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

    Neuronpedia

    INDEX
    Explanations

    ease, clarity, or improvement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     alınan
    0.46
     kullanılan
    0.46
     của
    0.46
     devons
    0.42
     mehrerer
    0.42
     diese
    0.41
     செய்யப்பட்ட
    0.41
     fer
    0.41
     pemb
    0.41
     ഇവ
    0.41
    POSITIVE LOGITS
    其
    0.47
    ，
    0.46
    его
    0.39
    ੀ
    0.38
    Ін
    0.38
    使其
    0.37
    턴
    0.37
    붉
    0.37
    ；
    0.36
     качестве
    0.35
    Act Density 0.139%

    No Known Activations