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

    Neuronpedia

    INDEX
    Explanations

    gpt-3.5-turbo model names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    MenuBar
    0.45
    клар
    0.41
     affirmations
    0.41
     hạn
    0.38
     الدین
    0.37
    龄
    0.37
    钧
    0.37
     கலோரிகள்
    0.36
     phận
    0.36
    拁
    0.36
    POSITIVE LOGITS
    ang
    0.41
     cosidd
    0.40
    ↵
    0.38
     pequeños
    0.38
     vencedor
    0.35
     last
    0.35
     privat
    0.35
     pequenos
    0.34
     dividing
    0.34
    esco
    0.34
    Act Density 0.133%

    No Known Activations