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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash

    The neuron flags (activates on) tokens that represent command-line options (i.e. the dash-prefixed “-x” flags).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _closure
    -0.07
     stature
    -0.07
    _cert
    -0.07
    -label
    -0.07
    initWith
    -0.06
    _books
    -0.06
    AppName
    -0.06
     Policies
    -0.06
     Lyft
    -0.06
    /TT
    -0.06
    POSITIVE LOGITS
     Puppy
    0.07
    Everything
    0.07
     jinak
    0.07
    全面
    0.07
    _Is
    0.07
    台灣
    0.06
     refreshed
    0.06
    现在
    0.06
    _Sh
    0.06
     anytime
    0.06
    Act Density 0.010%

    No Known Activations