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

    Neuronpedia

    INDEX
    Explanations

    hold onto

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    hedron
    0.42
    lád
    0.38
    ivado
    0.37
    ‌ന
    0.37
     hängt
    0.36
     Highlighter
    0.36
    之旅
    0.35
    发生在
    0.35
    രിച്ച
    0.35
    Richt
    0.34
    POSITIVE LOGITS
     onto
    1.12
     hold
    1.06
     aloft
    0.97
     sway
    0.96
     accountable
    0.95
     hostage
    0.95
     holds
    0.89
     Hold
    0.89
     tightly
    0.88
     tight
    0.88
    Act Density 0.031%

    No Known Activations