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

    Neuronpedia

    INDEX
    Explanations

    is family, permissions, ITEM, inches, if

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    曼
    0.46
    ま
    0.39
    маты
    0.39
     اك
    0.38
    со
    0.37
     경우는
    0.36
    馆
    0.36
    圾
    0.36
    core
    0.36
    اك
    0.36
    POSITIVE LOGITS
     BOT
    0.40
    প্রশ
    0.40
     MACH
    0.38
    擋
    0.38
    ophytic
    0.36
     Strategic
    0.36
     allot
    0.36
     Buffett
    0.35
    jdt
    0.35
     Printing
    0.34
    Act Density 0.001%

    No Known Activations