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

    Neuronpedia

    INDEX
    Explanations

    script and cursive fonts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    BOUNDARY
    0.43
    GRID
    0.41
     прямоуго
    0.40
    rectangle
    0.40
    চারের
    0.39
    [multimodal]
    0.39
     Taxonomy
    0.39
    ODE
    0.39
     уйнау
    0.38
     ойнотуу
    0.38
    POSITIVE LOGITS
     Script
    0.68
     swirl
    0.68
     swirls
    0.68
     curly
    0.66
     script
    0.63
     swo
    0.60
     swirling
    0.57
     sw
    0.57
     fancy
    0.56
    script
    0.56
    Act Density 0.003%

    No Known Activations