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

    Neuronpedia

    INDEX
    Explanations

    reports and studies

    np_max-act · gemini-2.0-flash

    The neuron activates on meta‐scientific reporting language in research papers—particularly verbs and phrases like “report(s),” “this report,” or “paper reports” that introduce the study’s findings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     photo
    -0.07
    AZE
    -0.06
     cart
    -0.06
    [`
    -0.06
     banned
    -0.06
    求购
    -0.06
    AVE
    -0.06
     dye
    -0.06
     fluor
    -0.06
     raced
    -0.06
    POSITIVE LOGITS
    ước
    0.07
     Ellie
    0.07
    _logits
    0.06
    Having
    0.06
    wordpress
    0.06
     thwart
    0.06
    ARGS
    0.06
     roaring
    0.06
    .Dictionary
    0.06
    <Expression
    0.06
    Act Density 0.046%

    No Known Activations