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

    Neuronpedia

    INDEX
    Explanations

    violating rules/boundaries

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     NHS
    -0.07
    .kafka
    -0.07
    铯
    -0.07
     כלי
    -0.07
    isempty
    -0.07
    祉
    -0.07
    Unix
    -0.07
     Алексан
    -0.07
     unix
    -0.07
    uv
    -0.06
    POSITIVE LOGITS
    Inverse
    0.08
    ()]↵
    0.08
    调查
    0.07
    天鹅
    0.07
    .navigationItem
    0.07
    \E
    0.07
     tightly
    0.07
    这里
    0.07
    偶
    0.07
     resc
    0.07
    Act Density 0.001%

    No Known Activations