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

    Neuronpedia

    INDEX
    Explanations

    phrases related to breaking out or breaking free

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating escape or breaking free from constraints

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     postp
    -0.78
     postponed
    -0.65
     Tsukuyomi
    -0.64
    ãĥĩãĤ£
    -0.63
     attendant
    -0.62
     Trip
    -0.60
     misses
    -0.59
     Jac
    -0.59
     dodged
    -0.58
     downgrade
    -0.57
    POSITIVE LOGITS
    ochet
    0.74
    ymes
    0.73
    taining
    0.73
    Sax
    0.73
    dden
    0.70
    Ò
    0.67
    EStreamFrame
    0.67
    tain
    0.65
    merce
    0.65
    sonian
    0.63
    Act Density 0.063%

    No Known Activations