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

    Neuronpedia

    INDEX
    Explanations

    negatively charged words or phrases

    oai_token-act-pair · gpt-3.5-turbo

    expressions of frustration or dismissal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atu
    -0.69
    imir
    -0.63
     plent
    -0.60
    æ©
    -0.59
    issance
    -0.59
    agnar
    -0.58
     immortality
    -0.57
     tremend
    -0.57
    oppable
    -0.56
    ometimes
    -0.55
    POSITIVE LOGITS
     slightest
    1.18
     dime
    1.05
    enance
    1.01
     nor
    1.01
     ounce
    1.00
     whatsoever
    0.98
     mention
    0.87
     penny
    0.87
     inch
    0.86
     bothered
    0.81
    Act Density 0.218%

    No Known Activations