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

    Neuronpedia

    INDEX
    Explanations

    blow

    np_max-act · gemini-2.0-flash

    The neuron primarily fires on the verb “blow” when used in idiomatic phrases like “blow away,” “blow out,” or “blow up” as an intensifier.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    attice
    -0.08
     domác
    -0.07
    ert
    -0.07
    -heart
    -0.07
     يتم
    -0.07
    َد
    -0.07
     Kart
    -0.07
    urpose
    -0.07
    олуч
    -0.06
     USC
    -0.06
    POSITIVE LOGITS
     Blow
    0.13
     blow
    0.11
     blowing
    0.10
     blown
    0.10
     blows
    0.09
     blew
    0.09
    I
    0.09
     inflated
    0.08
    OW
    0.08
    low
    0.07
    Act Density 0.011%

    No Known Activations