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

    Neuronpedia

    INDEX
    Explanations

    phrases related to instructions and guidelines for communication

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

    writing product descriptions or formal communication

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ]-->
    -0.31
     unheard
    -0.29
     Idee
    -0.28
     aandacht
    -0.28
    RefNanny
    -0.28
     liggen
    -0.28
     propOrder
    -0.28
     mejoras
    -0.27
    pektive
    -0.27
     exclusiva
    -0.27
    POSITIVE LOGITS
     documents
    0.88
     texts
    0.84
     emails
    0.80
     captions
    0.79
    MLLoader
    0.79
     descriptions
    0.78
     correspondence
    0.77
     inscriptions
    0.77
     documentation
    0.76
     conversations
    0.76
    Act Density 0.082%

    No Known Activations