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

    Neuronpedia

    INDEX
    Explanations

    actions involving manipulation or transformation of objects or information

    oai_token-act-pair · gpt-3.5-turbo

    actions that involve manipulating or processing information or materials

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Cosponsors
    -0.70
    ooters
    -0.67
    ooter
    -0.67
    theless
    -0.66
    nia
    -0.64
    BuyableInstoreAndOnline
    -0.63
    ieth
    -0.61
    udence
    -0.61
    enny
    -0.59
    uchin
    -0.57
    POSITIVE LOGITS
     them
    1.33
    them
    1.11
     accordingly
    1.11
     THEM
    1.08
     therein
    0.94
     Them
    0.94
     it
    0.93
     thereof
    0.84
     theirs
    0.84
     corresponding
    0.83
    Act Density 0.569%

    No Known Activations