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

    Neuronpedia

    INDEX
    Explanations

    acknowledging humanity, support, or options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )',
    0.47
    )、
    0.47
     başarılı
    0.45
     vanaf
    0.45
     sobra
    0.44
     MUL
    0.44
     акчага
    0.44
     alınd
    0.44
    ₂)
    0.43
     Kø
    0.43
    POSITIVE LOGITS
    ה
    0.60
    ל
    0.55
    Our
    0.54
    Supported
    0.54
    Support
    0.53
    ص
    0.53
    Provided
    0.52
    ח
    0.52
    Other
    0.51
    الي
    0.51
    Act Density 0.001%

    No Known Activations