trainer_state.json

{
  "best_metric": 1.55401611328125,
  "best_model_checkpoint": "vit-base-patch16-224-in21k-rotated-dungeons-v5/checkpoint-100",
  "epoch": 15.0,
  "eval_steps": 100,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.7339019775390625,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.8849,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.6627044677734375,
      "learning_rate": 8.666666666666668e-06,
      "loss": 1.69,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.5126166343688965,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.5971,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.220706462860107,
      "learning_rate": 7.333333333333333e-06,
      "loss": 1.4661,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.741865158081055,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.3621,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.6890857219696045,
      "learning_rate": 6e-06,
      "loss": 1.2873,
      "step": 60
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.308206081390381,
      "learning_rate": 5.333333333333334e-06,
      "loss": 1.2145,
      "step": 70
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.7429513931274414,
      "learning_rate": 4.666666666666667e-06,
      "loss": 1.1569,
      "step": 80
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.162903070449829,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.0965,
      "step": 90
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.555781364440918,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 1.0613,
      "step": 100
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.55401611328125,
      "eval_runtime": 0.0786,
      "eval_samples_per_second": 101.782,
      "eval_steps_per_second": 12.723,
      "step": 100
    },
    {
      "epoch": 11.0,
      "grad_norm": 4.494476318359375,
      "learning_rate": 2.666666666666667e-06,
      "loss": 1.0414,
      "step": 110
    },
    {
      "epoch": 12.0,
      "grad_norm": 3.3987624645233154,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.9951,
      "step": 120
    },
    {
      "epoch": 13.0,
      "grad_norm": 4.958319187164307,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.9781,
      "step": 130
    },
    {
      "epoch": 14.0,
      "grad_norm": 3.870296001434326,
      "learning_rate": 6.666666666666667e-07,
      "loss": 0.9437,
      "step": 140
    },
    {
      "epoch": 15.0,
      "grad_norm": 3.002636671066284,
      "learning_rate": 0.0,
      "loss": 0.9387,
      "step": 150
    },
    {
      "epoch": 15.0,
      "step": 150,
      "total_flos": 5.463430091836416e+16,
      "train_loss": 1.2475657399495443,
      "train_runtime": 27.9492,
      "train_samples_per_second": 25.224,
      "train_steps_per_second": 5.367
    }
  ],
  "logging_steps": 10,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.463430091836416e+16,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}