checkpoint-7623/trainer_state.json

{
  "best_metric": 1.394932508468628,
  "best_model_checkpoint": "cat_breed_image_detection/checkpoint-7623",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 7623,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19677292404565133,
      "grad_norm": 8.009295463562012,
      "learning_rate": 8.465205334741845e-07,
      "loss": 1.3833,
      "step": 500
    },
    {
      "epoch": 0.39354584809130266,
      "grad_norm": 7.348078727722168,
      "learning_rate": 7.870989040010564e-07,
      "loss": 1.377,
      "step": 1000
    },
    {
      "epoch": 0.5903187721369539,
      "grad_norm": 8.270244598388672,
      "learning_rate": 7.276772745279281e-07,
      "loss": 1.3842,
      "step": 1500
    },
    {
      "epoch": 0.7870916961826053,
      "grad_norm": 6.505548477172852,
      "learning_rate": 6.682556450547999e-07,
      "loss": 1.3704,
      "step": 2000
    },
    {
      "epoch": 0.9838646202282566,
      "grad_norm": 9.899617195129395,
      "learning_rate": 6.088340155816718e-07,
      "loss": 1.3789,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6332023176852672,
      "eval_loss": 1.415798544883728,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1403.0987,
      "eval_samples_per_second": 77.246,
      "eval_steps_per_second": 2.414,
      "step": 2541
    },
    {
      "epoch": 1.1806375442739079,
      "grad_norm": 7.209441661834717,
      "learning_rate": 5.494123861085435e-07,
      "loss": 1.3551,
      "step": 3000
    },
    {
      "epoch": 1.3774104683195592,
      "grad_norm": 6.6000471115112305,
      "learning_rate": 4.899907566354153e-07,
      "loss": 1.3677,
      "step": 3500
    },
    {
      "epoch": 1.5741833923652107,
      "grad_norm": 9.57729721069336,
      "learning_rate": 4.3056912716228705e-07,
      "loss": 1.3535,
      "step": 4000
    },
    {
      "epoch": 1.770956316410862,
      "grad_norm": 6.144454479217529,
      "learning_rate": 3.7114749768915883e-07,
      "loss": 1.357,
      "step": 4500
    },
    {
      "epoch": 1.9677292404565132,
      "grad_norm": 8.755939483642578,
      "learning_rate": 3.117258682160306e-07,
      "loss": 1.3443,
      "step": 5000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.636892899320933,
      "eval_loss": 1.4011926651000977,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1242.2087,
      "eval_samples_per_second": 87.251,
      "eval_steps_per_second": 2.727,
      "step": 5082
    },
    {
      "epoch": 2.1645021645021645,
      "grad_norm": 11.363967895507812,
      "learning_rate": 2.523042387429024e-07,
      "loss": 1.3423,
      "step": 5500
    },
    {
      "epoch": 2.3612750885478158,
      "grad_norm": 6.285542964935303,
      "learning_rate": 1.928826092697742e-07,
      "loss": 1.3412,
      "step": 6000
    },
    {
      "epoch": 2.558048012593467,
      "grad_norm": 7.968774795532227,
      "learning_rate": 1.3346097979664598e-07,
      "loss": 1.3396,
      "step": 6500
    },
    {
      "epoch": 2.7548209366391183,
      "grad_norm": 7.346720218658447,
      "learning_rate": 7.403935032351777e-08,
      "loss": 1.3324,
      "step": 7000
    },
    {
      "epoch": 2.9515938606847696,
      "grad_norm": 8.237432479858398,
      "learning_rate": 1.4617720850389543e-08,
      "loss": 1.3485,
      "step": 7500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6383137732506643,
      "eval_loss": 1.394932508468628,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1219.7548,
      "eval_samples_per_second": 88.857,
      "eval_steps_per_second": 2.777,
      "step": 7623
    }
  ],
  "logging_steps": 500,
  "max_steps": 7623,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 3.781059536914691e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}