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

    Neuronpedia

    INDEX
    Explanations

    refining words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     marrow
    0.64
     epochs
    0.63
     subter
    0.59
     warts
    0.59
     activos
    0.58
     arcane
    0.58
     epoch
    0.58
     supremacy
    0.57
     arrhythmias
    0.57
     fodder
    0.56
    POSITIVE LOGITS
    在这里
    0.53
    на
    0.52
    ဦ
    0.51
    ويل
    0.50
    Injury
    0.49
    ترك
    0.48
    خدام
    0.47
    ط
    0.47
    vação
    0.46
    션
    0.46
    Act Density 0.000%

    No Known Activations