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

    Neuronpedia

    INDEX
    Explanations

    VTuber descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     menghasilkan
    1.26
     bottlenecks
    1.08
     elettron
    1.05
     Optionen
    1.05
     relatif
    1.04
     otomatis
    1.03
     zostały
    1.01
    ukung
    1.00
    ✅
    1.00
     pulleys
    0.99
    POSITIVE LOGITS
    。」
    1.04
    Professor
    1.00
    Lay
    0.98
    ти
    0.94
    ।"
    0.92
    ే
    0.92
    ।”
    0.91
    r
    0.89
    ם
    0.89
     জন্যও
    0.88
    Act Density 0.568%

    No Known Activations