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

    Neuronpedia

    INDEX
    Explanations

    verbs indicating mental actions

    oai_token-act-pair · gpt-3.5-turbo

    verbs conveying knowledge, awareness, or emotional states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anwhile
    -0.69
     Annotations
    -0.68
    agher
    -0.67
    java
    -0.67
     opting
    -0.64
    merce
    -0.63
    enegger
    -0.63
    quart
    -0.62
     withdrawing
    -0.62
    arta
    -0.62
    POSITIVE LOGITS
     itself
    0.90
     occupants
    0.66
    erers
    0.66
    erer
    0.65
    ickets
    0.64
    lessly
    0.64
    ibly
    0.61
    rive
    0.60
     lur
    0.60
    wer
    0.58
    Act Density 0.676%

    No Known Activations