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

    Neuronpedia

    INDEX
    Explanations

    develop, calculate, confirm, generate, install

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     enak
    0.61
     نباش
    0.54
     udah
    0.54
     मत
    0.53
     אבל
    0.53
     هاي
    0.52
     sudah
    0.52
     berhasil
    0.51
     മാത്രം
    0.51
     כן
    0.51
    POSITIVE LOGITS
    한다
    0.78
    하여
    0.76
    하는
    0.71
    하며
    0.68
    하고
    0.67
    함으로써
    0.60
    받
    0.59
    并通过
    0.58
    并将
    0.57
    된다
    0.52
    Act Density 0.001%

    No Known Activations