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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     bubble
    -0.09
     plated
    -0.08
     exciting
    -0.08
     inserted
    -0.08
     bubb
    -0.07
     trendy
    -0.07
    结合
    -0.07
    Inserted
    -0.07
     kriter
    -0.07
    rut
    -0.07
    POSITIVE LOGITS
     siblings
    0.13
     spouses
    0.11
     родствен
    0.11
     निधन
    0.11
     aunt
    0.11
     classmates
    0.10
    兄
    0.10
    نۍ
    0.10
     spouse
    0.10
    姐妹
    0.10
    Act Density 0.054%

    No Known Activations