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

    Neuronpedia

    INDEX
    Explanations

    years in the 1900s

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric decade references, i.e. four-digit year prefixes immediately followed by “s” (e.g. “1980s,” “1990s,” “70s”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .lineEdit
    -0.07
    .opts
    -0.06
    _rel
    -0.06
     [];
    -0.06
     سایر
    -0.06
     hồng
    -0.06
    .dequeueReusableCell
    -0.06
    貴
    -0.06
     >&
    -0.06
     vinegar
    -0.06
    POSITIVE LOGITS
     Reich
    0.07
    dictions
    0.07
    По
    0.07
     docks
    0.07
     pass
    0.06
    δρα
    0.06
    thumbs
    0.06
     condolences
    0.06
    aight
    0.06
    (history
    0.06
    Act Density 0.013%

    No Known Activations