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

    Neuronpedia

    INDEX
    Explanations

    punctuation, especially periods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     poslov
    0.63
     kuchh
    0.51
     keepsake
    0.51
     multimillion
    0.50
     organi
    0.50
     odborn
    0.50
    㬶
    0.49
     convivial
    0.49
     quelquefois
    0.49
     businesswoman
    0.48
    POSITIVE LOGITS
     merda
    0.71
     fuck
    0.69
    fuck
    0.67
    Fuck
    0.67
     shit
    0.66
     Fuck
    0.64
     dunno
    0.64
     Fallout
    0.64
     Warhammer
    0.63
     lmao
    0.63
    Act Density 0.039%

    No Known Activations