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

    Neuronpedia

    INDEX
    Explanations

    steady performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Giải
    0.82
    漏洞
    0.81
     özel
    0.81
    Traité
    0.81
    Giải
    0.79
     öffentlich
    0.79
     Multipart
    0.78
    搭建
    0.77
     ஒப்பந்த
    0.77
    替换
    0.77
    POSITIVE LOGITS
     acceleration
    1.40
     throttle
    1.30
    Acceleration
    1.22
     accelerating
    1.19
     accelerate
    1.17
     rpm
    1.16
    acceleration
    1.16
     Acceleration
    1.15
     steady
    1.10
     acceler
    1.09
    Act Density 0.448%

    No Known Activations