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

    Neuronpedia

    INDEX
    Explanations

    adjectives describing thickness or density

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "thick."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    FAULT
    -0.83
    OTA
    -0.82
    DB
    -0.76
    Particip
    -0.74
    USA
    -0.70
    AMY
    -0.68
     Controlled
    -0.67
    GEN
    -0.66
    SEC
    -0.65
    428
    -0.65
    POSITIVE LOGITS
    ening
    1.32
    ened
    1.29
    nesses
    1.14
    ener
    1.02
    ness
    1.00
     thickness
    0.99
     thick
    0.98
     nodd
    0.96
     thicker
    0.95
    ciating
    0.94
    Act Density 0.009%

    No Known Activations