trainer_state.json

{
  "best_metric": 0.1469496637582779,
  "best_model_checkpoint": "gbert-base-definition_ext/checkpoint-1257",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 5028,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.39777247414478917,
      "grad_norm": 10.905921936035156,
      "learning_rate": 1.9204455051710422e-05,
      "loss": 0.3152,
      "step": 500
    },
    {
      "epoch": 0.7955449482895783,
      "grad_norm": 18.412729263305664,
      "learning_rate": 1.8408910103420846e-05,
      "loss": 0.2665,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9606635071090047,
      "eval_loss": 0.1469496637582779,
      "eval_runtime": 13.7605,
      "eval_samples_per_second": 153.337,
      "eval_steps_per_second": 19.185,
      "step": 1257
    },
    {
      "epoch": 1.1933174224343674,
      "grad_norm": 0.68099045753479,
      "learning_rate": 1.7613365155131266e-05,
      "loss": 0.2253,
      "step": 1500
    },
    {
      "epoch": 1.5910898965791567,
      "grad_norm": 0.110905721783638,
      "learning_rate": 1.681782020684169e-05,
      "loss": 0.2098,
      "step": 2000
    },
    {
      "epoch": 1.988862370723946,
      "grad_norm": 1.256501317024231,
      "learning_rate": 1.602227525855211e-05,
      "loss": 0.2219,
      "step": 2500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9606635071090047,
      "eval_loss": 0.15402448177337646,
      "eval_runtime": 13.6125,
      "eval_samples_per_second": 155.004,
      "eval_steps_per_second": 19.394,
      "step": 2514
    },
    {
      "epoch": 2.386634844868735,
      "grad_norm": 0.1358814239501953,
      "learning_rate": 1.5226730310262532e-05,
      "loss": 0.1629,
      "step": 3000
    },
    {
      "epoch": 2.7844073190135243,
      "grad_norm": 0.28879252076148987,
      "learning_rate": 1.4431185361972953e-05,
      "loss": 0.1723,
      "step": 3500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9587677725118483,
      "eval_loss": 0.2121916264295578,
      "eval_runtime": 13.5857,
      "eval_samples_per_second": 155.311,
      "eval_steps_per_second": 19.432,
      "step": 3771
    },
    {
      "epoch": 3.1821797931583133,
      "grad_norm": 20.142990112304688,
      "learning_rate": 1.3635640413683375e-05,
      "loss": 0.171,
      "step": 4000
    },
    {
      "epoch": 3.579952267303103,
      "grad_norm": 8.473916053771973,
      "learning_rate": 1.2840095465393797e-05,
      "loss": 0.1613,
      "step": 4500
    },
    {
      "epoch": 3.977724741447892,
      "grad_norm": 2.436373472213745,
      "learning_rate": 1.2044550517104217e-05,
      "loss": 0.1458,
      "step": 5000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9630331753554502,
      "eval_loss": 0.17300711572170258,
      "eval_runtime": 13.5779,
      "eval_samples_per_second": 155.4,
      "eval_steps_per_second": 19.443,
      "step": 5028
    }
  ],
  "logging_steps": 500,
  "max_steps": 12570,
  "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": 8942011133037720.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}