{
  "best_global_step": 750,
  "best_metric": 0.937,
  "best_model_checkpoint": "./output/checkpoint-750",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 4.507571697235107,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 3.6814801025390627,
      "step": 25
    },
    {
      "epoch": 0.2,
      "grad_norm": 5.898707389831543,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 3.506214599609375,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 8.463967323303223,
      "learning_rate": 7.4e-06,
      "loss": 3.1335317993164065,
      "step": 75
    },
    {
      "epoch": 0.4,
      "grad_norm": 12.804802894592285,
      "learning_rate": 9.9e-06,
      "loss": 2.86414794921875,
      "step": 100
    },
    {
      "epoch": 0.5,
      "grad_norm": 17.248327255249023,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 2.2183343505859376,
      "step": 125
    },
    {
      "epoch": 0.6,
      "grad_norm": 26.214523315429688,
      "learning_rate": 1.4900000000000001e-05,
      "loss": 1.6969413757324219,
      "step": 150
    },
    {
      "epoch": 0.7,
      "grad_norm": 25.541751861572266,
      "learning_rate": 1.7400000000000003e-05,
      "loss": 1.355454559326172,
      "step": 175
    },
    {
      "epoch": 0.8,
      "grad_norm": 29.073373794555664,
      "learning_rate": 1.9900000000000003e-05,
      "loss": 1.1928968811035157,
      "step": 200
    },
    {
      "epoch": 0.9,
      "grad_norm": 20.221284866333008,
      "learning_rate": 1.94e-05,
      "loss": 0.8974443054199219,
      "step": 225
    },
    {
      "epoch": 1.0,
      "grad_norm": 16.163564682006836,
      "learning_rate": 1.8775000000000002e-05,
      "loss": 0.8238192749023437,
      "step": 250
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.903,
      "eval_loss": 0.5729588866233826,
      "eval_runtime": 6.6811,
      "eval_samples_per_second": 299.352,
      "eval_steps_per_second": 2.395,
      "step": 250
    },
    {
      "epoch": 1.1,
      "grad_norm": 23.14790916442871,
      "learning_rate": 1.815e-05,
      "loss": 0.5758992385864258,
      "step": 275
    },
    {
      "epoch": 1.2,
      "grad_norm": 26.1975040435791,
      "learning_rate": 1.7525000000000002e-05,
      "loss": 0.589322738647461,
      "step": 300
    },
    {
      "epoch": 1.3,
      "grad_norm": 21.341732025146484,
      "learning_rate": 1.69e-05,
      "loss": 0.5247846984863281,
      "step": 325
    },
    {
      "epoch": 1.4,
      "grad_norm": 25.669605255126953,
      "learning_rate": 1.6275e-05,
      "loss": 0.542167854309082,
      "step": 350
    },
    {
      "epoch": 1.5,
      "grad_norm": 20.174819946289062,
      "learning_rate": 1.565e-05,
      "loss": 0.5534855651855469,
      "step": 375
    },
    {
      "epoch": 1.6,
      "grad_norm": 21.812488555908203,
      "learning_rate": 1.5025000000000001e-05,
      "loss": 0.4083445739746094,
      "step": 400
    },
    {
      "epoch": 1.7,
      "grad_norm": 12.833174705505371,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.4940061187744141,
      "step": 425
    },
    {
      "epoch": 1.8,
      "grad_norm": 18.026479721069336,
      "learning_rate": 1.3775000000000001e-05,
      "loss": 0.4515077590942383,
      "step": 450
    },
    {
      "epoch": 1.9,
      "grad_norm": 11.27204704284668,
      "learning_rate": 1.3150000000000001e-05,
      "loss": 0.5152847671508789,
      "step": 475
    },
    {
      "epoch": 2.0,
      "grad_norm": 14.681392669677734,
      "learning_rate": 1.2525000000000001e-05,
      "loss": 0.37565738677978516,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.934,
      "eval_loss": 0.3546564280986786,
      "eval_runtime": 6.7021,
      "eval_samples_per_second": 298.415,
      "eval_steps_per_second": 2.387,
      "step": 500
    },
    {
      "epoch": 2.1,
      "grad_norm": 10.114750862121582,
      "learning_rate": 1.1900000000000001e-05,
      "loss": 0.3353092575073242,
      "step": 525
    },
    {
      "epoch": 2.2,
      "grad_norm": 15.374469757080078,
      "learning_rate": 1.1275e-05,
      "loss": 0.3263074493408203,
      "step": 550
    },
    {
      "epoch": 2.3,
      "grad_norm": 15.985736846923828,
      "learning_rate": 1.065e-05,
      "loss": 0.29777233123779295,
      "step": 575
    },
    {
      "epoch": 2.4,
      "grad_norm": 3.142488956451416,
      "learning_rate": 1.0025e-05,
      "loss": 0.3357927703857422,
      "step": 600
    },
    {
      "epoch": 2.5,
      "grad_norm": 18.786211013793945,
      "learning_rate": 9.4e-06,
      "loss": 0.3292089462280273,
      "step": 625
    },
    {
      "epoch": 2.6,
      "grad_norm": 15.298687934875488,
      "learning_rate": 8.775e-06,
      "loss": 0.2830336380004883,
      "step": 650
    },
    {
      "epoch": 2.7,
      "grad_norm": 17.82750129699707,
      "learning_rate": 8.15e-06,
      "loss": 0.29575424194335936,
      "step": 675
    },
    {
      "epoch": 2.8,
      "grad_norm": 12.27523422241211,
      "learning_rate": 7.525e-06,
      "loss": 0.25342252731323245,
      "step": 700
    },
    {
      "epoch": 2.9,
      "grad_norm": 30.67465591430664,
      "learning_rate": 6.9e-06,
      "loss": 0.30169628143310545,
      "step": 725
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.317840576171875,
      "learning_rate": 6.275e-06,
      "loss": 0.24922727584838866,
      "step": 750
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.937,
      "eval_loss": 0.30609381198883057,
      "eval_runtime": 6.7042,
      "eval_samples_per_second": 298.32,
      "eval_steps_per_second": 2.387,
      "step": 750
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 1,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3157446057984000.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}