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

    Neuronpedia

    INDEX
    Explanations

    if you'd like

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tokens
    0.37
    TimeUnit
    0.36
    Roof
    0.35
     договора
    0.35
     यदि
    0.34
     if
    0.33
     berusaha
    0.33
     если
    0.33
    睃
    0.33
    Unable
    0.32
    POSITIVE LOGITS
    我可以
    0.44
     puedo
    0.40
     posso
    0.37
     можна
    0.35
    有什么
    0.33
    你想
    0.33
     minhas
    0.32
     언제
    0.31
    prix
    0.31
     boleh
    0.31
    Act Density 0.008%

    No Known Activations