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

    Neuronpedia

    INDEX
    Explanations

    words related to subversion and influence

    oai_token-act-pair · gpt-3.5-turbo

    prefixes and root words that imply subversion or opposition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onement
    -0.66
     Tsukuyomi
    -0.64
     pinch
    -0.63
     viewer
    -0.63
    ottage
    -0.62
    HOME
    -0.62
     reader
    -0.62
    ocene
    -0.61
    ings
    -0.61
    onian
    -0.60
    POSITIVE LOGITS
    etheless
    1.15
    hibited
    1.15
    formed
    1.11
    actively
    1.10
    ceed
    1.07
    cluded
    1.07
    operated
    1.02
    cedes
    1.00
    figured
    0.98
    versely
    0.98
    Act Density 0.105%

    No Known Activations