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

    Neuronpedia

    INDEX
    Explanations

    phrases related to hitting or impacting things, especially in a physical sense

    oai_token-act-pair · gpt-3.5-turbo

    references to knocking and its effects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    natureconservancy
    -0.94
    BIL
    -0.87
    bia
    -0.84
    erve
    -0.80
    ngth
    -0.79
    umn
    -0.78
    cia
    -0.77
    tions
    -0.76
    ESA
    -0.72
    URI
    -0.72
    POSITIVE LOGITS
    vu
    0.77
     pickup
    0.75
     tumble
    0.74
     Jensen
    0.74
     punch
    0.73
     Sasuke
    0.72
     blow
    0.72
     concrete
    0.67
     Hats
    0.66
     swat
    0.66
    Act Density 0.063%

    No Known Activations