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

    Neuronpedia

    INDEX
    Explanations

    Research paper abstracts

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric tokens (digit sequences) in author/address lines, effectively picking out postal codes and other numbers in affiliations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     fabulous
    -0.07
     mucho
    -0.07
     coloured
    -0.07
    .IsNullOr
    -0.06
    xFA
    -0.06
     stavu
    -0.06
     Neg
    -0.06
    _trigger
    -0.06
    셔
    -0.06
     jeg
    -0.06
    POSITIVE LOGITS
    VC
    0.07
    article
    0.07
    変
    0.07
    +h
    0.06
    	real
    0.06
    冲
    0.06
     painful
    0.06
    .Speed
    0.06
    必要
    0.06
    Venue
    0.06
    Act Density 0.004%

    No Known Activations