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

    Neuronpedia

    INDEX
    Explanations

    a role

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     상당히
    0.52
     довольно
    0.49
     quite
    0.44
     काफी
    0.43
     досить
    0.43
    かなり
    0.41
     весьма
    0.41
     dość
    0.40
    꽤
    0.39
     cukup
    0.38
    POSITIVE LOGITS
     solche
    0.39
    versive
    0.39
     desired
    0.36
     multitude
    0.36
     solchen
    0.35
     quelconque
    0.33
     grudge
    0.33
     vagina
    0.31
     thing
    0.31
    uras
    0.30
    Act Density 0.556%

    No Known Activations