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

    Neuronpedia

    INDEX
    Explanations

    the Prompt, Taxes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     crash
    0.10
     as
    0.09
     but
    0.09
    ãs
    0.09
     confronted
    0.08
    ędz
    0.08
     unexpectedly
    0.08
     됩니다
    0.08
    <a>
    0.08
    crash
    0.08
    POSITIVE LOGITS
     acide
    0.11
     স্মরণ
    0.10
     سرطان
    0.10
     бактери
    0.10
     улучшения
    0.10
     leuc
    0.10
    胀
    0.10
    asonic
    0.10
     intelligente
    0.10
     綿
    0.10
    Act Density 0.000%

    No Known Activations