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

    Neuronpedia

    INDEX
    Explanations

    entrances and exits

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ब्रिज
    0.36
    рог
    0.36
    तुरंग
    0.36
     مربع
    0.35
    胺
    0.35
     মহাপ
    0.35
     фай
    0.34
     mira
    0.33
    溱
    0.32
     amis
    0.32
    POSITIVE LOGITS
     yard
    0.59
    yard
    0.55
     porch
    0.50
     corner
    0.47
     door
    0.46
    yards
    0.46
     hall
    0.46
     entrance
    0.45
    street
    0.45
     room
    0.45
    Act Density 0.008%

    No Known Activations