{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8995743339113984,
  "eval_steps": 500,
  "global_step": 11412,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03941352672237112,
      "grad_norm": 9.434146881103516,
      "learning_rate": 1.9840000000000003e-05,
      "loss": 3.5972,
      "step": 500
    },
    {
      "epoch": 0.07882705344474224,
      "grad_norm": 8.484919548034668,
      "learning_rate": 1.973587517972203e-05,
      "loss": 3.0731,
      "step": 1000
    },
    {
      "epoch": 0.11824058016711335,
      "grad_norm": 8.578750610351562,
      "learning_rate": 1.9469620320570853e-05,
      "loss": 2.9624,
      "step": 1500
    },
    {
      "epoch": 0.15765410688948447,
      "grad_norm": 6.2477827072143555,
      "learning_rate": 1.920336546141967e-05,
      "loss": 2.8351,
      "step": 2000
    },
    {
      "epoch": 0.19706763361185559,
      "grad_norm": 6.6669602394104,
      "learning_rate": 1.8937110602268492e-05,
      "loss": 2.817,
      "step": 2500
    },
    {
      "epoch": 0.2364811603342267,
      "grad_norm": 4.857083797454834,
      "learning_rate": 1.8670855743117314e-05,
      "loss": 2.7781,
      "step": 3000
    },
    {
      "epoch": 0.2758946870565978,
      "grad_norm": 5.961228847503662,
      "learning_rate": 1.8404600883966135e-05,
      "loss": 2.7681,
      "step": 3500
    },
    {
      "epoch": 0.31530821377896895,
      "grad_norm": 3.896094560623169,
      "learning_rate": 1.8138346024814953e-05,
      "loss": 2.6833,
      "step": 4000
    },
    {
      "epoch": 0.35472174050134003,
      "grad_norm": 4.19826078414917,
      "learning_rate": 1.7872091165663774e-05,
      "loss": 2.7317,
      "step": 4500
    },
    {
      "epoch": 0.39413526722371117,
      "grad_norm": 4.363683223724365,
      "learning_rate": 1.7605836306512596e-05,
      "loss": 2.6456,
      "step": 5000
    },
    {
      "epoch": 0.4335487939460823,
      "grad_norm": 4.608914852142334,
      "learning_rate": 1.7339581447361414e-05,
      "loss": 2.6575,
      "step": 5500
    },
    {
      "epoch": 0.4729623206684534,
      "grad_norm": 5.031886577606201,
      "learning_rate": 1.707332658821024e-05,
      "loss": 2.6423,
      "step": 6000
    },
    {
      "epoch": 0.5123758473908245,
      "grad_norm": 4.503154277801514,
      "learning_rate": 1.6807071729059057e-05,
      "loss": 2.6062,
      "step": 6500
    },
    {
      "epoch": 0.5517893741131956,
      "grad_norm": 3.5765438079833984,
      "learning_rate": 1.654134937962618e-05,
      "loss": 2.6487,
      "step": 7000
    },
    {
      "epoch": 0.5912029008355668,
      "grad_norm": 3.3804619312286377,
      "learning_rate": 1.6275094520475e-05,
      "loss": 2.6378,
      "step": 7500
    },
    {
      "epoch": 0.6306164275579379,
      "grad_norm": 4.173604965209961,
      "learning_rate": 1.600883966132382e-05,
      "loss": 2.5821,
      "step": 8000
    },
    {
      "epoch": 0.670029954280309,
      "grad_norm": 3.0175459384918213,
      "learning_rate": 1.5743117311890943e-05,
      "loss": 2.5756,
      "step": 8500
    },
    {
      "epoch": 0.7094434810026801,
      "grad_norm": 4.371723651885986,
      "learning_rate": 1.5476862452739764e-05,
      "loss": 2.5365,
      "step": 9000
    },
    {
      "epoch": 0.7488570077250513,
      "grad_norm": 3.797525644302368,
      "learning_rate": 1.5210607593588584e-05,
      "loss": 2.5734,
      "step": 9500
    },
    {
      "epoch": 0.7882705344474223,
      "grad_norm": 2.7907519340515137,
      "learning_rate": 1.4944352734437404e-05,
      "loss": 2.5801,
      "step": 10000
    },
    {
      "epoch": 0.8276840611697934,
      "grad_norm": 3.6856374740600586,
      "learning_rate": 1.4678097875286225e-05,
      "loss": 2.5198,
      "step": 10500
    },
    {
      "epoch": 0.8670975878921646,
      "grad_norm": 3.692336320877075,
      "learning_rate": 1.4411843016135045e-05,
      "loss": 2.5294,
      "step": 11000
    }
  ],
  "logging_steps": 500,
  "max_steps": 38058,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1268,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2981864669184000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}