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

    Neuronpedia

    INDEX
    Explanations

    self-attention within sequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    baş
    0.51
    edged
    0.50
    hanging
    0.47
    başı
    0.46
    kový
    0.46
    ances
    0.46
    spent
    0.45
    pmap
    0.45
    verbs
    0.44
    transistors
    0.43
    POSITIVE LOGITS
    车
    0.47
    最
    0.44
     млад
    0.44
    \\
    0.43
    車
    0.42
     rô
    0.42
     læ
    0.41
    他にも
    0.40
    ة
    0.40
     cru
    0.40
    Act Density 0.004%

    No Known Activations