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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    conversational cues in a text-based interaction.

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

    The neuron fires on instruction or directive language—i.e. imperative verbs and formatting commands (e.g. “Do,” “Use,” “without,” etc.) in the prompt.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    फ
    -0.07
    intendo
    -0.06
    /groups
    -0.06
    mall
    -0.06
     двер
    -0.06
    hes
    -0.06
     fung
    -0.06
    Senator
    -0.06
    _SP
    -0.06
     �
    -0.06
    POSITIVE LOGITS
    ('<?
    0.07
    (Locale
    0.07
    .Information
    0.07
    -transition
    0.06
     Observation
    0.06
    date
    0.06
    -general
    0.06
    ้งาน
    0.06
     Qi
    0.06
    (&$
    0.06
    Act Density 0.017%

    No Known Activations