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

    Neuronpedia

    INDEX
    Explanations

    common words

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric and enumeration tokens (e.g. digits, decimal numbers, list indices, and similar number‐related fragments).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    大量
    -0.07
     Hosp
    -0.06
    CELL
    -0.06
     solving
    -0.06
    job
    -0.06
     openFileDialog
    -0.06
    W
    -0.06
     rad
    -0.06
     NORMAL
    -0.06
     rede
    -0.06
    POSITIVE LOGITS
    _controls
    0.08
    ीसर
    0.07
     کی
    0.06
    лом
    0.06
     brasile
    0.06
    icamente
    0.06
     MouseEvent
    0.06
    Fort
    0.06
     hton
    0.06
    �s
    0.06
    Act Density 0.066%

    No Known Activations