trainer_state.json

{
  "best_metric": 0.7206650376319885,
  "best_model_checkpoint": "./bert-multilingual-sdg-classification/checkpoint-2690",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 2690,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.929368029739777,
      "grad_norm": 6.992225646972656,
      "learning_rate": 8.316666666666668e-06,
      "loss": 2.1122,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.6814140148396195,
      "eval_loss": 1.0624698400497437,
      "eval_runtime": 10.5752,
      "eval_samples_per_second": 813.695,
      "eval_steps_per_second": 12.766,
      "step": 538
    },
    {
      "epoch": 1.858736059479554,
      "grad_norm": 6.2237749099731445,
      "learning_rate": 8.090909090909092e-06,
      "loss": 0.9564,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.7685561242660267,
      "eval_loss": 0.8072949051856995,
      "eval_runtime": 10.5617,
      "eval_samples_per_second": 814.737,
      "eval_steps_per_second": 12.782,
      "step": 1076
    },
    {
      "epoch": 2.7881040892193307,
      "grad_norm": 7.1777119636535645,
      "learning_rate": 5.698564593301436e-06,
      "loss": 0.7652,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.7885510737037653,
      "eval_loss": 0.7433095574378967,
      "eval_runtime": 10.6937,
      "eval_samples_per_second": 804.676,
      "eval_steps_per_second": 12.624,
      "step": 1614
    },
    {
      "epoch": 3.717472118959108,
      "grad_norm": 7.244228363037109,
      "learning_rate": 3.30622009569378e-06,
      "loss": 0.6619,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.7918831109586941,
      "eval_loss": 0.726136326789856,
      "eval_runtime": 10.5238,
      "eval_samples_per_second": 817.671,
      "eval_steps_per_second": 12.828,
      "step": 2152
    },
    {
      "epoch": 4.646840148698884,
      "grad_norm": 5.082164287567139,
      "learning_rate": 9.186602870813398e-07,
      "loss": 0.6038,
      "step": 2500
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.7925314519666281,
      "eval_loss": 0.7206650376319885,
      "eval_runtime": 10.5422,
      "eval_samples_per_second": 816.247,
      "eval_steps_per_second": 12.806,
      "step": 2690
    },
    {
      "epoch": 5.0,
      "step": 2690,
      "total_flos": 4.52871045169152e+16,
      "train_loss": 0.9877658092399512,
      "train_runtime": 691.6026,
      "train_samples_per_second": 248.842,
      "train_steps_per_second": 3.89
    }
  ],
  "logging_steps": 500,
  "max_steps": 2690,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.52871045169152e+16,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}