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

    Neuronpedia

    INDEX
    Explanations

    error correction and detection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ATH
    0.42
     丨
    0.40
    κρι
    0.40
    Biden
    0.39
     აღმასრულ
    0.37
    राष्ट्रपति
    0.37
    medallas
    0.36
    niem
    0.36
    ╔
    0.36
     역
    0.36
    POSITIVE LOGITS
     primo
    0.45
     vac
    0.44
     clumps
    0.43
     Campania
    0.43
     Воло
    0.42
     plazo
    0.40
    anja
    0.39
     vacuum
    0.39
     дре
    0.38
     ужа
    0.38
    Act Density 0.042%

    No Known Activations