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

    Neuronpedia

    INDEX
    Explanations

    characteristics related to physical appearance and identity descriptors

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

    physical descriptions of people

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    inSlope
    -0.45
    ņas
    -0.44
     vandens
    -0.44
     stabilisation
    -0.40
     način
    -0.40
    Diweddarwch
    -0.39
     título
    -0.39
     tiek
    -0.37
    lgari
    -0.37
     ferons
    -0.36
    POSITIVE LOGITS
     blonde
    0.52
     haired
    0.51
     grey
    0.50
     bearded
    0.49
     slim
    0.47
    Datuak
    0.47
     masked
    0.46
     gray
    0.46
     brunette
    0.46
    AccessorTable
    0.46
    Act Density 0.038%

    No Known Activations