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

    Neuronpedia

    INDEX
    Explanations

    instances of sentence punctuation and structural markers, particularly periods and parentheses

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

    questionnaire and xkcd discussions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lenker
    -0.74
    parsedMessage
    -0.71
    ագրություններ
    -0.69
     disambiguazione
    -0.68
    protoimpl
    -0.67
     poffe
    -0.66
     queſta
    -0.66
    -0.65
    ButterKnife
    -0.65
    rrggbb
    -0.64
    POSITIVE LOGITS
    <bos>
    0.47
    mathbb
    0.45
    .
    0.43
     .
    0.40
    <0xE2>
    0.40
     strict
    0.39
     quase
    0.38
    s
    0.37
     large
    0.36
     
    0.35
    Act Density 0.002%

    No Known Activations