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

    Neuronpedia

    INDEX
    Explanations

    an older

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    +}$
    0.76
     };
    0.75
    മസ
    0.71
    Зна
    0.70
    sampling
    0.70
    mps
    0.68
    Ад
    0.66
    acks
    0.66
    1
    0.66
    ؤال
    0.65
    POSITIVE LOGITS
     dintre
    0.72
    cinoma
    0.70
     такую
    0.69
    duğ
    0.69
    ম্ভ
    0.68
     মার্চের
    0.68
     тогда
    0.67
     prije
    0.67
     muerto
    0.66
     丨
    0.66
    Act Density 0.001%

    No Known Activations