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

    Neuronpedia

    INDEX
    Explanations

    ``` code blocks and CREATE statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    \
    0.49
     (
    0.48
    s
    0.45
     (\
    0.45
    na
    0.43
    يل
    0.42
     in
    0.42
     to
    0.41
    ll
    0.41
     are
    0.41
    POSITIVE LOGITS
    C
    0.50
    a
    0.49
    P
    0.48
    G
    0.48
    N
    0.47
     הח
    0.44
    示す
    0.43
     કર્યું
    0.42
    D
    0.42
    ه
    0.42
    Act Density 0.408%

    No Known Activations