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

    Neuronpedia

    INDEX
    Explanations

    part

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _forward
    -0.07
    .''
    -0.07
    =F
    -0.07
    FFE
    -0.07
    ']:↵
    -0.07
    .")↵↵
    -0.07
    。「
    -0.07
    :↵↵
    -0.07
    ()])↵
    -0.07
     auf
    -0.06
    POSITIVE LOGITS
    户型
    0.08
     extravag
    0.07
    在市场上
    0.07
    billing
    0.07
    -job
    0.07
     Năm
    0.07
     resize
    0.07
    reak
    0.07
     Gam
    0.07
    attention
    0.07
    Act Density 0.003%

    No Known Activations