trainer_state.json

{
  "best_metric": 0.28125,
  "best_model_checkpoint": "resnet-18-finetuned-resnet/checkpoint-6",
  "epoch": 2.6666666666666665,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8888888888888888,
      "eval_accuracy": 0.28125,
      "eval_loss": 3.1846918359351296e+16,
      "eval_runtime": 0.3291,
      "eval_samples_per_second": 291.681,
      "eval_steps_per_second": 9.115,
      "step": 6
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 34356422656.0,
      "learning_rate": 2.5e-05,
      "loss": 2.9760391893693236e+16,
      "step": 10
    },
    {
      "epoch": 1.925925925925926,
      "eval_accuracy": 0.28125,
      "eval_loss": 3.1846918359351296e+16,
      "eval_runtime": 0.3235,
      "eval_samples_per_second": 296.738,
      "eval_steps_per_second": 9.273,
      "step": 13
    },
    {
      "epoch": 2.6666666666666665,
      "eval_accuracy": 0.28125,
      "eval_loss": 3.1846918359351296e+16,
      "eval_runtime": 0.4136,
      "eval_samples_per_second": 232.081,
      "eval_steps_per_second": 7.253,
      "step": 18
    },
    {
      "epoch": 2.6666666666666665,
      "step": 18,
      "total_flos": 2.326039595266867e+16,
      "train_loss": 2.9779228284041444e+16,
      "train_runtime": 15.5478,
      "train_samples_per_second": 166.712,
      "train_steps_per_second": 1.158
    }
  ],
  "logging_steps": 10,
  "max_steps": 18,
  "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": 2.326039595266867e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}