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

    Neuronpedia

    INDEX
    Explanations

    code context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zippers
    0.48
     netizens
    0.46
     copyrights
    0.46
     inequalities
    0.44
    ByMerging
    0.44
     pronouns
    0.43
     organelles
    0.43
     diast
    0.43
     apparaissent
    0.43
     liabilities
    0.42
    POSITIVE LOGITS
    目
    0.44
    k
    0.43
    .
    0.43
    。
    0.43
    工作
    0.43
    ك
    0.43
    行
    0.41
    not
    0.40
    在
    0.40
    ف
    0.39
    Act Density 0.042%

    No Known Activations