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

    Neuronpedia

    INDEX
    Explanations

    When Where

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ergus
    -0.07
    ())[
    -0.06
    խ
    -0.06
     Course
    -0.06
     Mum
    -0.06
     frais
    -0.06
     Current
    -0.06
     Tools
    -0.06
    (conn
    -0.06
     menstr
    -0.06
    POSITIVE LOGITS
    zb
    0.08
     RED
    0.07
     automated
    0.07
    _LINEAR
    0.07
     "]";↵
    0.07
    יצה
    0.07
     adversary
    0.07
    DEL
    0.07
     ......
    0.06
    .BLUE
    0.06
    Act Density 0.013%

    No Known Activations