QA-DeBERTa-v3-large-diff-binary / test_results.json
saiteki-kai's picture
End of training
52d26e4 verified
{
"test_accuracy": 0.6766379207090669,
"test_animal_abuse/accuracy": 0.9931578632171517,
"test_animal_abuse/f1": 0.7676664972038637,
"test_animal_abuse/fpr": 0.005037019055710637,
"test_animal_abuse/precision": 0.6945722171113156,
"test_animal_abuse/recall": 0.8579545454545454,
"test_animal_abuse/threshold": 0.5,
"test_child_abuse/accuracy": 0.9963169241825368,
"test_child_abuse/f1": 0.6070287539936102,
"test_child_abuse/fpr": 0.0012642986152919908,
"test_child_abuse/precision": 0.6934306569343066,
"test_child_abuse/recall": 0.5397727272727273,
"test_child_abuse/threshold": 0.5,
"test_controversial_topics,politics/accuracy": 0.9715085638998683,
"test_controversial_topics,politics/f1": 0.45519610649871173,
"test_controversial_topics,politics/fpr": 0.011352768779885837,
"test_controversial_topics,politics/precision": 0.519268451992162,
"test_controversial_topics,politics/recall": 0.40519877675840976,
"test_controversial_topics,politics/threshold": 0.5,
"test_discrimination,stereotype,injustice/accuracy": 0.9543957360162894,
"test_discrimination,stereotype,injustice/f1": 0.7218772826880935,
"test_discrimination,stereotype,injustice/fpr": 0.023755877742946668,
"test_discrimination,stereotype,injustice/precision": 0.7309541420118343,
"test_discrimination,stereotype,injustice/recall": 0.713023088023088,
"test_discrimination,stereotype,injustice/threshold": 0.5,
"test_drug_abuse,weapons,banned_substance/accuracy": 0.9734848484848485,
"test_drug_abuse,weapons,banned_substance/f1": 0.7634566582075598,
"test_drug_abuse,weapons,banned_substance/fpr": 0.014630821418381233,
"test_drug_abuse,weapons,banned_substance/precision": 0.755884686590849,
"test_drug_abuse,weapons,banned_substance/recall": 0.7711818672423097,
"test_drug_abuse,weapons,banned_substance/threshold": 0.5,
"test_financial_crime,property_crime,theft/accuracy": 0.9579590370104204,
"test_financial_crime,property_crime,theft/f1": 0.7996575342465754,
"test_financial_crime,property_crime,theft/fpr": 0.0271945610877824,
"test_financial_crime,property_crime,theft/precision": 0.7744610281923715,
"test_financial_crime,property_crime,theft/recall": 0.8265486725663717,
"test_financial_crime,property_crime,theft/threshold": 0.5,
"test_flagged/accuracy": 0.8487393699844292,
"test_flagged/aucpr": 0.9047138471115324,
"test_flagged/f1": 0.8630417395312266,
"test_flagged/fpr": 0.15492622560685337,
"test_flagged/precision": 0.8747698480309983,
"test_flagged/recall": 0.8516239499170635,
"test_hate_speech,offensive_language/accuracy": 0.9497844053180021,
"test_hate_speech,offensive_language/f1": 0.686307519640853,
"test_hate_speech,offensive_language/fpr": 0.017700424021299645,
"test_hate_speech,offensive_language/precision": 0.7730720606826802,
"test_hate_speech,offensive_language/recall": 0.6170534813319879,
"test_hate_speech,offensive_language/threshold": 0.5,
"test_loss": 0.08679678291082382,
"test_macro_f1": 0.6252685023565255,
"test_macro_precision": 0.6847558918566842,
"test_macro_recall": 0.6093259557964672,
"test_micro_f1": 0.7468949012290593,
"test_micro_precision": 0.7705170480223624,
"test_micro_recall": 0.724678064292686,
"test_misinformation_regarding_ethics,laws_and_safety/accuracy": 0.987633249490957,
"test_misinformation_regarding_ethics,laws_and_safety/f1": 0.05707762557077625,
"test_misinformation_regarding_ethics,laws_and_safety/fpr": 0.0005304959379168174,
"test_misinformation_regarding_ethics,laws_and_safety/precision": 0.4166666666666667,
"test_misinformation_regarding_ethics,laws_and_safety/recall": 0.030637254901960783,
"test_misinformation_regarding_ethics,laws_and_safety/threshold": 0.5,
"test_non_violent_unethical_behavior/accuracy": 0.8818571086357647,
"test_non_violent_unethical_behavior/f1": 0.6774575924790517,
"test_non_violent_unethical_behavior/fpr": 0.050999362507968556,
"test_non_violent_unethical_behavior/precision": 0.7528845280276187,
"test_non_violent_unethical_behavior/recall": 0.6157675731906672,
"test_non_violent_unethical_behavior/threshold": 0.5,
"test_privacy_violation/accuracy": 0.9793687866810397,
"test_privacy_violation/f1": 0.8086111111111111,
"test_privacy_violation/fpr": 0.01268442169778534,
"test_privacy_violation/precision": 0.7837910608508347,
"test_privacy_violation/recall": 0.8350545037292025,
"test_privacy_violation/threshold": 0.5,
"test_runtime": 664.4818,
"test_samples_per_second": 100.517,
"test_self_harm/accuracy": 0.9963917834471194,
"test_self_harm/f1": 0.714792899408284,
"test_self_harm/fpr": 0.0011910505367265691,
"test_self_harm/precision": 0.7926509186351706,
"test_self_harm/recall": 0.6508620689655172,
"test_self_harm/threshold": 0.5,
"test_sexually_explicit,adult_content/accuracy": 0.9838902862618277,
"test_sexually_explicit,adult_content/f1": 0.6515544041450777,
"test_sexually_explicit,adult_content/fpr": 0.009186954524575089,
"test_sexually_explicit,adult_content/precision": 0.6264009962640099,
"test_sexually_explicit,adult_content/recall": 0.6788124156545209,
"test_sexually_explicit,adult_content/threshold": 0.5,
"test_steps_per_second": 1.571,
"test_terrorism,organized_crime/accuracy": 0.9913911845730028,
"test_terrorism,organized_crime/f1": 0.1958041958041958,
"test_terrorism,organized_crime/fpr": 0.001343680173923544,
"test_terrorism,organized_crime/precision": 0.44025157232704404,
"test_terrorism,organized_crime/recall": 0.12589928057553956,
"test_terrorism,organized_crime/threshold": 0.5,
"test_violence,aiding_and_abetting,incitement/accuracy": 0.9157533836387591,
"test_violence,aiding_and_abetting,incitement/f1": 0.8472708519935944,
"test_violence,aiding_and_abetting,incitement/fpr": 0.06457640343312376,
"test_violence,aiding_and_abetting,incitement/precision": 0.8322934997067136,
"test_violence,aiding_and_abetting,incitement/recall": 0.8627971254836927,
"test_violence,aiding_and_abetting,incitement/threshold": 0.5
}