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

    Neuronpedia

    INDEX
    Explanations

    presence of formatted text or structural elements in written content

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

    Punctuation separating parts of sentences

    np_acts-logits-general · gemini-2.0-flash

    gender symbols and context words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     par
    -0.81
     ad
    -0.77
     in
    -0.77
     
    -0.77
     re
    -0.73
     bu
    -0.72
    — 
    -0.71
     qu
    -0.71
    WEBPACK
    -0.71
     at
    -0.70
    POSITIVE LOGITS
     myſelf
    1.67
     Anſ
    1.65
     purpoſe
    1.65
     Monfieur
    1.61
     Jefus
    1.58
     pleaſure
    1.56
     Houſe
    1.55
     itſelf
    1.54
     auffi
    1.54
     Majefty
    1.53
    Act Density 0.594%

    No Known Activations