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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the underscore character, often in the context of naming or defining variables and functions

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

    Tokens starting with "__" followed by other characters

    np_acts-logits-general · gemini-2.0-flash

    C/C++ bit-field declarations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rem
    -0.61
     Felix
    -0.60
    Egypte
    -0.60
     rot
    -0.57
    ため
    -0.57
    瓜
    -0.55
    ,
    -0.55
    帖最后由
    -0.54
    Geplaatst
    -0.53
    SNAPSHOT
    -0.52
    POSITIVE LOGITS
    tagHelperRunner
    1.17
    /__
    0.94
    出版年
    0.93
     (!__
    0.92
    ->__
    0.90
    _.
    0.89
     (__
    0.88
     שוליים
    0.85
    [__
    0.83
    #__
    0.82
    Act Density 0.223%

    No Known Activations