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

    Neuronpedia

    INDEX
    Explanations

    emphatic phrases indicating a lack of exception or evidence

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

    absolute negation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     éc
    -0.52
    ruled
    -0.50
     Lens
    -0.49
     Biôgrafia
    -0.49
     FetchType
    -0.48
    ainville
    -0.47
    majánló
    -0.47
    榄
    -0.47
     Ephesus
    -0.47
    AnimationsModule
    -0.47
    POSITIVE LOGITS
     whatsoever
    1.84
    soever
    1.03
     whatever
    0.69
    whatever
    0.64
    osoever
    0.59
     mitään
    0.59
     jamás
    0.58
    一切
    0.56
     EVER
    0.56
     keinerlei
    0.55
    Act Density 0.003%

    No Known Activations