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

    Neuronpedia

    INDEX
    Explanations

    programming context after certain tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    it
    1.20
    ,
    1.13
    ;
    1.06
    ii
    0.97
    or
    0.95
    queries
    0.93
    controls
    0.92
    da
    0.91
    .
    0.91
    el
    0.90
    POSITIVE LOGITS
     frère
    1.22
     причина
    1.19
    야
    1.17
    ва
    1.13
    想
    1.13
    苼
    1.13
     échant
    1.12
    肐
    1.12
    ს
    1.11
    杪
    1.10
    Act Density 0.000%

    No Known Activations