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

    Neuronpedia

    INDEX
    Explanations

    phrases related to time or duration

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "through."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abet
    -0.79
    etus
    -0.71
    CTV
    -0.70
    \/
    -0.67
     inval
    -0.67
    inen
    -0.65
    istic
    -0.63
    IST
    -0.63
    uman
    -0.62
    ize
    -0.62
    POSITIVE LOGITS
    lace
    0.88
    put
    0.80
     intermedi
    0.76
    shaw
    0.71
    drive
    0.67
     Seasons
    0.67
     waivers
    0.67
     Labor
    0.66
    lined
    0.65
     clenched
    0.63
    Act Density 0.048%

    No Known Activations