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

    Neuronpedia

    INDEX
    Explanations

    explanation continues after phrasing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    م
    1.00
    äns
    0.76
    ات
    0.73
     jusque
    0.72
    aré
    0.68
     الحد
    0.68
     कामया
    0.68
     restitution
    0.66
    残酷
    0.66
    োপ
    0.65
    POSITIVE LOGITS
    ᠴ
    0.73
     그런
    0.70
     Calcul
    0.70
    ਣਾ
    0.70
     cần
    0.70
     drilled
    0.69
     여름
    0.68
    Ци
    0.68
    стана
    0.67
     лабора
    0.67
    Act Density 9.337%

    No Known Activations