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

    Neuronpedia

    INDEX
    Explanations

    the word "Tu" followed by various numbers

    oai_token-act-pair · gpt-3.5-turbo

    the word "Tu" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     SPD
    -0.70
     constants
    -0.67
     Rebirth
    -0.65
     Authorization
    -0.58
    rings
    -0.58
     angels
    -0.57
     Colossus
    -0.57
    Ö¼
    -0.56
     Reson
    -0.56
    âĸ¬
    -0.56
    POSITIVE LOGITS
    fts
    1.24
    ples
    1.21
    kson
    1.12
    culosis
    1.08
    pling
    1.05
    esta
    1.04
    Tu
    1.03
    cci
    1.01
    ition
    1.00
    fty
    0.99
    Act Density 0.023%

    No Known Activations