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

    Neuronpedia

    INDEX
    Explanations

    is

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    red
    -0.07
    arda
    -0.07
    言い
    -0.07
    Directory
    -0.06
     credibility
    -0.06
     bread
    -0.06
    φέρει
    -0.06
    ют
    -0.06
    odom
    -0.06
    roids
    -0.06
    POSITIVE LOGITS
     illustrated
    0.07
     ni
    0.06
    (cert
    0.06
     neut
    0.06
     cancellation
    0.06
     život
    0.06
    /oct
    0.06
     nå
    0.06
     bapt
    0.06
    „N
    0.06
    Act Density 0.027%

    No Known Activations