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

    Neuronpedia

    INDEX
    Explanations

    say "W"

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    าะ
    0.38
    éder
    0.38
     PMC
    0.37
    archer
    0.37
     Self
    0.37
    stedt
    0.37
    ipc
    0.36
     aces
    0.36
    மரு
    0.36
     ح
    0.35
    POSITIVE LOGITS
    わ
    0.43
    ORDON
    0.40
    큥
    0.38
     конститу
    0.36
     volunteered
    0.36
    Tür
    0.36
    HEX
    0.36
    ούς
    0.36
    段
    0.35
     {-
    0.35
    Act Density 0.001%

    No Known Activations