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

    Neuronpedia

    INDEX
    Explanations

    error correction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ת
    0.48
    ּ
    0.45
    వారు
    0.43
    ุ
    0.42
     dodge
    0.42
    ern
    0.40
    ‌
    0.40
    ד
    0.40
    ﺩ
    0.40
    ತ್ಮ
    0.39
    POSITIVE LOGITS
     Purchases
    0.51
     évident
    0.47
     références
    0.46
     merda
    0.45
     Blur
    0.44
     correctes
    0.44
    よね
    0.44
     wanting
    0.43
     multe
    0.43
     blindness
    0.43
    Act Density 0.003%

    No Known Activations