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

    Neuronpedia

    INDEX
    Explanations

    base conversions and math

    np_max-act · gemini-2.0-flash

    The neuron is looking for the numeric result token (the answer) in a “In base N, what is …?” arithmetic question.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    but
    -0.07
    MarshalAs
    -0.06
     Pe
    -0.06
     advoc
    -0.06
    igeria
    -0.06
     summers
    -0.06
    iscing
    -0.06
     Glad
    -0.06
    ‘
    -0.06
    Sw
    -0.06
    POSITIVE LOGITS
     pylab
    0.07
    /svg
    0.07
     */,
    0.07
    -rounded
    0.06
     HTMLElement
    0.06
    ilebilir
    0.06
    ocurrency
    0.06
     nghiệm
    0.06
    TB
    0.06
    }*/↵↵
    0.06
    Act Density 0.010%

    No Known Activations