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

    Neuronpedia

    INDEX
    Explanations

    reach

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𓐍
    -2.33
     سبک
    -2.17
     なら
    -2.08
    u
    -1.96
    你说
    -1.94
     мира
    -1.88
    enablog
    -1.85
    战胜
    -1.84
     том
    -1.82
    </
    -1.79
    POSITIVE LOGITS
    s
    2.63
    想到了
    2.36
    所以我
    2.25
    龗
    2.20
    禤
    2.19
    鬟
    2.19
    潆
    2.19
    ︓
    2.17
     estren
    2.17
     They
    2.16
    Act Density 0.012%

    No Known Activations