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

    Neuronpedia

    INDEX
    Explanations

    words that are being emphasized or highlighted

    oai_token-act-pair · gpt-3.5-turbo

    various forms of the suffix "ing" and related verb forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     diam
    -0.74
     exit
    -0.67
     cyn
    -0.62
     Pir
    -0.62
     Seym
    -0.61
     princip
    -0.60
    SOURCE
    -0.58
     Pwr
    -0.58
     Principal
    -0.57
     Mandatory
    -0.57
    POSITIVE LOGITS
    ings
    1.69
    INGS
    1.37
    ers
    1.25
    ING
    1.24
    ingly
    1.20
    als
    1.14
    ership
    1.12
    eless
    1.11
    ments
    1.10
    lers
    1.08
    Act Density 0.453%

    No Known Activations