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

    Neuronpedia

    INDEX
    Explanations

    data types and file names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ৬
    0.59
     ໜອງ
    0.56
    ដើម្បី
    0.55
    0.55
     Sparta
    0.51
     Straf
    0.51
    inci
    0.50
     江
    0.49
    ంగ్‌
    0.48
    -
    0.48
    POSITIVE LOGITS
    ان
    0.56
     estou
    0.55
     saluran
    0.50
    ش
    0.50
     masing
    0.49
    بر
    0.48
     høj
    0.48
     uang
    0.47
    ähne
    0.47
    лер
    0.46
    Act Density 0.063%

    No Known Activations