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

    Neuronpedia

    INDEX
    Explanations

    various numerical representations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nine
    1.64
     Nine
    1.50
    Nine
    1.42
    nine
    1.33
     nueve
    1.27
     eleven
    1.24
     nineties
    1.21
    九
    1.18
     neun
    1.17
     ninth
    1.13
    POSITIVE LOGITS
    2
    1.49
    1
    1.33
     ২
    1.33
     ১
    1.32
    3
    1.23
     одну
    1.22
     १
    1.21
     डेढ़
    1.18
     jedną
    1.18
     ۲
    1.14
    Act Density 0.415%

    No Known Activations