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

    Neuronpedia

    INDEX
    Explanations

    phrases related to caution or warnings in political discourse

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

    cloud or Match

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

    locations that mark the end of a turn or sentence (sentence/turn boundaries).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    tagHelperRunner
    -0.59
     autorytatywna
    -0.56
     surla
    -0.47
    SpringRunner
    -0.46
    AddTagHelper
    -0.46
     ujednoznacz
    -0.45
    estacks
    -0.45
    rungsseite
    -0.43
     transférez
    -0.43
     CanadaChoose
    -0.42
    POSITIVE LOGITS
     חיצוניים
    0.60
    braucher
    0.41
     oprot
    0.41
    ALG
    0.39
     поко
    0.38
    bietern
    0.37
    новниш
    0.37
     obsession
    0.37
    idoscope
    0.36
    eadilan
    0.36
    Act Density 0.255%

    No Known Activations