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

    Neuronpedia

    INDEX
    Explanations

    ,

    np_max-act · gemini-2.0-flash

    phrases describing betrayal or treachery in a narrative context.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects occurrences of direct speech marked by first‐ and second‐person pronouns and imperatives (e.g. “I,” “you,” “let,” “have,” “say”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    γον
    -0.07
     <$
    -0.06
     wallets
    -0.06
     Depend
    -0.06
    /services
    -0.06
    migration
    -0.06
     Rit
    -0.06
    _txn
    -0.06
    ootball
    -0.06
    чний
    -0.06
    POSITIVE LOGITS
     ≠
    0.07
     الکتر
    0.06
    .accel
    0.06
    _titles
    0.06
     З
    0.06
    elements
    0.06
    _render
    0.06
    _PARAMETERS
    0.06
     performing
    0.06
    uluğu
    0.06
    Act Density 0.015%

    No Known Activations