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

    Neuronpedia

    INDEX
    Explanations

    varies greatly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ussen
    0.48
    Shape
    0.44
    Sentence
    0.44
     científ
    0.42
    Direction
    0.42
     cliché
    0.41
    Bag
    0.40
     monstrous
    0.40
     régl
    0.39
     съем
    0.39
    POSITIVE LOGITS
     andolan
    0.51
    เลือก
    0.50
     setor
    0.48
     Alek
    0.46
     Andrey
    0.46
    印
    0.45
     prix
    0.44
     Paly
    0.44
    या
    0.44
     Skype
    0.43
    Act Density 0.003%

    No Known Activations