trainer_state.json

{
  "best_metric": 1.2963756322860718,
  "best_model_checkpoint": "bert-propaganda-ner/checkpoint-1750",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.57,
      "grad_norm": 3.605520486831665,
      "learning_rate": 4.714285714285714e-05,
      "loss": 1.2802,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7176097303014278,
      "eval_f1": 0.09410755148741418,
      "eval_loss": 1.2981841564178467,
      "eval_precision": 0.28986784140969163,
      "eval_recall": 0.05617210175857948,
      "eval_runtime": 4.2298,
      "eval_samples_per_second": 217.743,
      "eval_steps_per_second": 27.425,
      "step": 875
    },
    {
      "epoch": 1.14,
      "grad_norm": 2.260868787765503,
      "learning_rate": 4.428571428571428e-05,
      "loss": 1.1118,
      "step": 1000
    },
    {
      "epoch": 1.71,
      "grad_norm": 7.614594459533691,
      "learning_rate": 4.1428571428571437e-05,
      "loss": 0.9205,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6880236021041498,
      "eval_f1": 0.16797458893871453,
      "eval_loss": 1.2963756322860718,
      "eval_precision": 0.18547555188776563,
      "eval_recall": 0.15349154857435549,
      "eval_runtime": 4.3987,
      "eval_samples_per_second": 209.378,
      "eval_steps_per_second": 26.371,
      "step": 1750
    }
  ],
  "logging_steps": 500,
  "max_steps": 8750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 553357099061862.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}