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

    Neuronpedia

    INDEX
    Explanations

    A/B tests and specific concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ji
    0.56
    ttino
    0.48
     tiered
    0.48
    HC
    0.47
    ጋገብ
    0.46
     peaking
    0.45
    簸
    0.45
    Jf
    0.45
     flocked
    0.45
     nodded
    0.44
    POSITIVE LOGITS
    át
    0.50
     entier
    0.50
     stedet
    0.48
     учеб
    0.48
     оператив
    0.48
     место
    0.47
    chte
    0.47
     автомо
    0.47
    ãi
    0.46
    但这
    0.45
    Act Density 0.001%

    No Known Activations