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

    Neuronpedia

    INDEX
    Explanations

    say "x"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    /admin
    -0.08
    良
    -0.07
     aggregate
    -0.07
     invalid
    -0.07
    があった
    -0.06
    害
    -0.06
    诽
    -0.06
    怛
    -0.06
     form
    -0.06
    他人
    -0.06
    POSITIVE LOGITS
     �
    0.08
     armored
    0.07
     sorte
    0.07
     Debt
    0.07
     ורק
    0.07
     merciless
    0.07
    _doc
    0.07
     surged
    0.07
     Lore
    0.07
    🔀
    0.07
    Act Density 0.178%

    No Known Activations