{
  "best_metric": 1.2591369152069092,
  "best_model_checkpoint": "./test-mlm/checkpoint-3500",
  "epoch": 14.476614699331849,
  "global_step": 6500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.11,
      "learning_rate": 9.257609502598367e-05,
      "loss": 1.4927,
      "step": 500
    },
    {
      "epoch": 1.11,
      "eval_accuracy": 0.7092214125253411,
      "eval_loss": 1.3571860790252686,
      "eval_runtime": 22.3204,
      "eval_samples_per_second": 10.752,
      "eval_steps_per_second": 2.15,
      "step": 500
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.515219005196734e-05,
      "loss": 1.4401,
      "step": 1000
    },
    {
      "epoch": 2.23,
      "eval_accuracy": 0.7100038286933217,
      "eval_loss": 1.3240282535552979,
      "eval_runtime": 22.2663,
      "eval_samples_per_second": 10.779,
      "eval_steps_per_second": 2.156,
      "step": 1000
    },
    {
      "epoch": 3.34,
      "learning_rate": 7.7728285077951e-05,
      "loss": 1.421,
      "step": 1500
    },
    {
      "epoch": 3.34,
      "eval_accuracy": 0.7097248706074639,
      "eval_loss": 1.3301254510879517,
      "eval_runtime": 22.2775,
      "eval_samples_per_second": 10.773,
      "eval_steps_per_second": 2.155,
      "step": 1500
    },
    {
      "epoch": 4.45,
      "learning_rate": 7.030438010393468e-05,
      "loss": 1.4133,
      "step": 2000
    },
    {
      "epoch": 4.45,
      "eval_accuracy": 0.7174892821809388,
      "eval_loss": 1.2943295240402222,
      "eval_runtime": 22.3325,
      "eval_samples_per_second": 10.747,
      "eval_steps_per_second": 2.149,
      "step": 2000
    },
    {
      "epoch": 5.57,
      "learning_rate": 6.288047512991834e-05,
      "loss": 1.3937,
      "step": 2500
    },
    {
      "epoch": 5.57,
      "eval_accuracy": 0.7130121905375651,
      "eval_loss": 1.324472427368164,
      "eval_runtime": 22.2745,
      "eval_samples_per_second": 10.775,
      "eval_steps_per_second": 2.155,
      "step": 2500
    },
    {
      "epoch": 6.68,
      "learning_rate": 5.545657015590201e-05,
      "loss": 1.402,
      "step": 3000
    },
    {
      "epoch": 6.68,
      "eval_accuracy": 0.714921681780709,
      "eval_loss": 1.2966943979263306,
      "eval_runtime": 22.2535,
      "eval_samples_per_second": 10.785,
      "eval_steps_per_second": 2.157,
      "step": 3000
    },
    {
      "epoch": 7.8,
      "learning_rate": 4.803266518188567e-05,
      "loss": 1.4041,
      "step": 3500
    },
    {
      "epoch": 7.8,
      "eval_accuracy": 0.7209466427834772,
      "eval_loss": 1.2591369152069092,
      "eval_runtime": 22.299,
      "eval_samples_per_second": 10.763,
      "eval_steps_per_second": 2.153,
      "step": 3500
    },
    {
      "epoch": 8.91,
      "learning_rate": 4.060876020786934e-05,
      "loss": 1.3869,
      "step": 4000
    },
    {
      "epoch": 8.91,
      "eval_accuracy": 0.7187449018434934,
      "eval_loss": 1.2874624729156494,
      "eval_runtime": 22.103,
      "eval_samples_per_second": 10.858,
      "eval_steps_per_second": 2.172,
      "step": 4000
    },
    {
      "epoch": 10.02,
      "learning_rate": 3.318485523385301e-05,
      "loss": 1.3846,
      "step": 4500
    },
    {
      "epoch": 10.02,
      "eval_accuracy": 0.7178085187817814,
      "eval_loss": 1.290149450302124,
      "eval_runtime": 22.1418,
      "eval_samples_per_second": 10.839,
      "eval_steps_per_second": 2.168,
      "step": 4500
    },
    {
      "epoch": 11.14,
      "learning_rate": 2.5760950259836673e-05,
      "loss": 1.3752,
      "step": 5000
    },
    {
      "epoch": 11.14,
      "eval_accuracy": 0.7138363074252861,
      "eval_loss": 1.2945785522460938,
      "eval_runtime": 22.153,
      "eval_samples_per_second": 10.834,
      "eval_steps_per_second": 2.167,
      "step": 5000
    },
    {
      "epoch": 12.25,
      "learning_rate": 1.8337045285820342e-05,
      "loss": 1.3749,
      "step": 5500
    },
    {
      "epoch": 12.25,
      "eval_accuracy": 0.7154974280398381,
      "eval_loss": 1.3071742057800293,
      "eval_runtime": 22.2109,
      "eval_samples_per_second": 10.805,
      "eval_steps_per_second": 2.161,
      "step": 5500
    },
    {
      "epoch": 13.36,
      "learning_rate": 1.091314031180401e-05,
      "loss": 1.3731,
      "step": 6000
    },
    {
      "epoch": 13.36,
      "eval_accuracy": 0.7149739939775527,
      "eval_loss": 1.2725176811218262,
      "eval_runtime": 22.206,
      "eval_samples_per_second": 10.808,
      "eval_steps_per_second": 2.162,
      "step": 6000
    },
    {
      "epoch": 14.48,
      "learning_rate": 3.489235337787677e-06,
      "loss": 1.3717,
      "step": 6500
    },
    {
      "epoch": 14.48,
      "eval_accuracy": 0.7124154720043279,
      "eval_loss": 1.2897814512252808,
      "eval_runtime": 22.2741,
      "eval_samples_per_second": 10.775,
      "eval_steps_per_second": 2.155,
      "step": 6500
    }
  ],
  "max_steps": 6735,
  "num_train_epochs": 15,
  "total_flos": 8679955957456896.0,
  "trial_name": null,
  "trial_params": null
}