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

    Neuronpedia

    INDEX
    Explanations

    negative sentiment, Taiwanese classes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Й
    0.50
     ಸಾವಿರ
    0.49
    В
    0.46
    ailing
    0.46
     окружа
    0.46
    ለያዩ
    0.46
     рекоменду
    0.45
    पिछले
    0.42
    й
    0.42
    Если
    0.42
    POSITIVE LOGITS
    ’
    0.46
     Kirche
    0.46
     Là
    0.46
     queer
    0.46
     chem
    0.45
     Brave
    0.44
     Que
    0.44
     Según
    0.44
     Bedürfnisse
    0.43
     Prakt
    0.43
    Act Density 0.004%

    No Known Activations