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

    Neuronpedia

    INDEX
    Explanations

    contractions of 'have' like "I've" in the text data

    oai_token-act-pair · gpt-3.5-turbo

    instances of the contraction "we've"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    é¾įå¥ij士
    -0.70
    ologically
    -0.69
     fracturing
    -0.64
     lapt
    -0.60
     obser
    -0.60
    soever
    -0.58
    matically
    -0.57
     Anon
    -0.57
    ItemTracker
    -0.56
    interstitial
    -0.56
    POSITIVE LOGITS
    tt
    0.96
    ve
    0.85
    tti
    0.85
    mber
    0.83
    ggie
    0.81
    ices
    0.79
    gard
    0.78
    tta
    0.78
    ggies
    0.76
    tted
    0.76
    Act Density 0.025%

    No Known Activations