run-1/checkpoint-1070/trainer_state.json

{
  "best_metric": 0.0,
  "best_model_checkpoint": "hw-01-finetuned-cola/run-1/checkpoint-535",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1070,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.93,
      "learning_rate": 6.523832453871098e-07,
      "loss": 0.6196,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6081321835517883,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.7503,
      "eval_samples_per_second": 1390.123,
      "eval_steps_per_second": 87.966,
      "step": 535
    },
    {
      "epoch": 1.87,
      "learning_rate": 8.011724066157488e-08,
      "loss": 0.5922,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5986050963401794,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.4026,
      "eval_samples_per_second": 743.6,
      "eval_steps_per_second": 47.054,
      "step": 1070
    }
  ],
  "logging_steps": 500,
  "max_steps": 1070,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 85728657313800.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.2246492501126448e-06,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 16,
    "seed": 14
  }
}