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

    Neuronpedia

    INDEX
    Explanations

    references to male characters or subjects, particularly using third-person pronouns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ssz
    -0.67
     Noch
    -0.60
     aig
    -0.59
     inva
    -0.59
    acd
    -0.56
     comod
    -0.55
    setShow
    -0.54
    URS
    -0.53
     entro
    -0.51
     similar
    -0.51
    POSITIVE LOGITS
     himself
    1.38
     he
    1.35
    himself
    1.24
    He
    1.22
     she
    1.21
     He
    1.20
    She
    1.12
    she
    1.12
    hehe
    1.11
     Himself
    1.10
    Act Density 0.348%

    No Known Activations