trainer_state.json

{
  "best_metric": 0.8878306878306879,
  "best_model_checkpoint": "dinov2-finetuned-har/checkpoint-84",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 84,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 38.582855224609375,
      "learning_rate": 4.933333333333334e-05,
      "loss": 2.1812,
      "step": 10
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 28.47872543334961,
      "learning_rate": 4.266666666666667e-05,
      "loss": 1.1506,
      "step": 20
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 30.652055740356445,
      "learning_rate": 3.6e-05,
      "loss": 0.9425,
      "step": 30
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 24.053394317626953,
      "learning_rate": 2.9333333333333336e-05,
      "loss": 0.8886,
      "step": 40
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8470899470899471,
      "eval_loss": 0.49746325612068176,
      "eval_runtime": 38.384,
      "eval_samples_per_second": 49.239,
      "eval_steps_per_second": 0.782,
      "step": 42
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 23.17043685913086,
      "learning_rate": 2.2666666666666668e-05,
      "loss": 0.7596,
      "step": 50
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 21.568603515625,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.7396,
      "step": 60
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 14.64736270904541,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.6483,
      "step": 70
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 14.532594680786133,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.632,
      "step": 80
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8878306878306879,
      "eval_loss": 0.3669522702693939,
      "eval_runtime": 28.007,
      "eval_samples_per_second": 67.483,
      "eval_steps_per_second": 1.071,
      "step": 84
    },
    {
      "epoch": 2.0,
      "step": 84,
      "total_flos": 2.1883033324486656e+18,
      "train_loss": 0.9726932162330264,
      "train_runtime": 800.502,
      "train_samples_per_second": 26.758,
      "train_steps_per_second": 0.105
    }
  ],
  "logging_steps": 10,
  "max_steps": 84,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1883033324486656e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}