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

    Neuronpedia

    INDEX
    Explanations

    optimization and minimization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    的
    0.72
    在
    0.66
    س
    0.64
    󠁮
    0.61
     in
    0.60
    是
    0.58
    s
    0.58
    Α
    0.57
    专
    0.56
    倨
    0.56
    POSITIVE LOGITS
    anine
    0.63
     mondial
    0.62
     curtail
    0.60
    anın
    0.59
    лера
    0.59
    ла
    0.58
    at
    0.57
    ся
    0.57
    кси
    0.56
     Mondial
    0.55
    Act Density 0.058%

    No Known Activations