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

    Neuronpedia

    INDEX
    Explanations

    specific terms related to controlled experiments or interventions

    oai_token-act-pair · gpt-3.5-turbo

    terms related to controlled environments or experiments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eeks
    -0.79
    ilk
    -0.78
    ouf
    -0.78
    endi
    -0.77
    sters
    -0.75
    ittal
    -0.73
    warts
    -0.71
    issues
    -0.70
    ager
    -0.70
    osaurs
    -0.70
    POSITIVE LOGITS
     demolition
    0.95
     doping
    0.83
    uled
    0.81
     demol
    0.80
     clinical
    0.79
     suicide
    0.75
     deletion
    0.75
     manslaughter
    0.75
     combustion
    0.74
     Parenthood
    0.74
    Act Density 0.126%

    No Known Activations