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

    Neuronpedia

    INDEX
    Explanations

    words with the substring 'st' often occuring multiple times in combination with preceding characters

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "st."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     thumbs
    -0.70
     deaf
    -0.66
     Haram
    -0.64
     Peel
    -0.61
     Tate
    -0.60
     amy
    -0.59
     timely
    -0.59
    CENT
    -0.59
    sets
    -0.58
     disabling
    -0.57
    POSITIVE LOGITS
    oppers
    1.09
    rup
    1.09
    ellation
    1.06
    hetics
    1.05
    itute
    1.05
    alker
    1.03
    ral
    1.02
    retch
    1.01
    rophe
    1.01
    oppable
    1.00
    Act Density 0.029%

    No Known Activations