{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.444444444444445,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.44477811455726624,
      "learning_rate": 2.45e-05,
      "loss": 0.3655,
      "step": 50
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.3250960111618042,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 0.2586,
      "step": 100
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.2056521773338318,
      "learning_rate": 4.727777777777778e-05,
      "loss": 0.1309,
      "step": 150
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.16068345308303833,
      "learning_rate": 4.4500000000000004e-05,
      "loss": 0.0814,
      "step": 200
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.17723295092582703,
      "learning_rate": 4.172222222222222e-05,
      "loss": 0.0619,
      "step": 250
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.17438340187072754,
      "learning_rate": 3.894444444444444e-05,
      "loss": 0.0552,
      "step": 300
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 0.14024192094802856,
      "learning_rate": 3.6166666666666674e-05,
      "loss": 0.0467,
      "step": 350
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.18480558693408966,
      "learning_rate": 3.338888888888889e-05,
      "loss": 0.0394,
      "step": 400
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.15892359614372253,
      "learning_rate": 3.061111111111111e-05,
      "loss": 0.0372,
      "step": 450
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.13413561880588531,
      "learning_rate": 2.7833333333333333e-05,
      "loss": 0.0366,
      "step": 500
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 0.1481749713420868,
      "learning_rate": 2.5055555555555555e-05,
      "loss": 0.0327,
      "step": 550
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.14009827375411987,
      "learning_rate": 2.2277777777777778e-05,
      "loss": 0.0317,
      "step": 600
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.12962058186531067,
      "learning_rate": 1.9500000000000003e-05,
      "loss": 0.0309,
      "step": 650
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 0.168107807636261,
      "learning_rate": 1.6722222222222222e-05,
      "loss": 0.0293,
      "step": 700
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.11126367002725601,
      "learning_rate": 1.3944444444444446e-05,
      "loss": 0.0304,
      "step": 750
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 0.4637928009033203,
      "learning_rate": 1.1166666666666668e-05,
      "loss": 0.0279,
      "step": 800
    },
    {
      "epoch": 3.7777777777777777,
      "grad_norm": 0.12586413323879242,
      "learning_rate": 8.38888888888889e-06,
      "loss": 0.0265,
      "step": 850
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1142253577709198,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.028,
      "step": 900
    },
    {
      "epoch": 4.222222222222222,
      "grad_norm": 0.10988615453243256,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.0271,
      "step": 950
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 0.12187027931213379,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.0265,
      "step": 1000
    },
    {
      "epoch": 4.444444444444445,
      "step": 1000,
      "total_flos": 1.099960418304e+16,
      "train_loss": 0.07021328616142274,
      "train_runtime": 1219.6822,
      "train_samples_per_second": 13.118,
      "train_steps_per_second": 0.82
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.099960418304e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}