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

    Neuronpedia

    INDEX
    Explanations

    use

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens in sentences where the author states they can’t or must not use some feature or must set/use something under a constraint—that is, it spots expressions of usage restrictions or required “use” actions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Publisher
    -0.07
    ……。
    -0.07
    اهی
    -0.07
     trout
    -0.07
     caves
    -0.07
     preventive
    -0.07
     findBy
    -0.07
    (klass
    -0.07
    Gil
    -0.06
     Porter
    -0.06
    POSITIVE LOGITS
     noe
    0.06
     ticking
    0.06
    equality
    0.06
     Lever
    0.06
     stab
    0.06
     VIA
    0.06
    edia
    0.06
    ीड
    0.06
    liable
    0.06
     onFocus
    0.05
    Act Density 0.055%

    No Known Activations