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

    Neuronpedia

    INDEX
    Explanations

    at its core / at least

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     是否
    0.86
    อะไร
    0.79
    norod
    0.78
     putem
    0.78
    ෲ
    0.77
    ෂ
    0.77
    יש
    0.76
     继续
    0.75
    atoren
    0.75
    ရ
    0.73
    POSITIVE LOGITS
     least
    2.33
    least
    1.60
     times
    1.51
    Least
    1.34
     intervals
    1.30
     Least
    1.30
     home
    1.24
     odds
    1.24
     dawn
    1.23
     night
    1.21
    Act Density 0.135%

    No Known Activations