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

    Neuronpedia

    INDEX
    Explanations

    specific technical terms and definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    fe
    0.52
    ra
    0.51
    sp
    0.48
    k
    0.47
    sc
    0.46
    es
    0.46
    le
    0.45
    va
    0.45
    br
    0.44
    ha
    0.44
    POSITIVE LOGITS
    타
    0.61
     этих
    0.55
    이
    0.54
    ない
    0.54
    을
    0.53
    א
    0.51
     класса
    0.50
    лога
    0.49
    אי
    0.49
    내
    0.49
    Act Density 0.000%

    No Known Activations