run-2/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": 0.594516129032258,
      "eval_loss": 0.20276567339897156,
      "eval_runtime": 5.4162,
      "eval_samples_per_second": 572.36,
      "eval_steps_per_second": 12.001,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.535661518573761,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.3272,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8270967741935484,
      "eval_loss": 0.09640489518642426,
      "eval_runtime": 6.1013,
      "eval_samples_per_second": 508.09,
      "eval_steps_per_second": 10.653,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8841935483870967,
      "eval_loss": 0.06340694427490234,
      "eval_runtime": 5.5585,
      "eval_samples_per_second": 557.704,
      "eval_steps_per_second": 11.694,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.4495888650417328,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1121,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 260941334653608.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.07863039618542955,
    "num_train_epochs": 10,
    "temperature": 9
  }
}