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

    Neuronpedia

    INDEX
    Explanations

    pruning or retirement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     liability
    0.44
     wären
    0.41
    energy
    0.40
     Liability
    0.39
     intric
    0.39
     molybdenum
    0.39
    EUR
    0.38
     mood
    0.38
    mood
    0.38
    箬
    0.37
    POSITIVE LOGITS
     Silverman
    0.48
     সংখ্যাগ
    0.40
    𝘁
    0.39
    螌
    0.39
     மேலும்
    0.39
     their
    0.38
     ಸಾಮಾನ್ಯವಾಗಿ
    0.38
    暊
    0.38
    每个
    0.37
     திரும்பி
    0.37
    Act Density 0.000%

    No Known Activations