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

    Neuronpedia

    INDEX
    Explanations

    the word "tor" followed by varied suffixes and prefixes, indicating a focus on words related to torment or turmoil

    oai_token-act-pair · gpt-3.5-turbo

    references to torment or distress

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enegger
    -0.89
    ynski
    -0.82
    pired
    -0.69
    soDeliveryDate
    -0.69
    DragonMagazine
    -0.65
    Streamer
    -0.64
    ooks
    -0.64
    ortium
    -0.64
    wallet
    -0.64
    pires
    -0.63
    POSITIVE LOGITS
    mented
    0.96
    itational
    0.83
    rid
    0.78
    ching
    0.77
    vell
    0.76
    nel
    0.75
    iously
    0.74
    pid
    0.73
     Haunted
    0.73
    rik
    0.71
    Act Density 0.035%

    No Known Activations