run-5/checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.1446540880503147,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6535483870967742,
      "eval_loss": 0.28503966331481934,
      "eval_runtime": 1.889,
      "eval_samples_per_second": 1641.078,
      "eval_steps_per_second": 34.41,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.7276282906532288,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.4585,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8422580645161291,
      "eval_loss": 0.1152644231915474,
      "eval_runtime": 2.104,
      "eval_samples_per_second": 1473.356,
      "eval_steps_per_second": 30.893,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8938709677419355,
      "eval_loss": 0.06576532125473022,
      "eval_runtime": 3.2907,
      "eval_samples_per_second": 942.06,
      "eval_steps_per_second": 19.753,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.6005321741104126,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1354,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 260941334653608.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.45486992782879154,
    "num_train_epochs": 10,
    "temperature": 3
  }
}