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

    Neuronpedia

    INDEX
    Explanations

    Questions about advantages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Subscription
    -0.07
     portraying
    -0.07
     Gab
    -0.06
    Honda
    -0.06
     manhã
    -0.06
    関
    -0.06
    Johnson
    -0.06
    吸取
    -0.06
    Statics
    -0.06
     científ
    -0.06
    POSITIVE LOGITS
    .dy
    0.08
    .setData
    0.08
    匀
    0.07
    plit
    0.07
    资本
    0.07
    Attr
    0.07
    *pi
    0.07
    彤
    0.07
     dope
    0.07
    _ele
    0.06
    Act Density 0.132%

    No Known Activations