trainer_state.json

{
  "best_metric": 0.2857142857142857,
  "best_model_checkpoint": "resnet-18/checkpoint-5",
  "epoch": 2.7272727272727275,
  "eval_steps": 500,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9090909090909091,
      "eval_accuracy": 0.2857142857142857,
      "eval_loss": 1.142593873281104e+31,
      "eval_runtime": 0.3831,
      "eval_samples_per_second": 200.981,
      "eval_steps_per_second": 7.83,
      "step": 5
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 19.20919418334961,
      "learning_rate": 1.923076923076923e-05,
      "loss": 1.08768806515315e+31,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2857142857142857,
      "eval_loss": 1.142593873281104e+31,
      "eval_runtime": 0.2161,
      "eval_samples_per_second": 356.345,
      "eval_steps_per_second": 13.884,
      "step": 11
    },
    {
      "epoch": 2.7272727272727275,
      "eval_accuracy": 0.2857142857142857,
      "eval_loss": 1.142593873281104e+31,
      "eval_runtime": 0.2381,
      "eval_samples_per_second": 323.413,
      "eval_steps_per_second": 12.601,
      "step": 15
    },
    {
      "epoch": 2.7272727272727275,
      "step": 15,
      "total_flos": 1.912117618678579e+16,
      "train_loss": 1.0846118163935475e+31,
      "train_runtime": 12.9408,
      "train_samples_per_second": 160.191,
      "train_steps_per_second": 1.159
    }
  ],
  "logging_steps": 10,
  "max_steps": 15,
  "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": 1.912117618678579e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}