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

    Neuronpedia

    INDEX
    Explanations

    Data management

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of output artifacts or saving steps in code (e.g. “analysis report,” “model,” “save”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     средство
    -0.07
     pill
    -0.07
     plated
    -0.07
    Evento
    -0.06
    valuator
    -0.06
    ValueType
    -0.06
    utto
    -0.06
     contamination
    -0.06
     boutique
    -0.06
     hookers
    -0.06
    POSITIVE LOGITS
     Expected
    0.07
     theoretically
    0.07
     FileNotFoundException
    0.07
     superv
    0.06
    foreign
    0.06
     Facing
    0.06
    There
    0.06
     resentment
    0.06
     بان
    0.06
     Lovely
    0.06
    Act Density 0.073%

    No Known Activations