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

    Neuronpedia

    INDEX
    Explanations

    code blocks and comments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    1.52
    ir
    1.42
    an
    1.40
    ent
    1.37
    am
    1.35
    od
    1.35
    or
    1.33
    ol
    1.33
    on
    1.32
    et
    1.32
    POSITIVE LOGITS
     Tregs
    1.24
     Valores
    1.17
     disgraceful
    1.16
     simplices
    1.14
     eprintln
    1.13
    té
    1.12
     circumvent
    1.12
    ,_-
    1.11
     horrend
    1.11
     assassinated
    1.11
    Act Density 0.217%

    No Known Activations