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

    Neuronpedia

    INDEX
    Explanations

    non-English characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ר
    0.63
    лы
    0.57
    is
    0.53
    ף
    0.53
    ס
    0.52
    mathspace
    0.51
    ス
    0.51
    *}$
    0.50
    сть
    0.50
    .{
    0.50
    POSITIVE LOGITS
     ডু
    0.52
     appliqué
    0.51
     ten
    0.50
     pén
    0.49
     kuts
    0.48
     బ్యాంకు
    0.48
     redd
    0.47
     pod
    0.47
     ingot
    0.47
     בש
    0.45
    Act Density 0.001%

    No Known Activations