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

    Neuronpedia

    INDEX
    Explanations

    &

    np_max-act · gemini-2.0-flash

    The neuron activates on floating-point numeric values (especially decimals) in URL query parameters.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     torture
    -0.08
    IQUE
    -0.07
    EXPR
    -0.07
    ]={
    -0.07
     fracture
    -0.07
    ему
    -0.07
    )--
    -0.07
    ']=='
    -0.07
     blows
    -0.06
    _alias
    -0.06
    POSITIVE LOGITS
    .getName
    0.08
    &t
    0.06
     jed
    0.06
    &R
    0.06
     snacks
    0.06
     dan
    0.06
     &&
    0.06
    Sher
    0.06
    fonts
    0.06
    notated
    0.06
    Act Density 0.004%

    No Known Activations