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

    Neuronpedia

    INDEX
    Explanations

    python imports from libraries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hoge
    0.55
     is
    0.55
    in
    0.55
    d
    0.54
     e
    0.54
     icon
    0.53
     ikon
    0.53
     H
    0.52
     extract
    0.52
     o
    0.51
    POSITIVE LOGITS
    ாலும்
    0.57
    да
    0.56
    но
    0.55
    ม
    0.55
    ну
    0.54
    ৭০
    0.54
    اک
    0.54
    вся
    0.54
    ላይ
    0.53
    с
    0.53
    Act Density 0.006%

    No Known Activations