Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    woman words

    np_acts-logits-general · gemini-2.5-flash-lite

    references to female individuals, especially when the subject is a woman or girl and is referred to with feminine pronouns or names.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     способен
    0.57
     сам
    0.56
     sám
    0.56
    Escolhido
    0.55
     आला
    0.54
     равен
    0.52
     نفسه
    0.52
     который
    0.51
     должен
    0.50
     शकतो
    0.50
    POSITIVE LOGITS
     herself
    1.44
     woman
    1.05
     actresses
    1.02
     girl
    1.01
     businesswoman
    1.01
     heroine
    1.00
     women
    0.97
     xinh
    0.97
     femenina
    0.97
     نفسها
    0.96
    Act Density 0.431%

    No Known Activations