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

    Neuronpedia

    INDEX
    Explanations

    multilingual and diverse concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nicht
    1.13
     auch
    1.02
     allem
    0.96
     presque
    0.95
     asumir
    0.95
     menand
    0.95
     mwaka
    0.94
     donde
    0.93
     etwas
    0.93
     not
    0.92
    POSITIVE LOGITS
    ać
    1.01
    iculum
    1.00
    vation
    0.99
    Ди
    0.98
     Від
    0.98
    ה
    0.97
    ύ
    0.96
    𝘿
    0.94
    dling
    0.93
    localhost
    0.93
    Act Density 0.011%

    No Known Activations