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

    Neuronpedia

    INDEX
    Explanations

    grid

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     دفاع
    -0.09
    .clone
    -0.09
    .compiler
    -0.09
     lineback
    -0.08
     defense
    -0.08
    atriz
    -0.08
    召
    -0.08
    Reserved
    -0.08
     nuclé
    -0.08
     ultrices
    -0.08
    POSITIVE LOGITS
     underserved
    0.10
     outages
    0.09
     powered
    0.09
     outage
    0.09
     rural
    0.08
     stroom
    0.08
     flashlight
    0.08
     améliorer
    0.08
     reliably
    0.08
     powering
    0.08
    Act Density 0.012%

    No Known Activations