i_context/checkpoint-586/trainer_state.json

{
  "best_metric": 1.0713884830474854,
  "best_model_checkpoint": "clean_output/bert-large-cased/i_context/checkpoint-586",
  "epoch": 2.0,
  "global_step": 586,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.505982905982906,
      "eval_loss": 1.1140286922454834,
      "eval_runtime": 10.4473,
      "eval_samples_per_second": 223.981,
      "eval_steps_per_second": 28.045,
      "step": 293
    },
    {
      "epoch": 1.71,
      "learning_rate": 7.680097680097681e-06,
      "loss": 1.1604,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5525641025641026,
      "eval_loss": 1.0713884830474854,
      "eval_runtime": 10.4432,
      "eval_samples_per_second": 224.069,
      "eval_steps_per_second": 28.056,
      "step": 586
    }
  ],
  "max_steps": 1758,
  "num_train_epochs": 6,
  "total_flos": 6812609916643200.0,
  "trial_name": null,
  "trial_params": null
}