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

    Neuronpedia

    INDEX
    Explanations

    safety, ABS, iron, debugging

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     CTCF
    0.48
     Islington
    0.41
     Glor
    0.38
     Loire
    0.38
     Relating
    0.38
     peas
    0.37
     Rather
    0.37
     supremacy
    0.36
     दुसरी
    0.36
    ്യാപ
    0.36
    POSITIVE LOGITS
    िए
    0.41
    SdkError
    0.39
    ธ์
    0.38
     സാഹ
    0.35
    ósticos
    0.35
    Назад
    0.34
    ட்டிக்
    0.34
    рисов
    0.34
    িউ
    0.34
    ിലി
    0.34
    Act Density 0.001%

    No Known Activations