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

    Neuronpedia

    INDEX
    Explanations

    Salad discussions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     κατά
    0.74
    ม
    0.70
     subsid
    0.70
     langt
    0.68
     kuten
    0.68
     bruk
    0.64
    áno
    0.64
     sådan
    0.63
     públicos
    0.62
     zeker
    0.61
    POSITIVE LOGITS
    та
    1.02
    бург
    0.96
    ية
    0.94
    at
    0.90
    ot
    0.88
    shirts
    0.85
    的产品
    0.85
    lara
    0.83
    salad
    0.82
    EM
    0.80
    Act Density 0.002%

    No Known Activations