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

    Neuronpedia

    INDEX
    Explanations

    symmetry and reflections

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    -0.09
    ablemente
    -0.08
     enorme
    -0.08
    Bare
    -0.08
     luhur
    -0.08
     высокой
    -0.07
     Amnesty
    -0.07
     mined
    -0.07
     Belle
    -0.07
     Lv
    -0.07
    POSITIVE LOGITS
     invari
    0.08
     setups
    0.08
     Libraries
    0.08
     constructions
    0.08
    .Compare
    0.08
     comparisons
    0.07
     Compar
    0.07
     Geometry
    0.07
     geometry
    0.07
     involved
    0.07
    Act Density 0.031%
    No Known Activations