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

    Neuronpedia

    INDEX
    Explanations

    lead to consequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     BASED
    0.32
    ську
    0.32
    ষ্ট
    0.31
     svoju
    0.31
    ري
    0.31
     („
    0.31
    whatever
    0.30
     based
    0.29
    িলার
    0.29
     drugi
    0.29
    POSITIVE LOGITS
     astray
    0.70
     leads
    0.57
     lead
    0.55
     إلى
    0.54
     to
    0.54
    到
    0.49
    到一个
    0.48
     credence
    0.48
     منجر
    0.48
     toa
    0.47
    Act Density 0.006%

    No Known Activations