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

    Neuronpedia

    INDEX
    Explanations

    content-related actions or operations

    oai_token-act-pair · gpt-3.5-turbo

    technical terms or references related to programming and legal documentation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    orum
    -0.75
    ire
    -0.69
    advertising
    -0.68
    ]).
    -0.65
    Far
    -0.63
    ypes
    -0.62
    "}
    -0.62
    yang
    -0.61
    dor
    -0.61
     ..."
    -0.61
    POSITIVE LOGITS
     moreover
    0.93
     however
    0.89
     therefore
    0.78
     furthermore
    0.76
     though
    0.65
     accordingly
    0.63
     ital
    0.58
     extensively
    0.57
     amazed
    0.57
     reasoning
    0.55
    Act Density 1.075%

    No Known Activations