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

    Neuronpedia

    INDEX
    Explanations

    faults and negatives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    кция
    0.49
     परफॉर्म
    0.47
    स्तिष्क
    0.46
     нём
    0.46
    тря
    0.43
    чем
    0.43
    кции
    0.42
    effects
    0.42
     mischiev
    0.42
     неожидан
    0.41
    POSITIVE LOGITS
    :
    0.55
    *
    0.55
     caravan
    0.49
     alle
    0.45
    İ
    0.45
     testament
    0.44
    라
    0.42
    ;
    0.42
    #
    0.42
     jay
    0.42
    Act Density 0.000%

    No Known Activations