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

    Neuronpedia

    INDEX
    Explanations

    instances of actions related to physically striking or knocking on objects or surfaces

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "knock" or variations of it, particularly in the context of knocking actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    UF
    -0.75
    NT
    -0.73
    Cod
    -0.71
    NH
    -0.70
    Component
    -0.67
    Dist
    -0.67
    ELF
    -0.66
    MA
    -0.66
    States
    -0.65
    Online
    -0.65
    POSITIVE LOGITS
     behalf
    1.32
    erous
    1.00
    shore
    0.86
     steroids
    0.81
     autop
    0.81
    eness
    0.80
     fumes
    0.77
    etime
    0.77
     rooft
    0.75
    wards
    0.72
    Act Density 0.130%

    No Known Activations