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

    Neuronpedia

    INDEX
    Explanations

    number representation and levels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     niezwy
    0.73
    Ab
    0.68
     három
    0.68
    department
    0.66
     veoma
    0.66
    special
    0.66
    another
    0.65
    ামূলক
    0.64
    兩
    0.64
    avatth
    0.64
    POSITIVE LOGITS
     connotation
    0.89
     versions
    0.87
    😑
    0.85
     بودن
    0.83
     connotations
    0.83
     progression
    0.82
     depiction
    0.82
     เพราะ
    0.81
     버전
    0.80
     unless
    0.80
    Act Density 0.176%

    No Known Activations