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

    Neuronpedia

    INDEX
    Explanations

    Foreign language

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on Thai‐language text, i.e. tokens composed of Thai script characters.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _STENCIL
    -0.08
    _books
    -0.07
    .AUTO
    -0.07
     interoper
    -0.06
    tout
    -0.06
     marching
    -0.06
     REAL
    -0.06
     donn
    -0.06
    eman
    -0.06
    .writeString
    -0.06
    POSITIVE LOGITS
    ิ่
    0.07
    ือ
    0.07
     vague
    0.07
    Reporter
    0.06
    ื่
    0.06
     منظ
    0.06
     Zurich
    0.06
     wallet
    0.06
    ผม
    0.06
    seat
    0.06
    Act Density 0.016%

    No Known Activations