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

    Neuronpedia

    INDEX
    Explanations

    words followed by separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     =
    0.51
    ylabel
    0.49
     \
    0.49
    acak
    0.47
     bothering
    0.45
     ppl
    0.44
     તેઓ
    0.43
     a
    0.42
     affiliated
    0.42
    eeq
    0.42
    POSITIVE LOGITS
     circul
    0.51
     ścian
    0.51
    饮食
    0.49
     சினிமா
    0.49
     chimneys
    0.48
     filamentous
    0.48
     ventanas
    0.47
     ہوا
    0.46
    Flood
    0.45
     стру
    0.45
    Act Density 0.094%

    No Known Activations