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

    Neuronpedia

    INDEX
    Explanations

    Quantities and multiplicity

    np_max-act · gemini-2.0-flash

    the presence of explicit numeric quantifiers (words indicating counts or plurality such as “two,” “multiple,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Sweet
    -0.07
     Lives
    -0.07
     pla
    -0.07
    álních
    -0.07
     Curtain
    -0.06
    Asia
    -0.06
     Ultimate
    -0.06
    吃
    -0.06
     downward
    -0.06
     belir
    -0.06
    POSITIVE LOGITS
    și
    0.06
    \-
    0.06
    hack
    0.06
    icers
    0.06
    _MANAGER
    0.06
    fgang
    0.06
    -modules
    0.06
    TouchUpInside
    0.06
     fileId
    0.06
    obby
    0.06
    Act Density 0.021%

    No Known Activations