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

    Neuronpedia

    INDEX
    Explanations

    code queries

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ewing
    -0.07
    ?',
    -0.07
    發展
    -0.06
     математи
    -0.06
    病
    -0.06
    Protocol
    -0.06
    举
    -0.06
    aja
    -0.06
    handlers
    -0.06
     Ships
    -0.06
    POSITIVE LOGITS
     aktu
    0.06
    ~↵↵
    0.06
    .substr
    0.06
     refl
    0.06
    laz
    0.06
    pragma
    0.06
    (sockfd
    0.06
     "~
    0.05
    poses
    0.05
     geen
    0.05
    Act Density 0.111%

    No Known Activations