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

    Neuronpedia

    INDEX
    Explanations

    consisting of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     بابەت
    0.46
     අතර
    0.44
     ощу
    0.43
    urel
    0.42
     разгово
    0.42
    countery
    0.41
     मद्देनजर
    0.40
     गंभीर
    0.40
    textepsilon
    0.40
     আলোচ
    0.40
    POSITIVE LOGITS
     consists
    0.54
     consisting
    0.52
    =
    0.46
     comprised
    0.46
    With
    0.46
     consisted
    0.44
     consist
    0.43
     consiste
    0.43
    以
    0.42
     comprise
    0.42
    Act Density 0.049%

    No Known Activations