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

    Neuronpedia

    INDEX
    Explanations

    blocks, novel, complete, shame, sent back

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     yogurt
    0.55
     flavoring
    0.54
     Coronado
    0.51
    GoObject
    0.47
     എതി
    0.46
    ที่สุด
    0.46
     определён
    0.45
    商品的
    0.45
    လေ
    0.45
    slideDuplicate
    0.44
    POSITIVE LOGITS
     crossed
    0.48
    德
    0.47
    で
    0.46
    ન
    0.45
    ←
    0.45
    寬
    0.44
    瀨
    0.44
    sous
    0.44
    셈
    0.44
    安
    0.44
    Act Density 0.001%

    No Known Activations