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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron is triggered by numeric literal tokens (e.g. integers or floating‐point numbers) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Ting
    -0.07
    ongs
    -0.06
    Preference
    -0.06
     그렇
    -0.06
    tx
    -0.06
     transitional
    -0.06
    Reporter
    -0.06
    _alarm
    -0.06
     TAR
    -0.06
    ’T
    -0.06
    POSITIVE LOGITS
     emergencies
    0.07
     مطرح
    0.07
     Ngb
    0.06
     don
    0.06
     acess
    0.06
     채
    0.06
     wrists
    0.06
    bilder
    0.06
    合わせ
    0.06
    -､
    0.06
    Act Density 0.014%

    No Known Activations