checkpoint-56/trainer_state.json

{
  "best_metric": 0.83,
  "best_model_checkpoint": "distilhubert-finetuned-gtzan/checkpoint-56",
  "epoch": 1.9823008849557522,
  "eval_steps": 500,
  "global_step": 56,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "learning_rate": 3.5714285714285718e-06,
      "loss": 0.0338,
      "step": 5
    },
    {
      "epoch": 0.35,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.0053,
      "step": 10
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.920634920634922e-06,
      "loss": 0.0056,
      "step": 15
    },
    {
      "epoch": 0.71,
      "learning_rate": 9.523809523809525e-06,
      "loss": 0.0437,
      "step": 20
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.126984126984127e-06,
      "loss": 0.0017,
      "step": 25
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.82,
      "eval_loss": 1.0909959077835083,
      "eval_runtime": 72.1207,
      "eval_samples_per_second": 1.387,
      "eval_steps_per_second": 0.18,
      "step": 28
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.730158730158731e-06,
      "loss": 0.0013,
      "step": 30
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0373,
      "step": 35
    },
    {
      "epoch": 1.42,
      "learning_rate": 7.936507936507936e-06,
      "loss": 0.0041,
      "step": 40
    },
    {
      "epoch": 1.59,
      "learning_rate": 7.53968253968254e-06,
      "loss": 0.0012,
      "step": 45
    },
    {
      "epoch": 1.77,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.0338,
      "step": 50
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.746031746031747e-06,
      "loss": 0.0143,
      "step": 55
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.83,
      "eval_loss": 1.1379410028457642,
      "eval_runtime": 71.5138,
      "eval_samples_per_second": 1.398,
      "eval_steps_per_second": 0.182,
      "step": 56
    }
  ],
  "logging_steps": 5,
  "max_steps": 140,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.2267976549248e+17,
  "trial_name": null,
  "trial_params": null
}