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

    Neuronpedia

    INDEX
    Explanations

    repeated errors or issues across multiple instances or contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    comparison or lack thereof

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     betweenstory
    -0.62
     الرياضيه
    -0.59
     MenuView
    -0.59
     '\\;'
    -0.58
    AnchorTagHelper
    -0.57
    IUrlHelper
    -0.56
    发表于
    -0.56
     للاسماء
    -0.55
     المعيارى
    -0.54
    WireFormatLite
    -0.53
    POSITIVE LOGITS
     defaultstate
    0.37
     surla
    0.30
    moveToNext
    0.30
    pu
    0.29
     last
    0.29
     opge
    0.29
    permitAll
    0.28
     pu
    0.28
    ✷
    0.27
    اوية
    0.27
    Act Density 0.025%

    No Known Activations