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

    Neuronpedia

    INDEX
    Explanations

    speech

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    iktar
    0.86
     సమావేశ
    0.80
    갠
    0.76
     шар
    0.76
    خصص
    0.75
    লাষ
    0.73
     fornisce
    0.72
    шую
    0.72
    御
    0.72
     dùng
    0.72
    POSITIVE LOGITS
     démon
    0.56
     dis
    0.53
     People
    0.53
     switch
    0.53
     short
    0.51
     flag
    0.51
     changed
    0.51
     instructions
    0.51
     sections
    0.51
    ulfillment
    0.50
    Act Density 0.003%

    No Known Activations