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

    Neuronpedia

    INDEX
    Explanations

    programming-related syntax elements and parameters

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

    parameter declarations in code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     indeed
    -0.29
     acá
    -0.25
     asimismo
    -0.23
     atque
    -0.23
    ↵↵
    -0.23
     eikä
    -0.22
     here
    -0.22
     poème
    -0.22
    ulement
    -0.21
     aquí
    -0.21
    POSITIVE LOGITS
     ddelwed
    1.12
     للمعارف
    1.11
    <unused14>
    1.10
    [@BOS@]
    1.10
    <unused16>
    1.09
    <unused8>
    1.09
    <unused17>
    1.09
    <unused23>
    1.09
    <pad>
    1.09
    <unused3>
    1.09
    Act Density 0.009%

    No Known Activations