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

    Neuronpedia

    INDEX
    Explanations

    end-of-turn markers in dialogue or interaction

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

    preceding the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sobie
    -0.32
     hubieran
    -0.29
    warten
    -0.29
     mengenal
    -0.29
     olisi
    -0.29
     cuantas
    -0.29
    Tembelea
    -0.28
    îtra
    -0.28
    onSearch
    -0.28
    cing
    -0.28
    POSITIVE LOGITS
     aforementioned
    0.66
     disponibilités
    0.65
     preceding
    0.60
    دانشنامهٔ
    0.57
     foregoing
    0.56
     sumpay
    0.56
     gynhyrchwyd
    0.56
    بوابة
    0.55
     entirety
    0.54
    DispatchToProps
    0.54
    Act Density 0.121%

    No Known Activations