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

    Neuronpedia

    INDEX
    Explanations

    the exact word "Den" (capital D) in the text.

    oai_token-act-pair · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Negative Logits
     myſelf
    -1.89
     itſelf
    -1.80
     Efq
    -1.79
     Monfieur
    -1.70
     Theſe
    -1.66
     themſelves
    -1.65
     pleaſure
    -1.65
     himſelf
    -1.61
     Anſ
    -1.60
     Houſe
    -1.57
    POSITIVE LOGITS
     de
    1.00
     del
    0.90
     en
    0.85
     d
    0.81
     
    0.77
     und
    0.76
     di
    0.75
     par
    0.74
     des
    0.74
     du
    0.74
    Act Density 0.002%

    No Known Activations