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

    Neuronpedia

    INDEX
    Explanations

    instances of the verb "to be" in different declensions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "re."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Bernstein
    -0.69
    inav
    -0.66
    ulence
    -0.64
    romeda
    -0.63
     Kissinger
    -0.63
     Merchant
    -0.63
     Distribution
    -0.61
    igraph
    -0.61
     Lizard
    -0.61
     SetTextColor
    -0.61
    POSITIVE LOGITS
    selves
    0.99
    ngth
    0.94
    agine
    0.93
     gonna
    0.93
    aper
    0.82
    joice
    0.80
    asons
    0.77
    initely
    0.76
     thankful
    0.76
    ceive
    0.73
    Act Density 0.049%

    No Known Activations