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

    Neuronpedia

    INDEX
    Explanations

    jonathan

    np_max-act · gemini-2.0-flash

    assistant role/speaker labels in chat-style headers or metadata.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Donation
    -0.06
     Toll
    -0.06
     Ок
    -0.06
     Serial
    -0.06
    -0.06
     transcription
    -0.06
     vape
    -0.06
     گرف
    -0.06
     hon
    -0.06
     Düş
    -0.06
    POSITIVE LOGITS
    특별
    0.07
    insn
    0.07
    .LAZY
    0.07
     although
    0.06
    RetVal
    0.06
    itic
    0.06
     모든
    0.06
    .setDate
    0.06
     bý
    0.06
    symbol
    0.06
    Act Density 0.014%

    No Known Activations