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

    Neuronpedia

    INDEX
    Explanations

    processing, evidence, struggle, Fall, vowel

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ث
    0.49
    ش
    0.48
     والف
    0.45
    лимпи
    0.43
     flourishes
    0.42
     kredi
    0.41
    ﺜ
    0.40
    Tele
    0.39
    спублі
    0.39
     Télé
    0.39
    POSITIVE LOGITS
    %.
    0.43
     ongeveer
    0.42
    龵
    0.41
    ěl
    0.40
    substrate
    0.40
    ினார்கள்
    0.40
     THEY
    0.38
     siete
    0.38
     माफिया
    0.38
     छह
    0.38
    Act Density 0.005%

    No Known Activations