{
  "best_metric": 0.8440366972477065,
  "best_model_checkpoint": "microsoft/Multilingual-MiniLM-L12-H384-finetuned-sst2/run-1/checkpoint-4210",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4210,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.59,
      "grad_norm": 7.641239166259766,
      "learning_rate": 1.4914273614162126e-05,
      "loss": 0.5407,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8084862385321101,
      "eval_loss": 0.4513384699821472,
      "eval_runtime": 0.9147,
      "eval_samples_per_second": 953.342,
      "eval_steps_per_second": 60.131,
      "step": 842
    },
    {
      "epoch": 1.19,
      "grad_norm": 20.01938247680664,
      "learning_rate": 1.2904263693116017e-05,
      "loss": 0.4323,
      "step": 1000
    },
    {
      "epoch": 1.78,
      "grad_norm": 2.568437337875366,
      "learning_rate": 1.089425377206991e-05,
      "loss": 0.3959,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8256880733944955,
      "eval_loss": 0.45341619849205017,
      "eval_runtime": 1.2452,
      "eval_samples_per_second": 700.289,
      "eval_steps_per_second": 44.17,
      "step": 1684
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.6975702047348022,
      "learning_rate": 8.884243851023799e-06,
      "loss": 0.361,
      "step": 2000
    },
    {
      "epoch": 2.97,
      "grad_norm": 0.5081818699836731,
      "learning_rate": 6.874233929977691e-06,
      "loss": 0.3226,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8348623853211009,
      "eval_loss": 0.5020601153373718,
      "eval_runtime": 0.9128,
      "eval_samples_per_second": 955.316,
      "eval_steps_per_second": 60.255,
      "step": 2526
    },
    {
      "epoch": 3.56,
      "grad_norm": 39.4764518737793,
      "learning_rate": 4.864224008931583e-06,
      "loss": 0.2707,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8394495412844036,
      "eval_loss": 0.5194000005722046,
      "eval_runtime": 1.1521,
      "eval_samples_per_second": 756.852,
      "eval_steps_per_second": 47.737,
      "step": 3368
    },
    {
      "epoch": 4.16,
      "grad_norm": 0.2999955117702484,
      "learning_rate": 2.8542140878854744e-06,
      "loss": 0.2608,
      "step": 3500
    },
    {
      "epoch": 4.75,
      "grad_norm": 37.41545867919922,
      "learning_rate": 8.442041668393656e-07,
      "loss": 0.2394,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8440366972477065,
      "eval_loss": 0.5257198810577393,
      "eval_runtime": 0.9113,
      "eval_samples_per_second": 956.866,
      "eval_steps_per_second": 60.353,
      "step": 4210
    }
  ],
  "logging_steps": 500,
  "max_steps": 4210,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 143595990102684.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.6924283535208234e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 8,
    "seed": 25
  }
}