checkpoint-7623/trainer_state.json

{
  "best_metric": 0.8069068789482117,
  "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": 9.868806838989258,
      "learning_rate": 6.584048593688102e-07,
      "loss": 0.7382,
      "step": 500
    },
    {
      "epoch": 0.39354584809130266,
      "grad_norm": 9.517308235168457,
      "learning_rate": 6.12188036445266e-07,
      "loss": 0.7407,
      "step": 1000
    },
    {
      "epoch": 0.5903187721369539,
      "grad_norm": 8.953051567077637,
      "learning_rate": 5.659712135217218e-07,
      "loss": 0.7344,
      "step": 1500
    },
    {
      "epoch": 0.7870916961826053,
      "grad_norm": 11.933226585388184,
      "learning_rate": 5.197543905981778e-07,
      "loss": 0.7296,
      "step": 2000
    },
    {
      "epoch": 0.9838646202282566,
      "grad_norm": 10.369135856628418,
      "learning_rate": 4.7353756767463353e-07,
      "loss": 0.7334,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7683883229997047,
      "eval_loss": 0.8134992718696594,
      "eval_model_preparation_time": 0.0064,
      "eval_runtime": 1217.5266,
      "eval_samples_per_second": 89.02,
      "eval_steps_per_second": 2.782,
      "step": 2541
    },
    {
      "epoch": 1.1806375442739079,
      "grad_norm": 8.827561378479004,
      "learning_rate": 4.273207447510894e-07,
      "loss": 0.7246,
      "step": 3000
    },
    {
      "epoch": 1.3774104683195592,
      "grad_norm": 13.630207061767578,
      "learning_rate": 3.8110392182754523e-07,
      "loss": 0.7408,
      "step": 3500
    },
    {
      "epoch": 1.5741833923652107,
      "grad_norm": 9.284893989562988,
      "learning_rate": 3.34887098904001e-07,
      "loss": 0.7237,
      "step": 4000
    },
    {
      "epoch": 1.770956316410862,
      "grad_norm": 11.693731307983398,
      "learning_rate": 2.886702759804569e-07,
      "loss": 0.7261,
      "step": 4500
    },
    {
      "epoch": 1.9677292404565132,
      "grad_norm": 9.744012832641602,
      "learning_rate": 2.424534530569127e-07,
      "loss": 0.7188,
      "step": 5000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7696062149394745,
      "eval_loss": 0.808617889881134,
      "eval_model_preparation_time": 0.0064,
      "eval_runtime": 1216.7178,
      "eval_samples_per_second": 89.079,
      "eval_steps_per_second": 2.784,
      "step": 5082
    },
    {
      "epoch": 2.1645021645021645,
      "grad_norm": 13.421453475952148,
      "learning_rate": 1.9623663013336853e-07,
      "loss": 0.7312,
      "step": 5500
    },
    {
      "epoch": 2.3612750885478158,
      "grad_norm": 13.031763076782227,
      "learning_rate": 1.5001980720982439e-07,
      "loss": 0.7223,
      "step": 6000
    },
    {
      "epoch": 2.558048012593467,
      "grad_norm": 13.5600004196167,
      "learning_rate": 1.0380298428628021e-07,
      "loss": 0.7265,
      "step": 6500
    },
    {
      "epoch": 2.7548209366391183,
      "grad_norm": 9.442146301269531,
      "learning_rate": 5.7586161362736035e-08,
      "loss": 0.7174,
      "step": 7000
    },
    {
      "epoch": 2.9515938606847696,
      "grad_norm": 9.767436981201172,
      "learning_rate": 1.1369338439191866e-08,
      "loss": 0.7124,
      "step": 7500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7697907440212578,
      "eval_loss": 0.8069068789482117,
      "eval_model_preparation_time": 0.0064,
      "eval_runtime": 1205.4504,
      "eval_samples_per_second": 89.912,
      "eval_steps_per_second": 2.81,
      "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
}