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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "depart."

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

    mathematical expressions and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     loren
    -0.43
    ździer
    -0.42
     Conse
    -0.41
    gués
    -0.40
    .`
    -0.39
    ."""
    -0.39
     pylint
    -0.38
    ."</
    -0.38
     logr
    -0.36
    inheritDoc
    -0.36
    POSITIVE LOGITS
    <bos>
    0.68
     engraçadas
    0.60
     Мексичка
    0.53
    OGND
    0.52
     المعيارى
    0.52
     ​​
    0.52
     مرئيه
    0.52
    queryInterface
    0.52
    providedIn
    0.51
    ########.
    0.50
    Act Density 0.020%

    No Known Activations