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

    Neuronpedia

    INDEX
    Explanations

    DOI URLs

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric sequences and punctuation that form DOI links (e.g., digit-dot-digit patterns and slashes in “doi.org/10.xxxx/…” URLs).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Coc
    -0.07
     nin
    -0.07
     cat
    -0.07
     přih
    -0.06
     Lan
    -0.06
     Cant
    -0.06
     ptr
    -0.06
    _pad
    -0.06
    hodob
    -0.06
     Ngh
    -0.06
    POSITIVE LOGITS
    λού
    0.07
    ığı
    0.07
    rei
    0.07
     '">'
    0.07
    -support
    0.06
    び
    0.06
    Visual
    0.06
    	error
    0.06
     Alibaba
    0.06
    ­i
    0.06
    Act Density 0.002%

    No Known Activations