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

    Neuronpedia

    INDEX
    Explanations

    =

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     prisoner
    -0.07
    盖
    -0.07
     Й
    -0.06
    /business
    -0.06
    StateException
    -0.06
    ฮ
    -0.06
     holes
    -0.06
     hill
    -0.06
    SenderId
    -0.06
    anager
    -0.06
    POSITIVE LOGITS
     STEM
    0.07
     Rolling
    0.07
    ILTER
    0.06
    /generated
    0.06
     작업
    0.06
    242
    0.06
     >>>
    0.06
    cling
    0.06
     _:
    0.06
    ιν
    0.06
    Act Density 0.004%

    No Known Activations