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

    Neuronpedia

    INDEX
    Explanations

    genitals or russian/arabic/chinese words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    न
    1.67
    पीरियंस
    1.59
     persuasion
    1.55
    anager
    1.48
    ../
    1.40
    ુ
    1.38
    ногда
    1.36
    ्रे
    1.34
     والإ
    1.33
     awal
    1.33
    POSITIVE LOGITS
    NIH
    1.58
     oligos
    1.50
     marginTop
    1.42
    Đ
    1.36
     padrão
    1.32
    𝑻
    1.30
     dai
    1.30
    לת
    1.30
    VITY
    1.28
    Į
    1.28
    Act Density 0.001%

    No Known Activations