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

    Neuronpedia

    INDEX
    Explanations

    Questions needing explanations/analysis

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     affairs
    -0.08
    Bounds
    -0.07
     Très
    -0.07
    wear
    -0.07
     enemigo
    -0.07
    Encontr
    -0.07
     níveis
    -0.07
    wicklung
    -0.07
    Enemy
    -0.07
     sentiments
    -0.07
    POSITIVE LOGITS
     why
    0.23
    为什么
    0.21
    理由
    0.21
     waarom
    0.21
     Why
    0.20
    Why
    0.20
    why
    0.20
     warum
    0.20
     이유
    0.19
     ఎంద
    0.18
    Act Density 0.171%

    No Known Activations