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

    Neuronpedia

    INDEX
    Explanations

    words related to tendencies or inclinations

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating common tendencies or behaviors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     wish
    -0.67
    heed
    -0.66
     grant
    -0.63
     Siem
    -0.62
     retract
    -0.61
     complete
    -0.60
    ierce
    -0.58
     susp
    -0.57
    Sir
    -0.57
     blasp
    -0.56
    POSITIVE LOGITS
     tends
    2.69
     tended
    2.30
     tending
    1.53
     varies
    1.37
    usually
    1.23
    typically
    1.22
    pmwiki
    1.21
     usually
    1.19
     prefers
    1.17
    often
    1.16
    Act Density 0.024%

    No Known Activations