{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.1026072329688814,
  "eval_steps": 1000,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21026072329688814,
      "grad_norm": 2.809398889541626,
      "learning_rate": 4.5e-05,
      "loss": 0.6523,
      "step": 1000
    },
    {
      "epoch": 0.21026072329688814,
      "eval_accuracy": 0.6497723106038205,
      "eval_loss": 0.6366201639175415,
      "eval_runtime": 70.6325,
      "eval_samples_per_second": 239.394,
      "eval_steps_per_second": 7.489,
      "step": 1000
    },
    {
      "epoch": 0.42052144659377627,
      "grad_norm": 1.8323402404785156,
      "learning_rate": 4e-05,
      "loss": 0.643,
      "step": 2000
    },
    {
      "epoch": 0.42052144659377627,
      "eval_accuracy": 0.6556271807912946,
      "eval_loss": 0.6256184577941895,
      "eval_runtime": 70.7729,
      "eval_samples_per_second": 238.919,
      "eval_steps_per_second": 7.475,
      "step": 2000
    },
    {
      "epoch": 0.6307821698906644,
      "grad_norm": 2.800808906555176,
      "learning_rate": 3.5e-05,
      "loss": 0.6258,
      "step": 3000
    },
    {
      "epoch": 0.6307821698906644,
      "eval_accuracy": 0.6687562836359335,
      "eval_loss": 0.6168072819709778,
      "eval_runtime": 70.6278,
      "eval_samples_per_second": 239.41,
      "eval_steps_per_second": 7.49,
      "step": 3000
    },
    {
      "epoch": 0.8410428931875525,
      "grad_norm": 3.3484714031219482,
      "learning_rate": 3e-05,
      "loss": 0.6153,
      "step": 4000
    },
    {
      "epoch": 0.8410428931875525,
      "eval_accuracy": 0.6773315985569814,
      "eval_loss": 0.6092528104782104,
      "eval_runtime": 70.616,
      "eval_samples_per_second": 239.45,
      "eval_steps_per_second": 7.491,
      "step": 4000
    },
    {
      "epoch": 1.0513036164844407,
      "grad_norm": 2.8150999546051025,
      "learning_rate": 2.5e-05,
      "loss": 0.6059,
      "step": 5000
    },
    {
      "epoch": 1.0513036164844407,
      "eval_accuracy": 0.6848423916257614,
      "eval_loss": 0.5986923575401306,
      "eval_runtime": 70.5783,
      "eval_samples_per_second": 239.578,
      "eval_steps_per_second": 7.495,
      "step": 5000
    },
    {
      "epoch": 1.2615643397813288,
      "grad_norm": 3.897254467010498,
      "learning_rate": 2e-05,
      "loss": 0.5815,
      "step": 6000
    },
    {
      "epoch": 1.2615643397813288,
      "eval_accuracy": 0.6794015021586137,
      "eval_loss": 0.6166995763778687,
      "eval_runtime": 70.7635,
      "eval_samples_per_second": 238.951,
      "eval_steps_per_second": 7.476,
      "step": 6000
    },
    {
      "epoch": 1.471825063078217,
      "grad_norm": 3.7805373668670654,
      "learning_rate": 1.5e-05,
      "loss": 0.577,
      "step": 7000
    },
    {
      "epoch": 1.471825063078217,
      "eval_accuracy": 0.6845466911112426,
      "eval_loss": 0.6091504693031311,
      "eval_runtime": 70.9254,
      "eval_samples_per_second": 238.405,
      "eval_steps_per_second": 7.459,
      "step": 7000
    },
    {
      "epoch": 1.682085786375105,
      "grad_norm": 2.818917751312256,
      "learning_rate": 1e-05,
      "loss": 0.5777,
      "step": 8000
    },
    {
      "epoch": 1.682085786375105,
      "eval_accuracy": 0.6845466911112426,
      "eval_loss": 0.5907013416290283,
      "eval_runtime": 70.6651,
      "eval_samples_per_second": 239.284,
      "eval_steps_per_second": 7.486,
      "step": 8000
    },
    {
      "epoch": 1.8923465096719934,
      "grad_norm": 3.7022032737731934,
      "learning_rate": 5e-06,
      "loss": 0.5739,
      "step": 9000
    },
    {
      "epoch": 1.8923465096719934,
      "eval_accuracy": 0.688154237388373,
      "eval_loss": 0.5975272059440613,
      "eval_runtime": 70.5471,
      "eval_samples_per_second": 239.684,
      "eval_steps_per_second": 7.499,
      "step": 9000
    },
    {
      "epoch": 2.1026072329688814,
      "grad_norm": 3.167426824569702,
      "learning_rate": 0.0,
      "loss": 0.5533,
      "step": 10000
    },
    {
      "epoch": 2.1026072329688814,
      "eval_accuracy": 0.6901058607841978,
      "eval_loss": 0.6055890917778015,
      "eval_runtime": 70.6977,
      "eval_samples_per_second": 239.173,
      "eval_steps_per_second": 7.483,
      "step": 10000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}