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 activates on numeric literals in source code, i.e. tokens representing numbers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    cursor
    -0.07
    lahoma
    -0.06
    ляється
    -0.06
    _Left
    -0.06
     дом
    -0.06
     облі
    -0.06
    _append
    -0.06
     Annunci
    -0.06
    ,ev
    -0.06
    Left
    -0.06
    POSITIVE LOGITS
     pot
    0.06
    aped
    0.06
    -paced
    0.06
        ↵    ↵    ↵    ↵
    0.06
    guns
    0.06
    (torch
    0.06
    Deploy
    0.06
     backs
    0.06
     значительно
    0.06
    强
    0.06
    Act Density 0.006%

    No Known Activations