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

    Neuronpedia

    INDEX
    Explanations

    overwhelming sense, Option 1

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    エンジン
    0.45
     engine
    0.43
     nuts
    0.42
    croft
    0.41
    อส
    0.40
    te
    0.40
     fift
    0.40
    alla
    0.39
     am
    0.39
     crusade
    0.39
    POSITIVE LOGITS
    <bos>
    0.40
     टू
    0.39
     ظ
    0.39
     Each
    0.38
    BeforeText
    0.38
     वर्ण
    0.38
     riqueza
    0.38
     föl
    0.38
     Spread
    0.38
     Sampling
    0.38
    Act Density 0.095%

    No Known Activations