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

    Neuronpedia

    INDEX
    Explanations

    Destination Unreachable

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sacré
    0.38
     bese
    0.38
    .${
    0.36
     partake
    0.36
     मारी
    0.36
    晚上
    0.35
    ﹀
    0.35
    行政
    0.35
     жить
    0.35
    㣙
    0.34
    POSITIVE LOGITS
    erjee
    0.41
    ulas
    0.40
     Mane
    0.39
    eliz
    0.39
     maneuver
    0.39
     Counsel
    0.38
     राव
    0.38
    fahr
    0.37
    mane
    0.37
     ҳу
    0.37
    Act Density 0.001%

    No Known Activations