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

    Neuronpedia

    INDEX
    Explanations

    latex math and regex

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     social
    1.12
     ch
    1.00
     craftsman
    0.98
     like
    0.98
     san
    0.95
     re
    0.95
     lik
    0.95
     rele
    0.94
     marketing
    0.93
     th
    0.92
    POSITIVE LOGITS
    tilde
    1.77
    sqrt
    1.63
    textbf
    1.61
    text
    1.60
    varphi
    1.58
    ldots
    1.58
    overline
    1.54
    mathbb
    1.52
    textit
    1.51
    textrm
    1.49
    Act Density 0.063%

    No Known Activations