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

    Neuronpedia

    INDEX
    Explanations

    indications of gratitude and acknowledgment within the text

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

    the neuron responds to document-structure tokens such as paragraph/line breaks and other structural or boilerplate boundaries (e.g., section starts, <bos>, empty lines).

    oai_token-act-pair · gpt-5-miniTriggered by @phamvanluong2003
    New Auto-Interp
    Negative Logits
    ugeot
    -1.02
     FormsModule
    -0.88
    Rhestr
    -0.86
     FetchType
    -0.78
    cửa
    -0.78
     createSprite
    -0.78
    ruptedException
    -0.77
     BoxFit
    -0.75
    daction
    -0.74
    iffance
    -0.74
    POSITIVE LOGITS
    ↵↵
    1.83
    <h2>
    1.40
    ↵↵↵
    1.35
    ↵↵↵↵
    1.31
    <eos>
    1.27
    ↵↵↵↵↵
    1.26
    ↵
    1.22
    <h3>
    1.17
    <blockquote>
    1.15
    ↵↵↵↵↵↵
    1.14
    Act Density 1.050%

    No Known Activations