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

    Neuronpedia

    INDEX
    Explanations

    recall

    np_max-act · gemini-2.0-flash

    This neuron specifically responds to document structural headings or section titles (e.g. “Abstract,” “Preliminaries,” “Definitions,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    PLAY
    -0.07
    рам
    -0.07
    jwt
    -0.06
     Drew
    -0.06
     Leap
    -0.06
    prites
    -0.06
    _mul
    -0.06
    Big
    -0.06
    敬
    -0.06
     Reynolds
    -0.06
    POSITIVE LOGITS
    之
    0.07
     undo
    0.07
    aseline
    0.07
     رفتار
    0.07
     appId
    0.07
    <usize
    0.07
    Spi
    0.07
    (space
    0.06
     之
    0.06
     unto
    0.06
    Act Density 0.009%

    No Known Activations