run-17/checkpoint-1500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.716981132075472,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6283870967741936,
      "eval_loss": 0.234601229429245,
      "eval_runtime": 5.418,
      "eval_samples_per_second": 572.171,
      "eval_steps_per_second": 11.997,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.3718,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8419354838709677,
      "eval_loss": 0.10685674846172333,
      "eval_runtime": 5.5263,
      "eval_samples_per_second": 560.955,
      "eval_steps_per_second": 11.762,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8980645161290323,
      "eval_loss": 0.06742739677429199,
      "eval_runtime": 5.4202,
      "eval_samples_per_second": 571.937,
      "eval_steps_per_second": 11.992,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1236,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9070967741935484,
      "eval_loss": 0.04948155954480171,
      "eval_runtime": 5.4608,
      "eval_samples_per_second": 567.684,
      "eval_steps_per_second": 11.903,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0748,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2862,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 389479376069112.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7441510056067615,
    "num_train_epochs": 9,
    "temperature": 5
  }
}