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

    Neuronpedia

    INDEX
    Explanations

    code and symbols

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     treaty
    -0.07
    dae
    -0.07
     dou
    -0.06
    wel
    -0.06
    ريل
    -0.06
    serializer
    -0.06
     wartime
    -0.06
     juven
    -0.06
    /cmd
    -0.06
     collaborators
    -0.06
    POSITIVE LOGITS
    uces
    0.07
     plausible
    0.07
    !
↵
    0.07
    请
    0.07
    iking
    0.06
    (tc
    0.06
    (){}↵
    0.06
    ，请
    0.06
    /am
    0.06
    igated
    0.06
    Act Density 0.000%

    No Known Activations