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

    Neuronpedia

    INDEX
    Explanations

    ties, FT, hugging, bugs, lats, RAM, TLD, denoising, LMSYS, pangolin

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tob
    0.28
    THAN
    0.27
     sogen
    0.27
     bekas
    0.27
    tet
    0.27
     switchTo
    0.27
     годах
    0.26
     llamado
    0.26
    ៕
    0.26
     বসিয়া
    0.25
    POSITIVE LOGITS
    al
    0.48
    an
    0.46
    ש
    0.43
    er
    0.42
    es
    0.39
    و
    0.39
    at
    0.38
    el
    0.36
    em
    0.35
    ن
    0.35
    Act Density 0.028%

    No Known Activations