eval_results.json

{
    "eval_HasAns_exact": 22.97081167532987,
    "eval_HasAns_f1": 29.82730501295977,
    "eval_HasAns_total": 5002,
    "eval_NoAns_exact": 61.644657863145255,
    "eval_NoAns_f1": 61.644657863145255,
    "eval_NoAns_total": 4998,
    "eval_best_exact": 50.0,
    "eval_best_exact_thresh": 0.0,
    "eval_best_f1": 50.00666666666667,
    "eval_best_f1_thresh": 0.0,
    "eval_exact": 42.3,
    "eval_f1": 45.72961796748245,
    "eval_runtime": 188.1784,
    "eval_samples": 10000,
    "eval_samples_per_second": 53.141,
    "eval_steps_per_second": 6.643,
    "eval_total": 10000
}