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

    Neuronpedia

    INDEX
    Explanations

    `->` or `>` after `head`, `item`, `object`, `current`, `Private`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     交換
    0.63
     Eating
    0.58
     desk
    0.58
     आर
    0.56
    ٛ
    0.55
    青年
    0.55
     صن
    0.55
     "$
    0.55
     dill
    0.54
     आग्रह
    0.53
    POSITIVE LOGITS
    ->
    0.78
    вища
    0.78
    Нор
    0.60
    ilir
    0.60
    iters
    0.60
    →
    0.59
    ారు
    0.59
    toires
    0.59
    真空
    0.58
    败
    0.57
    Act Density 0.154%

    No Known Activations