{
  "best_metric": 0.022105108946561813,
  "best_model_checkpoint": "./vit-base-beans/checkpoint-200",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 260,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "grad_norm": 1.248571515083313,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.724,
      "step": 10
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.3788352012634277,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.2693,
      "step": 20
    },
    {
      "epoch": 0.46,
      "grad_norm": 3.0690412521362305,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2057,
      "step": 30
    },
    {
      "epoch": 0.62,
      "grad_norm": 5.163357257843018,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2455,
      "step": 40
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.3465059995651245,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.13,
      "step": 50
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.7656354904174805,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2854,
      "step": 60
    },
    {
      "epoch": 1.08,
      "grad_norm": 3.6763484477996826,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1867,
      "step": 70
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.21904461085796356,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.132,
      "step": 80
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.25529107451438904,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0762,
      "step": 90
    },
    {
      "epoch": 1.54,
      "grad_norm": 1.4750274419784546,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0717,
      "step": 100
    },
    {
      "epoch": 1.54,
      "eval_accuracy": 1.0,
      "eval_loss": 0.03406519070267677,
      "eval_runtime": 1.8018,
      "eval_samples_per_second": 73.813,
      "eval_steps_per_second": 9.435,
      "step": 100
    },
    {
      "epoch": 1.69,
      "grad_norm": 1.6858314275741577,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.1123,
      "step": 110
    },
    {
      "epoch": 1.85,
      "grad_norm": 2.8304824829101562,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1281,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.1025494933128357,
      "learning_rate": 0.0001,
      "loss": 0.0345,
      "step": 130
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.08757635951042175,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0815,
      "step": 140
    },
    {
      "epoch": 2.31,
      "grad_norm": 4.281369686126709,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0501,
      "step": 150
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.11551445722579956,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0401,
      "step": 160
    },
    {
      "epoch": 2.62,
      "grad_norm": 2.2672688961029053,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0211,
      "step": 170
    },
    {
      "epoch": 2.77,
      "grad_norm": 0.12683047354221344,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0387,
      "step": 180
    },
    {
      "epoch": 2.92,
      "grad_norm": 2.012118339538574,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0165,
      "step": 190
    },
    {
      "epoch": 3.08,
      "grad_norm": 0.06524123251438141,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.038,
      "step": 200
    },
    {
      "epoch": 3.08,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.022105108946561813,
      "eval_runtime": 1.7507,
      "eval_samples_per_second": 75.968,
      "eval_steps_per_second": 9.71,
      "step": 200
    },
    {
      "epoch": 3.23,
      "grad_norm": 0.06961577385663986,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0139,
      "step": 210
    },
    {
      "epoch": 3.38,
      "grad_norm": 0.0649501383304596,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0117,
      "step": 220
    },
    {
      "epoch": 3.54,
      "grad_norm": 0.06359098851680756,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0123,
      "step": 230
    },
    {
      "epoch": 3.69,
      "grad_norm": 0.07035394012928009,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0111,
      "step": 240
    },
    {
      "epoch": 3.85,
      "grad_norm": 0.07381828874349594,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0113,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05646485462784767,
      "learning_rate": 0.0,
      "loss": 0.0112,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11380191872326227,
      "train_runtime": 79.0957,
      "train_samples_per_second": 52.291,
      "train_steps_per_second": 3.287
    }
  ],
  "logging_steps": 10,
  "max_steps": 260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "total_flos": 3.205097416476426e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}