{
  "best_metric": 0.9067,
  "best_model_checkpoint": "test-cifar100-vit-b-16/checkpoint-4692",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 5474,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.64,
      "learning_rate": 1.8173182316404826e-05,
      "loss": 4.0836,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8138,
      "eval_loss": 2.878326416015625,
      "eval_runtime": 79.7364,
      "eval_samples_per_second": 125.413,
      "eval_steps_per_second": 1.969,
      "step": 782
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.6346364632809646e-05,
      "loss": 3.1298,
      "step": 1000
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.4519546949214468e-05,
      "loss": 2.5016,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 1.9474856853485107,
      "eval_runtime": 51.8641,
      "eval_samples_per_second": 192.811,
      "eval_steps_per_second": 3.027,
      "step": 1564
    },
    {
      "epoch": 2.56,
      "learning_rate": 1.2692729265619292e-05,
      "loss": 2.0681,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8838,
      "eval_loss": 1.40227210521698,
      "eval_runtime": 72.9429,
      "eval_samples_per_second": 137.093,
      "eval_steps_per_second": 2.152,
      "step": 2346
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.0865911582024116e-05,
      "loss": 1.7523,
      "step": 2500
    },
    {
      "epoch": 3.84,
      "learning_rate": 9.039093898428938e-06,
      "loss": 1.5165,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8986,
      "eval_loss": 1.0694773197174072,
      "eval_runtime": 47.8979,
      "eval_samples_per_second": 208.777,
      "eval_steps_per_second": 3.278,
      "step": 3128
    },
    {
      "epoch": 4.48,
      "learning_rate": 7.21227621483376e-06,
      "loss": 1.3514,
      "step": 3500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.903,
      "eval_loss": 0.8822135925292969,
      "eval_runtime": 61.8514,
      "eval_samples_per_second": 161.678,
      "eval_steps_per_second": 2.538,
      "step": 3910
    },
    {
      "epoch": 5.12,
      "learning_rate": 5.385458531238583e-06,
      "loss": 1.2353,
      "step": 4000
    },
    {
      "epoch": 5.75,
      "learning_rate": 3.5586408476434055e-06,
      "loss": 1.1393,
      "step": 4500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9067,
      "eval_loss": 0.7875523567199707,
      "eval_runtime": 63.4983,
      "eval_samples_per_second": 157.484,
      "eval_steps_per_second": 2.473,
      "step": 4692
    },
    {
      "epoch": 6.39,
      "learning_rate": 1.731823164048228e-06,
      "loss": 1.0874,
      "step": 5000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9066,
      "eval_loss": 0.7552067637443542,
      "eval_runtime": 67.2383,
      "eval_samples_per_second": 148.725,
      "eval_steps_per_second": 2.335,
      "step": 5474
    }
  ],
  "logging_steps": 500,
  "max_steps": 5474,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 2.71460189564928e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}