{
  "best_metric": 0.4556220769882202,
  "best_model_checkpoint": "output/checkpoint-600",
  "epoch": 0.7061790668348046,
  "eval_steps": 50,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.5465,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 7.5e-05,
      "loss": 0.7053,
      "step": 50
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.6561679790026247,
      "eval_loss": 0.6764907836914062,
      "eval_runtime": 59.7766,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.606,
      "step": 50
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00015666666666666666,
      "loss": 0.6576,
      "step": 100
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7217847769028871,
      "eval_loss": 0.574826717376709,
      "eval_runtime": 59.8708,
      "eval_samples_per_second": 6.364,
      "eval_steps_per_second": 1.603,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.5874,
      "step": 150
    },
    {
      "epoch": 0.15,
      "eval_accuracy": 0.7480314960629921,
      "eval_loss": 0.5501317381858826,
      "eval_runtime": 55.7487,
      "eval_samples_per_second": 6.834,
      "eval_steps_per_second": 1.722,
      "step": 150
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002948780487804878,
      "loss": 0.5775,
      "step": 200
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.7401574803149606,
      "eval_loss": 0.5592061877250671,
      "eval_runtime": 54.7265,
      "eval_samples_per_second": 6.962,
      "eval_steps_per_second": 1.754,
      "step": 200
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00027658536585365855,
      "loss": 0.5821,
      "step": 250
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.7664041994750657,
      "eval_loss": 0.5005777478218079,
      "eval_runtime": 54.7418,
      "eval_samples_per_second": 6.96,
      "eval_steps_per_second": 1.754,
      "step": 250
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00025829268292682926,
      "loss": 0.5699,
      "step": 300
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.6745406824146981,
      "eval_loss": 0.7000961303710938,
      "eval_runtime": 54.7375,
      "eval_samples_per_second": 6.96,
      "eval_steps_per_second": 1.754,
      "step": 300
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.5079,
      "step": 350
    },
    {
      "epoch": 0.35,
      "eval_accuracy": 0.7322834645669292,
      "eval_loss": 0.5262107253074646,
      "eval_runtime": 54.7143,
      "eval_samples_per_second": 6.963,
      "eval_steps_per_second": 1.755,
      "step": 350
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00022170731707317072,
      "loss": 0.5826,
      "step": 400
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.7296587926509186,
      "eval_loss": 0.5798280835151672,
      "eval_runtime": 54.7163,
      "eval_samples_per_second": 6.963,
      "eval_steps_per_second": 1.755,
      "step": 400
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00020341463414634146,
      "loss": 0.5009,
      "step": 450
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.7506561679790026,
      "eval_loss": 0.5315915942192078,
      "eval_runtime": 54.7235,
      "eval_samples_per_second": 6.962,
      "eval_steps_per_second": 1.754,
      "step": 450
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00018512195121951218,
      "loss": 0.4756,
      "step": 500
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.7349081364829396,
      "eval_loss": 0.5232318639755249,
      "eval_runtime": 54.7185,
      "eval_samples_per_second": 6.963,
      "eval_steps_per_second": 1.754,
      "step": 500
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00016682926829268292,
      "loss": 0.4791,
      "step": 550
    },
    {
      "epoch": 0.55,
      "eval_accuracy": 0.7847769028871391,
      "eval_loss": 0.447153776884079,
      "eval_runtime": 54.7183,
      "eval_samples_per_second": 6.963,
      "eval_steps_per_second": 1.754,
      "step": 550
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00014853658536585364,
      "loss": 0.4415,
      "step": 600
    },
    {
      "epoch": 0.61,
      "eval_accuracy": 0.7847769028871391,
      "eval_loss": 0.4556220769882202,
      "eval_runtime": 54.7334,
      "eval_samples_per_second": 6.961,
      "eval_steps_per_second": 1.754,
      "step": 600
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00013024390243902438,
      "loss": 0.4528,
      "step": 650
    },
    {
      "epoch": 0.66,
      "eval_accuracy": 0.7690288713910761,
      "eval_loss": 0.47037920355796814,
      "eval_runtime": 54.7397,
      "eval_samples_per_second": 6.96,
      "eval_steps_per_second": 1.754,
      "step": 650
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00011195121951219511,
      "loss": 0.5177,
      "step": 700
    },
    {
      "epoch": 0.71,
      "eval_accuracy": 0.7742782152230971,
      "eval_loss": 0.4634643793106079,
      "eval_runtime": 54.7337,
      "eval_samples_per_second": 6.961,
      "eval_steps_per_second": 1.754,
      "step": 700
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 1.2097024614875136e+17,
  "trial_name": null,
  "trial_params": null
}