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

    Neuronpedia

    INDEX
    Explanations

    math and symbol packages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    スの
    1.01
    いる
    0.96
    デ
    0.89
     paramilitary
    0.85
     recid
    0.84
     it
    0.84
    𝓑
    0.82
    ಿದ್ದೇವೆ
    0.82
    いた
    0.80
    قة
    0.78
    POSITIVE LOGITS
    or
    1.36
    il
    1.30
    r
    1.27
    in
    1.26
    er
    1.23
    o
    1.21
    н
    1.20
    im
    1.18
    amssymb
    1.13
    an
    1.12
    Act Density 0.000%

    No Known Activations