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

    Neuronpedia

    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron spikes strongly on sentence-ending periods (full stops), effectively marking sentence boundaries.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    terms related to chemical safety.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cold
    -0.08
    Ps
    -0.08
     osobních
    -0.07
     допомоги
    -0.07
    _publish
    -0.07
     QDir
    -0.07
    زارش
    -0.07
    екту
    -0.07
    -prev
    -0.06
     prevalent
    -0.06
    POSITIVE LOGITS
     requester
    0.06
    ]?
    0.06
     möchten
    0.06
     информа
    0.06
    ौर
    0.06
    .longitude
    0.06
    tal
    0.06
    0.06
    '
↵
↵
    0.06
    .segment
    0.05
    Act Density 0.335%

    No Known Activations