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

    Neuronpedia

    INDEX
    Explanations

    sexual attributes and conformity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     truly
    0.66
     genuine
    0.63
     genuinely
    0.62
    Truly
    0.55
     véritable
    0.50
     echte
    0.49
    真实
    0.48
    真正的
    0.48
     really
    0.47
     உண்மைய
    0.47
    POSITIVE LOGITS
     якобы
    0.80
    conform
    0.71
     möglichst
    0.68
     supposedly
    0.68
     conform
    0.66
     superficially
    0.65
     Conform
    0.62
     conforms
    0.61
     conformity
    0.60
     marketable
    0.59
    Act Density 0.085%

    No Known Activations