test_results.json

{
    "epoch": 10.0,
    "eval_gate1_accuracy": 0.9251295336787565,
    "eval_gate1_f1": 0.9199371922438249,
    "eval_gate1_precision": 0.920966070966071,
    "eval_gate1_recall": 0.9189513092953367,
    "eval_gate1_support": 3860,
    "eval_gate2_accuracy": 0.8709844559585492,
    "eval_gate2_f1": 0.8487797208181732,
    "eval_gate2_precision": 0.8500529357712848,
    "eval_gate2_recall": 0.8532136788618205,
    "eval_gate2_support": 3860,
    "eval_gate3_accuracy": 0.8920086393088553,
    "eval_gate3_f1": 0.6992827001481418,
    "eval_gate3_precision": 0.715591871957983,
    "eval_gate3_recall": 0.7063340262769819,
    "eval_gate3_support": 6945,
    "eval_gate4_span_FN": 196,
    "eval_gate4_span_FP": 204,
    "eval_gate4_span_TP": 815,
    "eval_gate4_span_f1": 0.8029556600167974,
    "eval_gate4_span_precision": 0.7998037291383728,
    "eval_gate4_span_recall": 0.8061325420296129,
    "eval_loss": NaN,
    "eval_runtime": 10.4274,
    "eval_samples_per_second": 1133.841,
    "eval_steps_per_second": 35.483
}