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

    Neuronpedia

    INDEX
    Explanations

    access, see, order, changed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     erled
    0.59
     terlebih
    0.57
     geeign
    0.55
     simple
    0.53
     Simple
    0.53
     Leave
    0.52
     oversized
    0.52
     übernahm
    0.52
    Simple
    0.51
     templates
    0.51
    POSITIVE LOGITS
     interag
    0.63
     interacción
    0.62
    interacting
    0.62
    leyebilirsiniz
    0.60
     distinguishable
    0.59
     interacts
    0.59
    可以看到
    0.57
    粒子
    0.54
     interação
    0.54
     intelligible
    0.54
    Act Density 0.002%

    No Known Activations