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

    Neuronpedia

    INDEX
    Explanations

    impact your, indicates automated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    मंगल
    1.68
    ಕ
    1.62
    !\!\
    1.61
     beschäftigt
    1.55
     blijven
    1.52
     信
    1.49
    irmat
    1.48
    厸
    1.48
    堹
    1.47
    耺
    1.46
    POSITIVE LOGITS
    ang
    1.39
    es
    1.32
    unya
    1.25
     gates
    1.24
     gate
    1.23
    സ്തു
    1.22
    ner
    1.20
    hab
    1.20
    NA
    1.20
    ca
    1.20
    Act Density 0.255%

    No Known Activations