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

    Neuronpedia

    INDEX
    Explanations

    articles and prepositions

    np_max-act · gemini-2.0-flash

    The neuron flags the ordinary English “narrative” words that appear in math‐heavy exposition—i.e. the connective and method-description tokens (“we,” “using,” “call,” “involve,” etc.) in a formal mathematics text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    іт
    -0.07
     Judy
    -0.06
    (cols
    -0.06
     Sith
    -0.06
     Ath
    -0.06
    (K
    -0.06
    _negative
    -0.06
     migraine
    -0.06
    \R
    -0.06
     venture
    -0.06
    POSITIVE LOGITS
     státy
    0.07
    IÓN
    0.06
    νό
    0.06
    Tool
    0.06
    _SYN
    0.06
     xlim
    0.06
    rafted
    0.06
    STRUCTION
    0.06
     CAST
    0.06
    _projection
    0.06
    Act Density 0.028%

    No Known Activations