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

    Neuronpedia

    INDEX
    Explanations

    bash scripts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     isIn
    0.59
    빨
    0.59
    valuate
    0.58
    shallow
    0.57
    clusion
    0.57
    嗇
    0.57
     pronouns
    0.56
    📍
    0.56
     cui
    0.55
    சையில்
    0.55
    POSITIVE LOGITS
    fileExistsAtPath
    0.58
     ordinarily
    0.57
     پیچ
    0.56
    ת
    0.54
    irled
    0.53
    टु
    0.53
     cominci
    0.53
     gitu
    0.53
     Firmware
    0.53
    ет
    0.52
    Act Density 0.042%

    No Known Activations