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

    Neuronpedia

    INDEX
    Explanations

    handling mixed data types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     receipts
    0.41
     رسید
    0.41
     provisioning
    0.40
    궂
    0.39
     legacy
    0.38
     অধিদপ্তর
    0.38
     averaging
    0.38
    꾹
    0.38
    Sung
    0.37
     Šk
    0.37
    POSITIVE LOGITS
    ınıza
    0.45
    slice
    0.42
    opropyl
    0.40
    oshape
    0.40
     ля
    0.40
    тувати
    0.39
     pretzel
    0.38
    тельная
    0.38
     хе
    0.38
    hearing
    0.38
    Act Density 0.000%

    No Known Activations