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

    Neuronpedia

    INDEX
    Explanations

    type (occasional "

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Races
    0.45
     Race
    0.44
     Racing
    0.44
     specifico
    0.44
    ifika
    0.43
    trail
    0.43
    ных
    0.43
    antik
    0.43
     spécifique
    0.42
     fino
    0.41
    POSITIVE LOGITS
    让人
    0.41
    ↵↵
    0.41
    hém
    0.40
    FreeBuf
    0.38
    últ
    0.38
    句话
    0.37
    áng
    0.37
    がい
    0.36
    句話
    0.36
     hom
    0.35
    Act Density 0.001%

    No Known Activations