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

    Neuronpedia

    INDEX
    Explanations

    growth

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on domain-specific jargon or technical nouns (e.g. “growth,” “mechanism”) rather than common function words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _matching
    -0.07
    .Custom
    -0.07
     climbs
    -0.06
    NAV
    -0.06
    при
    -0.06
    devices
    -0.06
    _best
    -0.06
    =com
    -0.06
     tat
    -0.06
    LOCITY
    -0.06
    POSITIVE LOGITS
     využí
    0.07
     BCHP
    0.06
    bf
    0.06
    页面存档备份
    0.06
     chvíli
    0.06
    删除成功
    0.06
    (-
    0.06
     文章
    0.06
     uma
    0.06
    688
    0.06
    Act Density 0.031%

    No Known Activations