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

    Neuronpedia

    INDEX
    Explanations

    thought experiment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    鱚
    -2.34
    ⃙
    -2.31
    ised
    -2.27
    蕯
    -2.23
    皤
    -2.22
    萣
    -2.20
     was
    -2.17
    iy
    -2.17
     doigt
    -2.08
    ؟
    -2.05
    POSITIVE LOGITS
    '
    3.19
    .
    3.06
    𐄁
    2.78
    ↵↵
    2.50
    mathrm
    2.50
    但是
    2.23
    骉
    2.22
     Fakten
    2.20
     italianos
    2.16
    亊
    2.13
    Act Density 0.025%

    No Known Activations