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

    Neuronpedia

    INDEX
    Explanations

    phrases that indicate suitability or perfection for a purpose

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

    perfect for a purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    WriteLiteral
    -0.46
     Pemain
    -0.45
     Kebijakan
    -0.36
    parsedMessage
    -0.36
     fowl
    -0.35
    stateMutability
    -0.35
    codings
    -0.35
    把自己
    -0.34
    ündete
    -0.34
     oiseaux
    -0.34
    POSITIVE LOGITS
     perfect
    1.15
    perfect
    1.08
     ideal
    1.05
     PERFECT
    1.04
     Perfect
    0.98
    Perfect
    0.97
    ideal
    0.96
     Ideal
    0.91
    Ideal
    0.90
     IDEAL
    0.90
    Act Density 0.005%

    No Known Activations