trainer_state.json

{
  "best_metric": 0.7596717467760844,
  "best_model_checkpoint": "deberta-classification-transcripts/checkpoint-4325",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4325,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.58,
      "learning_rate": 8.84393063583815e-06,
      "loss": 0.4483,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.09210526315789473,
      "eval_f1": 0.6887254901960784,
      "eval_loss": 0.2805722951889038,
      "eval_roc_auc": 0.7836829389177116,
      "eval_runtime": 1.7776,
      "eval_samples_per_second": 42.755,
      "eval_steps_per_second": 42.755,
      "step": 865
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.687861271676302e-06,
      "loss": 0.339,
      "step": 1000
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.531791907514451e-06,
      "loss": 0.2866,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.10526315789473684,
      "eval_f1": 0.7407407407407407,
      "eval_loss": 0.22884871065616608,
      "eval_roc_auc": 0.8162958199606408,
      "eval_runtime": 1.7174,
      "eval_samples_per_second": 44.252,
      "eval_steps_per_second": 44.252,
      "step": 1730
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.375722543352601e-06,
      "loss": 0.265,
      "step": 2000
    },
    {
      "epoch": 2.89,
      "learning_rate": 4.219653179190752e-06,
      "loss": 0.2368,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.06578947368421052,
      "eval_f1": 0.7450058754406581,
      "eval_loss": 0.20490357279777527,
      "eval_roc_auc": 0.8221279142285448,
      "eval_runtime": 1.7181,
      "eval_samples_per_second": 44.235,
      "eval_steps_per_second": 44.235,
      "step": 2595
    },
    {
      "epoch": 3.47,
      "learning_rate": 3.063583815028902e-06,
      "loss": 0.2351,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.09210526315789473,
      "eval_f1": 0.7520759193357057,
      "eval_loss": 0.19738489389419556,
      "eval_roc_auc": 0.8239552326387778,
      "eval_runtime": 1.7202,
      "eval_samples_per_second": 44.18,
      "eval_steps_per_second": 44.18,
      "step": 3460
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.907514450867052e-06,
      "loss": 0.2232,
      "step": 3500
    },
    {
      "epoch": 4.62,
      "learning_rate": 7.514450867052025e-07,
      "loss": 0.2127,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.10526315789473684,
      "eval_f1": 0.7596717467760844,
      "eval_loss": 0.19270123541355133,
      "eval_roc_auc": 0.830700986111798,
      "eval_runtime": 1.7179,
      "eval_samples_per_second": 44.241,
      "eval_steps_per_second": 44.241,
      "step": 4325
    }
  ],
  "logging_steps": 500,
  "max_steps": 4325,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1138312891315200.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}