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

    Neuronpedia

    INDEX
    Explanations

    .

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    бург
    0.61
     बदलाव
    0.60
     catastrophe
    0.59
     change
    0.58
     disastrous
    0.58
    »),
    0.58
     cat
    0.58
     edge
    0.58
    يدي
    0.58
     trans
    0.57
    POSITIVE LOGITS
     attentions
    0.88
    ার্ট
    0.82
     yon
    0.76
     Attention
    0.75
    ሓ
    0.74
     놓고
    0.72
     attention
    0.72
    บน
    0.72
     representado
    0.72
    Attention
    0.71
    Act Density 0.001%

    No Known Activations