trainer_state.json

{
  "best_metric": 0.02877461351454258,
  "best_model_checkpoint": "./vit-base-beans/checkpoint-100",
  "epoch": 3.076923076923077,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8579,
      "step": 10
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3328,
      "step": 20
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2982,
      "step": 30
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2092,
      "step": 40
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.2919,
      "step": 50
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2561,
      "step": 60
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.175,
      "step": 70
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1027,
      "step": 80
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0535,
      "step": 90
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.1347,
      "step": 100
    },
    {
      "epoch": 1.54,
      "eval_accuracy": 1.0,
      "eval_loss": 0.02877461351454258,
      "eval_runtime": 2.6473,
      "eval_samples_per_second": 50.239,
      "eval_steps_per_second": 6.422,
      "step": 100
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.1644,
      "step": 110
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0519,
      "step": 120
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001,
      "loss": 0.0451,
      "step": 130
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0227,
      "step": 140
    },
    {
      "epoch": 2.31,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0537,
      "step": 150
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.02,
      "step": 160
    },
    {
      "epoch": 2.62,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0149,
      "step": 170
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0195,
      "step": 180
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0449,
      "step": 190
    },
    {
      "epoch": 3.08,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.013,
      "step": 200
    },
    {
      "epoch": 3.08,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.03942460939288139,
      "eval_runtime": 3.3453,
      "eval_samples_per_second": 39.757,
      "eval_steps_per_second": 5.082,
      "step": 200
    }
  ],
  "max_steps": 260,
  "num_train_epochs": 4,
  "total_flos": 2.465817209677947e+17,
  "trial_name": null,
  "trial_params": null
}