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

    Neuronpedia

    INDEX
    Explanations

    splitting into two

    np_max-act · gemini-2.0-flash

    This neuron activates on technical code and algorithm terms describing data structures and partition steps—tokens like “array,” “two,” “sub-array,” and “pointers.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (scores
    -0.07
    _WIDGET
    -0.07
    りと
    -0.06
    _cou
    -0.06
    shots
    -0.06
    Sector
    -0.06
    (editor
    -0.06
    Filters
    -0.06
    *angstrom
    -0.06
    /pre
    -0.06
    POSITIVE LOGITS
     dining
    0.07
    itical
    0.07
    Fa
    0.06
     Ikea
    0.06
     taxed
    0.06
    電話
    0.06
    usunda
    0.06
     advertised
    0.06
    shima
    0.06
    ugas
    0.06
    Act Density 0.009%

    No Known Activations