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

    Neuronpedia

    INDEX
    Explanations

    Conditionals

    np_max-act · gemini-2.0-flash

    This neuron detects conditional markers—words like “if,” “when,” “even if,” and other tokens signaling a conditional clause.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Animation
    -0.07
    有的
    -0.07
    -neutral
    -0.07
     Nutzung
    -0.07
    цвет
    -0.06
    -ranked
    -0.06
     {}↵
    -0.06
    .path
    -0.06
     TypeName
    -0.06
     fourn
    -0.06
    POSITIVE LOGITS
    34
    0.07
    annot
    0.07
    باب
    0.06
    ���
    0.06
     nelze
    0.06
     ’
    0.06
     Supplement
    0.06
    _css
    0.06
     vern
    0.06
    ONGLONG
    0.06
    Act Density 0.030%

    No Known Activations