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

    Neuronpedia

    INDEX
    Explanations

    keep up with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    نا
    0.75
    зя
    0.64
    g
    0.63
    at
    0.61
    م
    0.61
    ку
    0.60
    ோ
    0.60
    ا
    0.58
    ↵↵
    0.58
    in
    0.57
    POSITIVE LOGITS
     procured
    0.79
    isierten
    0.77
    .。
    0.75
    itate
    0.73
    ፦
    0.73
     levelled
    0.70
    werking
    0.70
     endeavoured
    0.69
    ։
    0.69
     affixed
    0.69
    Act Density 0.003%

    No Known Activations