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

    Neuronpedia

    INDEX
    Explanations

    replace placeholders

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    šķ
    0.67
     meille
    0.63
     मल्टी
    0.63
     agitated
    0.62
    ilidad
    0.60
     jsme
    0.60
     parlement
    0.60
    책
    0.59
     우선
    0.58
     Gobierno
    0.57
    POSITIVE LOGITS
    Replace
    1.47
    替换
    1.47
     replace
    1.45
     Replace
    1.45
     remplacer
    1.34
     заменить
    1.32
     replacing
    1.30
     Replacing
    1.30
     vervangen
    1.30
     erset
    1.29
    Act Density 0.016%

    No Known Activations