_rate": 1e-05,
      "loss": 1.1201,
      "step": 10
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 5.665889263153076,
      "learning_rate": 1e-05,
      "loss": 1.0909,
      "step": 11
    },
    {
      "epoch": 2.8421052631578947,
      "grad_norm": 3.9199063777923584,
      "learning_rate": 1e-05,
      "loss": 1.0534,
      "step": 12
    },
    {
      "epoch": 3.2105263157894735,
      "grad_norm": 2.902587652206421,
      "learning_rate": 1e-05,
      "loss": 1.2778,
      "step": 13
    },
    {
      "epoch": 3.4210526315789473,
      "grad_norm": 3.5333783626556396,
      "learning_rate": 1e-05,
      "loss": 0.9648,
      "step": 14
    },
    {
      "epoch": 3.6315789473684212,
      "grad_norm": 3.87031888961792,
      "learning_rate": 1e-05,
      "loss": 0.9722,
      "step": 15
    },
    {
      "epoch": 3.8421052631578947,
      "grad_norm": 3.942998170852661,
      "learning_rate": 1e-05,
      "loss": 0.9615,
      "step": 16
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 2.89681339263916,
      "learning_rate": 1e-05,
      "loss": 1.1407,
      "step": 17
    },
    {
      "epoch": 4.421052631578947,
      "grad_norm": 3.372147560119629,
      "learning_rate": 1e-05,
      "loss": 0.9027,
      "step": 18
    },
    {
      "epoch": 4.631578947368421,
      "grad_norm": 4.703608989715576,
      "learning_rate": 1e-05,
      "loss": 0.8606,
      "step": 19
    },
    {
      "epoch": 4.842105263157895,
      "grad_norm": 3.7259323596954346,
      "learning_rate": 1e-05,
      "loss": 0.854,
      "step": 20
    },
    {
      "epoch": 5.2105263157894735,
      "grad_norm": 4.665022373199463,
      "learning_rate": 1e-05,
      "loss": 1.0444,
      "step": 21
    },
    {
      "epoch": 5.421052631578947,
      "grad_norm": 3.989835262298584,
      "learning_rate": 1e-05,
      "loss": 0.8102,
      "step": 22
    },
    {
      "epoch": 5.631578947368421,
      "grad_norm": 3.183357000350952,
      "learning_rate": 1e-05,
      "loss": 0.7954,
      "step": 23
    },
    {
      "epoch": 5.842105263157895,
      "grad_norm": 3.0854287147521973,
      "learning_rate": 1e-05,
      "loss": 0.7586,
      "step": 24
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 4,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 127196161114112.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}