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

    Neuronpedia

    INDEX
    Explanations

    variable names and types in programming code

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

    variable assignments or initialization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ब्रेकडाउन
    -1.02
    MLLoader
    -0.99
    <unused14>
    -0.95
    <unused41>
    -0.95
    [@BOS@]
    -0.95
    <unused52>
    -0.94
    <unused74>
    -0.94
    <unused23>
    -0.94
    <unused8>
    -0.94
    <unused16>
    -0.94
    POSITIVE LOGITS
     is
    0.40
    ↵↵
    0.35
     alone
    0.32
    ↵
    0.30
     fijo
    0.30
    holdet
    0.28
     seul
    0.27
    .
    0.27
     package
    0.26
    '
    0.26
    Act Density 0.003%

    No Known Activations