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

    Neuronpedia

    INDEX
    Explanations

    various forms of punctuation and question marks in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    RegressionTest
    -0.94
    -0.72
    RegistryLite
    -0.67
    InjectAttribute
    -0.61
    <bos>
    -0.60
    ...
    -0.56
     umano
    -0.56
     personalidade
    -0.55
     רבה
    -0.54
    onAttach
    -0.54
    POSITIVE LOGITS
    ?)
    1.34
    ?]
    1.34
    ?,
    1.22
    ?),
    1.20
    ?";
    1.20
    ?):
    1.15
    ?",
    1.07
    ?).
    1.07
    ?;
    1.06
    !,
    1.06
    Act Density 0.198%

    No Known Activations