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

    Neuronpedia

    INDEX
    Explanations

    Welsh, welts, Welocalize, Welterweight

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    पछि
    0.40
    임
    0.40
    ưu
    0.38
    ື
    0.38
    彘
    0.37
     thuyết
    0.37
     Taille
    0.36
     Reveals
    0.36
     మధ్య
    0.36
     subsequence
    0.36
    POSITIVE LOGITS
    Wel
    0.64
     Wel
    0.54
     wel
    0.54
     Welch
    0.49
    COME
    0.44
    wel
    0.44
    हट
    0.44
     WEL
    0.42
     welder
    0.40
    coming
    0.40
    Act Density 0.003%

    No Known Activations