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

    Neuronpedia

    INDEX
    Explanations

    way

    np_max-act · gemini-2.0-flash

    This neuron detects descriptive modifiers of manner or degree—i.e. adverbs and adjectival phrases that qualify how something is done or the extent of a quality.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     unter
    -0.07
    _TRNS
    -0.07
     �
    -0.06
     チ
    -0.06
     sixty
    -0.06
     ten
    -0.06
     плеч
    -0.06
     Monter
    -0.06
    _CURRENT
    -0.06
     Arist
    -0.06
    POSITIVE LOGITS
     way
    0.07
     Ways
    0.07
     mundo
    0.07
     ways
    0.07
     Love
    0.06
    StreamReader
    0.06
    аков
    0.06
    لل
    0.06
     이해
    0.06
    한국
    0.06
    Act Density 0.026%

    No Known Activations