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

    Neuronpedia

    INDEX
    Explanations

    end punctuation and structure within the text

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

    ending phrases like `right`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    -0.30
     everything
    -0.28
    maxcdn
    -0.26
      
    -0.25
     inter
    -0.24
     $
    -0.24
     you
    -0.24
     polish
    -0.24
    टे
    -0.23
     or
    -0.23
    POSITIVE LOGITS
     kasarigan
    1.00
    oredCriteria
    0.93
     queſta
    0.84
     surla
    0.83
    AddHtmlAttribute
    0.79
    BufferException
    0.77
    <unused41>
    0.75
    <unused43>
    0.75
    <unused8>
    0.75
    [@BOS@]
    0.75
    Act Density 0.043%

    No Known Activations