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

    Neuronpedia

    INDEX
    Explanations

    Beach activities

    np_max-act · gemini-2.0-flash

    This neuron fires on gerund (-ing) verbs that describe leisure or relaxation activities (e.g., “lounging,” “relaxing”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     kale
    -0.06
    alizace
    -0.06
     liar
    -0.06
     Stress
    -0.06
    avings
    -0.06
    нка
    -0.06
     circle
    -0.06
     molest
    -0.06
    .Text
    -0.06
     enriched
    -0.06
    POSITIVE LOGITS
    .plugin
    0.07
    onitor
    0.07
     شماره
    0.06
    producer
    0.06
    ewith
    0.06
    .mon
    0.06
    connector
    0.06
     message
    0.06
    _ONE
    0.06
    accumulate
    0.06
    Act Density 0.017%

    No Known Activations