{
  "best_metric": 0.9523385300668151,
  "best_model_checkpoint": "finetune-bert-SST2\\checkpoint-948",
  "epoch": 5.0,
  "global_step": 4740,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.53,
      "learning_rate": 4.4725738396624474e-05,
      "loss": 0.2416,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9523385300668151,
      "eval_loss": 0.14288583397865295,
      "eval_runtime": 20.4577,
      "eval_samples_per_second": 329.216,
      "eval_steps_per_second": 5.181,
      "step": 948
    },
    {
      "epoch": 1.05,
      "learning_rate": 3.945147679324895e-05,
      "loss": 0.166,
      "step": 1000
    },
    {
      "epoch": 1.58,
      "learning_rate": 3.4177215189873416e-05,
      "loss": 0.1049,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9515961395694135,
      "eval_loss": 0.15094411373138428,
      "eval_runtime": 17.1692,
      "eval_samples_per_second": 392.271,
      "eval_steps_per_second": 6.174,
      "step": 1896
    },
    {
      "epoch": 2.11,
      "learning_rate": 2.8902953586497894e-05,
      "loss": 0.0921,
      "step": 2000
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3628691983122365e-05,
      "loss": 0.0597,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9518930957683742,
      "eval_loss": 0.16512708365917206,
      "eval_runtime": 17.5031,
      "eval_samples_per_second": 384.788,
      "eval_steps_per_second": 6.056,
      "step": 2844
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.8354430379746836e-05,
      "loss": 0.053,
      "step": 3000
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.308016877637131e-05,
      "loss": 0.036,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9515961395694135,
      "eval_loss": 0.21384631097316742,
      "eval_runtime": 17.6197,
      "eval_samples_per_second": 382.243,
      "eval_steps_per_second": 6.016,
      "step": 3792
    },
    {
      "epoch": 4.22,
      "learning_rate": 7.805907172995782e-06,
      "loss": 0.0292,
      "step": 4000
    },
    {
      "epoch": 4.75,
      "learning_rate": 2.531645569620253e-06,
      "loss": 0.0209,
      "step": 4500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9505567928730512,
      "eval_loss": 0.25926005840301514,
      "eval_runtime": 17.2903,
      "eval_samples_per_second": 389.525,
      "eval_steps_per_second": 6.131,
      "step": 4740
    }
  ],
  "max_steps": 4740,
  "num_train_epochs": 5,
  "total_flos": 6931566197923680.0,
  "trial_name": null,
  "trial_params": null
}