trainer_state.json

{
  "best_metric": 0.9211956521739131,
  "best_model_checkpoint": "dinov2-small-imagenet1k-1-layer-finetuned-eurosat/checkpoint-78",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 255.04713439941406,
      "learning_rate": 4.8571428571428576e-05,
      "loss": 3.56,
      "step": 10
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 167.83641052246094,
      "learning_rate": 4.1428571428571437e-05,
      "loss": 1.9886,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7717391304347826,
      "eval_loss": 0.8828014731407166,
      "eval_runtime": 4.0892,
      "eval_samples_per_second": 89.992,
      "eval_steps_per_second": 2.935,
      "step": 26
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 140.04786682128906,
      "learning_rate": 3.428571428571429e-05,
      "loss": 1.1548,
      "step": 30
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 134.09054565429688,
      "learning_rate": 2.714285714285714e-05,
      "loss": 0.8504,
      "step": 40
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 112.90784454345703,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 50
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8858695652173914,
      "eval_loss": 0.4112096130847931,
      "eval_runtime": 4.2464,
      "eval_samples_per_second": 86.661,
      "eval_steps_per_second": 2.826,
      "step": 52
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 121.52313995361328,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 0.5629,
      "step": 60
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 93.50647735595703,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.4834,
      "step": 70
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9211956521739131,
      "eval_loss": 0.2994256019592285,
      "eval_runtime": 5.3545,
      "eval_samples_per_second": 68.727,
      "eval_steps_per_second": 2.241,
      "step": 78
    },
    {
      "epoch": 3.0,
      "step": 78,
      "total_flos": 2.588582390393733e+17,
      "train_loss": 1.2386249945713923,
      "train_runtime": 258.1788,
      "train_samples_per_second": 38.485,
      "train_steps_per_second": 0.302
    }
  ],
  "logging_steps": 10,
  "max_steps": 78,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2.588582390393733e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}