checkpoint-5082/trainer_state.json

{
  "best_metric": 0.8857481479644775,
  "best_model_checkpoint": "cat_breed_image_detection/checkpoint-5082",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 5082,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19677292404565133,
      "grad_norm": 8.470013618469238,
      "learning_rate": 6.374006359300477e-07,
      "loss": 0.8262,
      "step": 500
    },
    {
      "epoch": 0.39354584809130266,
      "grad_norm": 10.178301811218262,
      "learning_rate": 5.67845786963434e-07,
      "loss": 0.8147,
      "step": 1000
    },
    {
      "epoch": 0.5903187721369539,
      "grad_norm": 8.706421852111816,
      "learning_rate": 4.982909379968204e-07,
      "loss": 0.8178,
      "step": 1500
    },
    {
      "epoch": 0.7870916961826053,
      "grad_norm": 11.302477836608887,
      "learning_rate": 4.2873608903020664e-07,
      "loss": 0.8268,
      "step": 2000
    },
    {
      "epoch": 0.9838646202282566,
      "grad_norm": 7.3357696533203125,
      "learning_rate": 3.59181240063593e-07,
      "loss": 0.8107,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.750784248597579,
      "eval_loss": 0.8888508677482605,
      "eval_model_preparation_time": 0.0072,
      "eval_runtime": 1235.6701,
      "eval_samples_per_second": 87.713,
      "eval_steps_per_second": 2.741,
      "step": 2541
    },
    {
      "epoch": 1.1806375442739079,
      "grad_norm": 8.730005264282227,
      "learning_rate": 2.896263910969793e-07,
      "loss": 0.8139,
      "step": 3000
    },
    {
      "epoch": 1.3774104683195592,
      "grad_norm": 10.811681747436523,
      "learning_rate": 2.2007154213036567e-07,
      "loss": 0.8152,
      "step": 3500
    },
    {
      "epoch": 1.5741833923652107,
      "grad_norm": 11.135604858398438,
      "learning_rate": 1.5051669316375198e-07,
      "loss": 0.8004,
      "step": 4000
    },
    {
      "epoch": 1.770956316410862,
      "grad_norm": 17.724515914916992,
      "learning_rate": 8.096184419713831e-08,
      "loss": 0.802,
      "step": 4500
    },
    {
      "epoch": 1.9677292404565132,
      "grad_norm": 16.812559127807617,
      "learning_rate": 1.1406995230524641e-08,
      "loss": 0.8072,
      "step": 5000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7516330823737821,
      "eval_loss": 0.8857481479644775,
      "eval_model_preparation_time": 0.0072,
      "eval_runtime": 1238.8905,
      "eval_samples_per_second": 87.485,
      "eval_steps_per_second": 2.734,
      "step": 5082
    }
  ],
  "logging_steps": 500,
  "max_steps": 5082,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 2.5207063579431272e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}