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

    Neuronpedia

    INDEX
    Explanations

    instances of verbs related to actions or operations

    oai_token-act-pair · gpt-3.5-turbo

    instances of hacking and related technological vulnerabilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    itaire
    -0.65
    pex
    -0.58
     Doyle
    -0.57
    emale
    -0.56
    lished
    -0.55
    llah
    -0.54
    lance
    -0.54
    etheless
    -0.54
    jri
    -0.54
    aml
    -0.54
    POSITIVE LOGITS
     themselves
    0.88
     selves
    0.80
    MpServer
    0.70
    selves
    0.67
     orbits
    0.62
     mouths
    0.62
     helmets
    0.62
     coats
    0.59
     uniforms
    0.58
     necks
    0.57
    Act Density 0.700%

    No Known Activations