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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "multi" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    the prefix "mult" indicating something that is multiple or various

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    OPLE
    -0.82
     Defenders
    -0.74
     Wolves
    -0.73
    çĶ°
    -0.71
    HOME
    -0.71
     Warriors
    -0.69
     Mane
    -0.67
     Tigers
    -0.66
    ãģį
    -0.66
    =-=-=-=-=-=-=-=-
    -0.66
    POSITIVE LOGITS
    icol
    1.33
    ilingual
    1.21
    itud
    1.17
    iversal
    1.12
    ivari
    1.12
    ivariate
    1.12
    ilateral
    1.09
    ipl
    1.06
    imedia
    1.05
    icultural
    1.02
    Act Density 0.012%

    No Known Activations