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

    Neuronpedia

    INDEX
    Explanations

    foreign pronunciations of mythical beings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused2121>
    0.68
    <unused2125>
    0.68
    <unused2164>
    0.64
    <unused254>
    0.63
    <unused656>
    0.62
     supernat
    0.62
    <unused2130>
    0.62
    <unused499>
    0.60
     buddhist
    0.59
    <unused574>
    0.59
    POSITIVE LOGITS
     kom
    0.93
     kor
    0.88
     tiga
    0.87
     ма
    0.85
     yap
    0.84
     ק
    0.83
     zur
    0.82
     פ
    0.82
     nach
    0.82
     zu
    0.82
    Act Density 0.407%

    No Known Activations