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

    Neuronpedia

    INDEX
    Explanations

    Descriptive

    np_max-act · gemini-2.0-flash

    This neuron activates on descriptive adjectives and adverbs (i.e. qualifiers of degree or traits).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     halluc
    -0.07
     blow
    -0.07
    utex
    -0.06
    包括
    -0.06
    tır
    -0.06
    _sec
    -0.06
    �认
    -0.06
     reply
    -0.06
     Dirk
    -0.06
     Blizzard
    -0.06
    POSITIVE LOGITS
     البحر
    0.06
    dirname
    0.06
     arts
    0.06
     thân
    0.06
     traveler
    0.06
    .mar
    0.06
    Pow
    0.06
     straw
    0.06
     무
    0.05
     Angiosper
    0.05
    Act Density 0.064%

    No Known Activations