checkpoint-600/trainer_state.json

{
  "best_metric": 0.39216598868370056,
  "best_model_checkpoint": "/scratch/hlv8980/Attack_Benchmark/models/hyena/tf4/origin/checkpoint-600",
  "epoch": 2.0202020202020203,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34,
      "learning_rate": 2.8760984182776802e-05,
      "loss": 0.5992,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.615114235500879e-05,
      "loss": 0.4813,
      "step": 200
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.774,
      "eval_f1": 0.7713328260834371,
      "eval_loss": 0.48207539319992065,
      "eval_matthews_correlation": 0.5579338694412199,
      "eval_precision": 0.785067107786007,
      "eval_recall": 0.772997299729973,
      "eval_runtime": 0.1057,
      "eval_samples_per_second": 9462.679,
      "eval_steps_per_second": 151.403,
      "step": 200
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.3514938488576452e-05,
      "loss": 0.4431,
      "step": 300
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.087873462214411e-05,
      "loss": 0.377,
      "step": 400
    },
    {
      "epoch": 1.35,
      "eval_accuracy": 0.816,
      "eval_f1": 0.8159933757615274,
      "eval_loss": 0.427643358707428,
      "eval_matthews_correlation": 0.6320128653971173,
      "eval_precision": 0.815991263965056,
      "eval_recall": 0.816021602160216,
      "eval_runtime": 0.1039,
      "eval_samples_per_second": 9625.637,
      "eval_steps_per_second": 154.01,
      "step": 400
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.82688927943761e-05,
      "loss": 0.3443,
      "step": 500
    },
    {
      "epoch": 2.02,
      "learning_rate": 1.563268892794376e-05,
      "loss": 0.33,
      "step": 600
    },
    {
      "epoch": 2.02,
      "eval_accuracy": 0.824,
      "eval_f1": 0.8239746523499383,
      "eval_loss": 0.39216598868370056,
      "eval_matthews_correlation": 0.6479558982194922,
      "eval_precision": 0.8239935027265344,
      "eval_recall": 0.8239623962396239,
      "eval_runtime": 0.1031,
      "eval_samples_per_second": 9696.512,
      "eval_steps_per_second": 155.144,
      "step": 600
    }
  ],
  "max_steps": 1188,
  "num_train_epochs": 4,
  "total_flos": 76909184335872.0,
  "trial_name": null,
  "trial_params": null
}