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

    Neuronpedia

    INDEX
    Explanations

    had

    np_max-act · gemini-2.0-flash

    The neuron activates on phrases that state someone “had” a number of “children,” i.e. mentions of a person’s offspring count.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -end
    -0.07
     MAN
    -0.07
    _SCROLL
    -0.07
    นเต
    -0.06
    271
    -0.06
    тин
    -0.06
    εις
    -0.06
    _BITS
    -0.06
     dull
    -0.06
    <w
    -0.06
    POSITIVE LOGITS
     StringIO
    0.07
     QLD
    0.07
     midway
    0.07
     Blockly
    0.07
    ($.
    0.07
    gements
    0.07
    ='')
    0.06
    诗
    0.06
    .compiler
    0.06
     Nicholson
    0.06
    Act Density 0.009%

    No Known Activations