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

    Neuronpedia

    INDEX
    Explanations

    names and relationships within family contexts

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

    Family relationships and names

    np_acts-logits-general · gemini-2.0-flash

    family members and deaths

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    AndEndTag
    -0.61
    amoan
    -0.57
    Rather
    -0.56
    |}{$
    -0.53
    RectangleBorder
    -0.53
    Instead
    -0.52
    rather
    -0.52
    Using
    -0.52
    vellous
    -0.51
    whatever
    -0.50
    POSITIVE LOGITS
     predeceased
    1.07
    deceased
    0.92
     whom
    0.88
     grandchildren
    0.85
     nieces
    0.80
     granddaughter
    0.80
     daughters
    0.80
     deceased
    0.80
     grandsons
    0.77
     nephews
    0.76
    Act Density 0.060%

    No Known Activations