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

    Neuronpedia

    INDEX
    Explanations

    references to political events or figures

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

    past tense verbs and sequences

    np_acts-logits-general · gemini-2.5-flash-lite

    the neuron highlights salient, information-dense tokens—important content words (main verbs, nouns, numbers) and emphatic punctuation that carry the core facts or claims.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    MethodManager
    -0.73
     kasarigan
    -0.64
    IsMutable
    -0.63
    HandlerContext
    -0.59
    ImageContext
    -0.52
     CallOverrides
    -0.52
    ValueStyle
    -0.51
    WebVitals
    -0.51
     ProtoMessage
    -0.50
    AnchorStyles
    -0.50
    POSITIVE LOGITS
    บ้าง
    0.35
    magini
    0.35
    ⓧ
    0.34
     Schles
    0.32
     SOME
    0.31
    ありがとうございます
    0.31
     tec
    0.30
     aussch
    0.30
     non
    0.30
     svet
    0.30
    Act Density 0.307%

    No Known Activations