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

    Neuronpedia

    INDEX
    Explanations

    symbols and special characters in structured mathematical or formal text

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

    quoted text starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     betweenstory
    -0.94
    rungsseite
    -0.94
     ujednoznacz
    -0.91
    ſſung
    -0.89
     imagui
    -0.89
    majánló
    -0.88
    uxxxx
    -0.88
    SharedDtor
    -0.87
    <unused20>
    -0.86
    <unused3>
    -0.85
    POSITIVE LOGITS
    Therefore
    0.30
     therefore
    0.30
     there
    0.27
     we
    0.26
     once
    0.25
    while
    0.24
    Once
    0.23
     Therefore
    0.23
     it
    0.23
     while
    0.23
    Act Density 0.004%

    No Known Activations