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

    Neuronpedia

    INDEX
    Explanations

    errors, lacks, very, well

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    uert
    0.64
    :
    0.57
    ulers
    0.56
    ى
    0.53
    ur
    0.52
    uers
    0.51
    ße
    0.51
    ಾ
    0.51
    ues
    0.50
     مهر
    0.50
    POSITIVE LOGITS
    ч
    0.61
    YOU
    0.59
    к
    0.59
    South
    0.58
    It
    0.57
    Racing
    0.57
    Jim
    0.55
     harán
    0.55
    𝘭
    0.55
    teacher
    0.54
    Act Density 0.406%

    No Known Activations