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

    Neuronpedia

    INDEX
    Explanations

    phrases with the word "these"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "these."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    achus
    -0.74
    cohol
    -0.72
    obe
    -0.71
    andum
    -0.70
    iosis
    -0.68
    ë
    -0.67
    Adds
    -0.67
    ģ«
    -0.67
    atform
    -0.66
    sein
    -0.65
    POSITIVE LOGITS
     guys
    1.29
     kinds
    1.09
     sorts
    1.01
     folks
    1.00
     dudes
    1.00
     fellows
    1.00
     days
    0.96
     gentlemen
    0.92
     idiots
    0.91
     kids
    0.89
    Act Density 0.102%

    No Known Activations