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

    Neuronpedia

    INDEX
    Explanations

    code reproducibility

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    /problems
    -0.08
     bull
    -0.08
    غة
    -0.08
     Оч
    -0.08
    bull
    -0.08
    вей
    -0.08
    Wis
    -0.07
     aboard
    -0.07
    thol
    -0.07
    智慧
    -0.07
    POSITIVE LOGITS
     reproduc
    0.16
     predictable
    0.10
    -repeat
    0.10
     reuse
    0.09
     deterministic
    0.09
    重复
    0.09
     repeat
    0.09
     reproducción
    0.09
     debugging
    0.09
     reproduction
    0.09
    Act Density 0.002%

    No Known Activations