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

    Neuronpedia

    INDEX
    Explanations

    expressions of family dynamics and protective relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    protective parental concerns

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

    mentions of parental concern/protectiveness or parents trying to control/limit a child's behavior.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    piram
    -0.43
     ProtoMessage
    -0.41
    windowFixed
    -0.40
     lenker
    -0.38
    脚注の使い方
    -0.38
    cửa
    -0.37
     hospod
    -0.36
     aikaa
    -0.36
     rétro
    -0.36
     oprot
    -0.36
    POSITIVE LOGITS
     safety
    0.74
    Safety
    0.68
     Safety
    0.66
     protective
    0.62
    safety
    0.60
    protective
    0.58
     security
    0.57
    Protective
    0.56
     risk
    0.53
    afety
    0.53
    Act Density 0.055%

    No Known Activations