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

    Neuronpedia

    INDEX
    Explanations

    programmatic structures and constructs in code

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

    code and math delimiters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Datuak
    -0.97
    <pad>
    -0.89
    <unused23>
    -0.89
    <unused79>
    -0.89
    <unused3>
    -0.89
    [@BOS@]
    -0.89
    <unused16>
    -0.88
    <unused28>
    -0.88
    <unused14>
    -0.88
    <unused8>
    -0.88
    POSITIVE LOGITS
    ↵
    0.36
    ↵↵
    0.35
        
    0.27
    //
    0.25
     precies
    0.25
    .
    0.24
     brief
    0.23
     
    0.23
    1
    0.23
    沿
    0.23
    Act Density 0.014%

    No Known Activations