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

    Neuronpedia

    INDEX
    Explanations

    disclaimers or negative prefixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )
    0.53
    ς
    0.51
    \
    0.48
     merken
    0.48
     roku
    0.47
    }
    0.46
     refers
    0.45
     jedes
    0.45
    $
    0.45
    \">
    0.43
    POSITIVE LOGITS
    otica
    0.50
    ưng
    0.50
    وية
    0.48
    懑
    0.47
    被
    0.47
     poca
    0.46
    ار
    0.46
    ش
    0.45
    𝔱
    0.45
    اند
    0.44
    Act Density 0.274%

    No Known Activations