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

    Neuronpedia

    INDEX
    Explanations

    programming constructs and syntax elements

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

    code constants and definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    parsedMessage
    -0.94
    IsMutable
    -0.90
    <unused14>
    -0.85
    <unused41>
    -0.85
    ſehen
    -0.84
    <unused52>
    -0.84
    <unused28>
    -0.84
    <unused8>
    -0.84
    <unused3>
    -0.84
    [@BOS@]
    -0.84
    POSITIVE LOGITS
    .
    0.44
    ↵↵
    0.41
     The
    0.39
     and
    0.36
     disciplina
    0.35
    ).
    0.30
    ↵
    0.30
     Learning
    0.30
    Kön
    0.30
     Mark
    0.30
    Act Density 0.043%

    No Known Activations