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

    Neuronpedia

    INDEX
    Explanations

    risk and risks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    思い
    1.36
     लगातार
    1.30
     wahrscheinlich
    1.28
     vidro
    1.21
    Storia
    1.17
    มี
    1.14
     adunay
    1.13
     konnten
    1.12
     meningkat
    1.11
     zum
    1.11
    POSITIVE LOGITS
    т
    1.18
    ナット
    1.15
     overcoming
    1.15
    sman
    1.11
    tura
    1.10
     contention
    1.08
     deduct
    1.08
     exacting
    1.05
     restricting
    1.05
     elevating
    1.04
    Act Density 0.000%

    No Known Activations