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

    Neuronpedia

    INDEX
    Explanations

    moving/switching

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Political
    -0.07
     Political
    -0.06
     '|
    -0.06
     negatively
    -0.06
     represents
    -0.06
    UNITY
    -0.06
     Meredith
    -0.06
    ิจ
    -0.06
     legitimately
    -0.06
    *(
    -0.06
    POSITIVE LOGITS
     statt
    0.07
    �
    0.07
    ๒
    0.06
    restart
    0.06
    ekk
    0.06
    лож
    0.06
     rr
    0.06
    _mid
    0.06
     дру
    0.06
    ่อน
    0.06
    Act Density 0.031%

    No Known Activations