trainer_state.json

{
  "best_metric": 0.031983137130737305,
  "best_model_checkpoint": "./vit-base-beans/checkpoint-150",
  "epoch": 3.488372093023256,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 0.00018837209302325584,
      "loss": 0.4709,
      "step": 10
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017674418604651164,
      "loss": 0.1674,
      "step": 20
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016511627906976747,
      "loss": 0.166,
      "step": 30
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015348837209302327,
      "loss": 0.1066,
      "step": 40
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001418604651162791,
      "loss": 0.0363,
      "step": 50
    },
    {
      "epoch": 1.16,
      "eval_accuracy": 0.9842931937172775,
      "eval_loss": 0.05189552530646324,
      "eval_runtime": 3.2845,
      "eval_samples_per_second": 58.151,
      "eval_steps_per_second": 7.307,
      "step": 50
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001302325581395349,
      "loss": 0.0399,
      "step": 60
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011860465116279071,
      "loss": 0.0512,
      "step": 70
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00010697674418604651,
      "loss": 0.0703,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.534883720930233e-05,
      "loss": 0.0224,
      "step": 90
    },
    {
      "epoch": 2.33,
      "learning_rate": 8.372093023255814e-05,
      "loss": 0.0225,
      "step": 100
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.9842931937172775,
      "eval_loss": 0.056159574538469315,
      "eval_runtime": 3.4745,
      "eval_samples_per_second": 54.972,
      "eval_steps_per_second": 6.908,
      "step": 100
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.209302325581396e-05,
      "loss": 0.0112,
      "step": 110
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.0465116279069765e-05,
      "loss": 0.0087,
      "step": 120
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.883720930232558e-05,
      "loss": 0.0059,
      "step": 130
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.7209302325581394e-05,
      "loss": 0.0055,
      "step": 140
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.5581395348837212e-05,
      "loss": 0.0053,
      "step": 150
    },
    {
      "epoch": 3.49,
      "eval_accuracy": 0.9947643979057592,
      "eval_loss": 0.031983137130737305,
      "eval_runtime": 3.2734,
      "eval_samples_per_second": 58.348,
      "eval_steps_per_second": 7.332,
      "step": 150
    }
  ],
  "max_steps": 172,
  "num_train_epochs": 4,
  "total_flos": 1.859807750750208e+17,
  "trial_name": null,
  "trial_params": null
}