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

    Neuronpedia

    INDEX
    Explanations

    strengths

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ashes
    0.67
     bfs
    0.64
     bj
    0.64
    Ь
    0.62
     baff
    0.61
    ashi
    0.61
     Command
    0.60
     অফ
    0.59
     Catar
    0.59
     Kain
    0.59
    POSITIVE LOGITS
     parche
    0.68
    romat
    0.64
    patch
    0.61
    ग़
    0.60
    聽到
    0.60
    romet
    0.59
    laan
    0.58
    롬
    0.57
     сек
    0.57
    リーチ
    0.56
    Act Density 0.187%

    No Known Activations