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

    Neuronpedia

    INDEX
    Explanations

    words related to actions that involve separation or isolation, as well as words related to exaggeration or infiltration

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to isolation and exaggeration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Dragonbound
    -0.72
    master
    -0.70
     Hath
    -0.69
    dash
    -0.69
    SHIP
    -0.68
     Maid
    -0.68
     miscarriage
    -0.68
    CHAT
    -0.67
    dress
    -0.67
    beard
    -0.66
    POSITIVE LOGITS
    ating
    1.19
    ates
    1.18
    ators
    1.14
    atable
    1.11
    isions
    1.10
    asion
    1.09
    ator
    1.08
    ations
    1.07
    ading
    1.07
    ocations
    1.06
    Act Density 0.054%

    No Known Activations