{
  "best_metric": 0.9577080491132333,
  "best_model_checkpoint": "/content/drive/MyDrive/mbert/2025_02_02/t1/checkpoint-2418",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2418,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.461277961730957,
      "learning_rate": 4.2159929973664246e-05,
      "loss": 0.0875,
      "step": 806
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9399813895781638,
      "eval_f1": 0.9364009860312243,
      "eval_loss": 0.27949896454811096,
      "eval_precision": 0.9041574103459219,
      "eval_recall": 0.9710293115201091,
      "eval_runtime": 34.1792,
      "eval_samples_per_second": 188.653,
      "eval_steps_per_second": 5.91,
      "step": 806
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5872113108634949,
      "learning_rate": 1.5100460852686277e-05,
      "loss": 0.0357,
      "step": 1612
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9561104218362283,
      "eval_f1": 0.9519769217715934,
      "eval_loss": 0.19011780619621277,
      "eval_precision": 0.9479553903345725,
      "eval_recall": 0.9560327198364008,
      "eval_runtime": 34.1574,
      "eval_samples_per_second": 188.773,
      "eval_steps_per_second": 5.914,
      "step": 1612
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.051353733986616135,
      "learning_rate": 0.0,
      "loss": 0.0036,
      "step": 2418
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9615384615384616,
      "eval_f1": 0.9577080491132333,
      "eval_loss": 0.20939238369464874,
      "eval_precision": 0.9583617747440273,
      "eval_recall": 0.9570552147239264,
      "eval_runtime": 34.5385,
      "eval_samples_per_second": 186.69,
      "eval_steps_per_second": 5.849,
      "step": 2418
    }
  ],
  "logging_steps": 500,
  "max_steps": 2418,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.272477999836365e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}