{
  "best_metric": 0.9543516428392275,
  "best_model_checkpoint": "ky-finetuned-skin/checkpoint-2800",
  "epoch": 9.966131907308379,
  "eval_steps": 500,
  "global_step": 2800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 94.771484375,
      "learning_rate": 4.998015873015873e-05,
      "loss": 1.4572,
      "step": 281
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7524454477050414,
      "eval_loss": 0.7686631679534912,
      "eval_runtime": 57.1799,
      "eval_samples_per_second": 69.727,
      "eval_steps_per_second": 2.186,
      "step": 281
    },
    {
      "epoch": 2.0,
      "grad_norm": 128.3647003173828,
      "learning_rate": 4.440476190476191e-05,
      "loss": 0.7019,
      "step": 562
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8532731376975169,
      "eval_loss": 0.46154719591140747,
      "eval_runtime": 57.2234,
      "eval_samples_per_second": 69.674,
      "eval_steps_per_second": 2.184,
      "step": 562
    },
    {
      "epoch": 3.0,
      "grad_norm": 75.07339477539062,
      "learning_rate": 3.8829365079365085e-05,
      "loss": 0.5125,
      "step": 843
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8941560070228242,
      "eval_loss": 0.3146187961101532,
      "eval_runtime": 57.2348,
      "eval_samples_per_second": 69.66,
      "eval_steps_per_second": 2.184,
      "step": 843
    },
    {
      "epoch": 4.0,
      "grad_norm": 68.2723617553711,
      "learning_rate": 3.3253968253968256e-05,
      "loss": 0.4267,
      "step": 1124
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9179834462001505,
      "eval_loss": 0.2546910047531128,
      "eval_runtime": 57.2275,
      "eval_samples_per_second": 69.669,
      "eval_steps_per_second": 2.184,
      "step": 1124
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.489927291870117,
      "learning_rate": 2.767857142857143e-05,
      "loss": 0.3441,
      "step": 1405
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9332831703034863,
      "eval_loss": 0.20282942056655884,
      "eval_runtime": 57.3383,
      "eval_samples_per_second": 69.535,
      "eval_steps_per_second": 2.18,
      "step": 1405
    },
    {
      "epoch": 6.0,
      "grad_norm": 138.46681213378906,
      "learning_rate": 2.2103174603174602e-05,
      "loss": 0.3034,
      "step": 1686
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9320290945573113,
      "eval_loss": 0.19920165836811066,
      "eval_runtime": 57.2471,
      "eval_samples_per_second": 69.645,
      "eval_steps_per_second": 2.184,
      "step": 1686
    },
    {
      "epoch": 7.0,
      "grad_norm": 37.29979705810547,
      "learning_rate": 1.6527777777777777e-05,
      "loss": 0.2581,
      "step": 1967
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.945322297466767,
      "eval_loss": 0.15903744101524353,
      "eval_runtime": 57.3097,
      "eval_samples_per_second": 69.569,
      "eval_steps_per_second": 2.181,
      "step": 1967
    },
    {
      "epoch": 8.0,
      "grad_norm": 16.976856231689453,
      "learning_rate": 1.0952380952380953e-05,
      "loss": 0.2167,
      "step": 2248
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.946827188362177,
      "eval_loss": 0.14902354776859283,
      "eval_runtime": 57.395,
      "eval_samples_per_second": 69.466,
      "eval_steps_per_second": 2.178,
      "step": 2248
    },
    {
      "epoch": 9.0,
      "grad_norm": 38.325687408447266,
      "learning_rate": 5.3769841269841275e-06,
      "loss": 0.1879,
      "step": 2529
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9541008276899925,
      "eval_loss": 0.13178889453411102,
      "eval_runtime": 57.4694,
      "eval_samples_per_second": 69.376,
      "eval_steps_per_second": 2.175,
      "step": 2529
    },
    {
      "epoch": 9.966131907308379,
      "grad_norm": 36.65696716308594,
      "learning_rate": 0.0,
      "loss": 0.16,
      "step": 2800
    },
    {
      "epoch": 9.966131907308379,
      "eval_accuracy": 0.9543516428392275,
      "eval_loss": 0.12597672641277313,
      "eval_runtime": 57.5603,
      "eval_samples_per_second": 69.266,
      "eval_steps_per_second": 2.172,
      "step": 2800
    },
    {
      "epoch": 9.966131907308379,
      "step": 2800,
      "total_flos": 3.6547618270315807e+19,
      "train_loss": 0.4579032461983817,
      "train_runtime": 8071.3814,
      "train_samples_per_second": 44.457,
      "train_steps_per_second": 0.347
    }
  ],
  "logging_steps": 10,
  "max_steps": 2800,
  "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": 3.6547618270315807e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}