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

    Neuronpedia

    INDEX
    Explanations

    Members versus non-members

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ank
    -0.06
     Tup
    -0.06
     лише
    -0.06
    省
    -0.06
    included
    -0.06
     Marina
    -0.06
     websocket
    -0.06
     ux
    -0.06
          ↵
    -0.06
     humidity
    -0.06
    POSITIVE LOGITS
    [word
    0.07
    [dir
    0.06
    [op
    0.06
    [char
    0.06
     toJSON
    0.06
     nev
    0.06
    주는
    0.06
    [n
    0.06
     sürdür
    0.06
     endiş
    0.06
    Act Density 0.020%

    No Known Activations