trainer_state.json

{
  "best_metric": 0.9803571428571428,
  "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-0-4-28Nov24-005/checkpoint-60",
  "epoch": 28.0,
  "eval_steps": 10,
  "global_step": 70,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.0,
      "grad_norm": 27.934415817260742,
      "learning_rate": 9.682539682539683e-06,
      "loss": 1.5698,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6214285714285714,
      "eval_loss": 1.2315565347671509,
      "eval_runtime": 7.1029,
      "eval_samples_per_second": 78.841,
      "eval_steps_per_second": 9.855,
      "step": 10
    },
    {
      "epoch": 8.0,
      "grad_norm": 19.06743049621582,
      "learning_rate": 8.095238095238097e-06,
      "loss": 0.7717,
      "step": 20
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8696428571428572,
      "eval_loss": 0.6292753219604492,
      "eval_runtime": 7.5831,
      "eval_samples_per_second": 73.848,
      "eval_steps_per_second": 9.231,
      "step": 20
    },
    {
      "epoch": 12.0,
      "grad_norm": 13.279122352600098,
      "learning_rate": 6.507936507936509e-06,
      "loss": 0.2702,
      "step": 30
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9571428571428572,
      "eval_loss": 0.325614333152771,
      "eval_runtime": 7.9533,
      "eval_samples_per_second": 70.411,
      "eval_steps_per_second": 8.801,
      "step": 30
    },
    {
      "epoch": 16.0,
      "grad_norm": 3.726670026779175,
      "learning_rate": 4.920634920634921e-06,
      "loss": 0.0817,
      "step": 40
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9678571428571429,
      "eval_loss": 0.20824483036994934,
      "eval_runtime": 8.0925,
      "eval_samples_per_second": 69.199,
      "eval_steps_per_second": 8.65,
      "step": 40
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.42984938621521,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.0301,
      "step": 50
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9660714285714286,
      "eval_loss": 0.1580493152141571,
      "eval_runtime": 8.3437,
      "eval_samples_per_second": 67.116,
      "eval_steps_per_second": 8.39,
      "step": 50
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.5897208452224731,
      "learning_rate": 1.746031746031746e-06,
      "loss": 0.0123,
      "step": 60
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9803571428571428,
      "eval_loss": 0.11675503849983215,
      "eval_runtime": 7.5667,
      "eval_samples_per_second": 74.009,
      "eval_steps_per_second": 9.251,
      "step": 60
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.4860112965106964,
      "learning_rate": 1.5873015873015874e-07,
      "loss": 0.0074,
      "step": 70
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9785714285714285,
      "eval_loss": 0.11327803134918213,
      "eval_runtime": 7.7681,
      "eval_samples_per_second": 72.09,
      "eval_steps_per_second": 9.011,
      "step": 70
    },
    {
      "epoch": 28.0,
      "step": 70,
      "total_flos": 6.136187839178342e+17,
      "train_loss": 0.3918761981385095,
      "train_runtime": 448.0037,
      "train_samples_per_second": 6.25,
      "train_steps_per_second": 0.156
    }
  ],
  "logging_steps": 10,
  "max_steps": 70,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 35,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.136187839178342e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}