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

    Neuronpedia

    INDEX
    Explanations

    Nuances, like muscle, not valid

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    apped
    0.67
     আব্দ
    0.61
    ვით
    0.59
    atisme
    0.59
    hatt
    0.58
     ಕ
    0.58
     Corn
    0.57
     fa
    0.57
    ্ডিং
    0.57
    akwa
    0.56
    POSITIVE LOGITS
    straw
    0.68
     अंज
    0.67
    anei
    0.67
     straw
    0.62
     स्वीकार
    0.61
     Detox
    0.61
    同意
    0.61
     BufferedWriter
    0.60
    oxi
    0.60
     banane
    0.60
    Act Density 0.176%

    No Known Activations