checkpoint-1000/trainer_state.json

{
  "best_metric": 0.4822468161582947,
  "best_model_checkpoint": "eesuan/imdb-spoiler-distilbert\\checkpoint-1000",
  "epoch": 0.6968641114982579,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35,
      "grad_norm": 3.155640125274658,
      "learning_rate": 1.3031358885017422e-05,
      "loss": 0.5139,
      "step": 500
    },
    {
      "epoch": 0.35,
      "eval_accuracy": 0.7761324041811847,
      "eval_loss": 0.495978981256485,
      "eval_runtime": 1985.4773,
      "eval_samples_per_second": 2.891,
      "eval_steps_per_second": 0.181,
      "step": 500
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.961811900138855,
      "learning_rate": 6.062717770034844e-06,
      "loss": 0.4732,
      "step": 1000
    },
    {
      "epoch": 0.7,
      "eval_accuracy": 0.7794425087108013,
      "eval_loss": 0.4822468161582947,
      "eval_runtime": 1990.215,
      "eval_samples_per_second": 2.884,
      "eval_steps_per_second": 0.18,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 1435,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2106550388308416.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}