trainer_state.json

{
  "best_metric": 0.5306122448979592,
  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-qlora-houseplant/checkpoint-63",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 70,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_loss": 0.6916753053665161,
      "eval_runtime": 2.0071,
      "eval_samples_per_second": 48.827,
      "eval_steps_per_second": 1.993,
      "step": 7
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.5469765067100525,
      "learning_rate": 9.523809523809525e-06,
      "loss": 0.688,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.6907908916473389,
      "eval_runtime": 1.9735,
      "eval_samples_per_second": 49.657,
      "eval_steps_per_second": 2.027,
      "step": 14
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.44840723276138306,
      "learning_rate": 7.936507936507936e-06,
      "loss": 0.6843,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.6899289488792419,
      "eval_runtime": 2.7667,
      "eval_samples_per_second": 35.421,
      "eval_steps_per_second": 1.446,
      "step": 21
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.689208984375,
      "eval_runtime": 1.9903,
      "eval_samples_per_second": 49.24,
      "eval_steps_per_second": 2.01,
      "step": 28
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.486519992351532,
      "learning_rate": 6.349206349206349e-06,
      "loss": 0.6871,
      "step": 30
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.688566267490387,
      "eval_runtime": 2.1429,
      "eval_samples_per_second": 45.733,
      "eval_steps_per_second": 1.867,
      "step": 35
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.487440288066864,
      "learning_rate": 4.761904761904762e-06,
      "loss": 0.6833,
      "step": 40
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.688008189201355,
      "eval_runtime": 2.0813,
      "eval_samples_per_second": 47.085,
      "eval_steps_per_second": 1.922,
      "step": 42
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.6875921487808228,
      "eval_runtime": 1.9963,
      "eval_samples_per_second": 49.091,
      "eval_steps_per_second": 2.004,
      "step": 49
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.5109758973121643,
      "learning_rate": 3.1746031746031746e-06,
      "loss": 0.6812,
      "step": 50
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5204081632653061,
      "eval_loss": 0.687288224697113,
      "eval_runtime": 2.4156,
      "eval_samples_per_second": 40.57,
      "eval_steps_per_second": 1.656,
      "step": 56
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 0.45402806997299194,
      "learning_rate": 1.5873015873015873e-06,
      "loss": 0.6807,
      "step": 60
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5306122448979592,
      "eval_loss": 0.6871312856674194,
      "eval_runtime": 1.9727,
      "eval_samples_per_second": 49.679,
      "eval_steps_per_second": 2.028,
      "step": 63
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.6085575222969055,
      "learning_rate": 0.0,
      "loss": 0.6829,
      "step": 70
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5306122448979592,
      "eval_loss": 0.6870341300964355,
      "eval_runtime": 1.9967,
      "eval_samples_per_second": 49.082,
      "eval_steps_per_second": 2.003,
      "step": 70
    },
    {
      "epoch": 10.0,
      "step": 70,
      "total_flos": 6.788424757963162e+17,
      "train_loss": 0.6839384623936244,
      "train_runtime": 322.9839,
      "train_samples_per_second": 27.029,
      "train_steps_per_second": 0.217
    }
  ],
  "logging_steps": 10,
  "max_steps": 70,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 6.788424757963162e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}