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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating options or comparisons

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions that indicate options or alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ito
    -0.65
    ouf
    -0.57
     stripes
    -0.57
     Trip
    -0.56
     rentals
    -0.56
     gunmen
    -0.55
    ievers
    -0.55
    %%
    -0.55
    akin
    -0.54
     Ukrain
    -0.54
    POSITIVE LOGITS
    worldly
    0.83
     Accessed
    0.63
    ALE
    0.63
     happening
    0.62
    mentioned
    0.62
     Pastebin
    0.61
     Helpful
    0.61
    chance
    0.60
     happens
    0.59
     Thumbnails
    0.58
    Act Density 0.241%

    No Known Activations