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

    Neuronpedia

    INDEX
    Explanations

    words related to descriptions or objects that one can own or possess

    oai_token-act-pair · gpt-3.5-turbo

    references to locations or origins associated with a variety of subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ratulations
    -0.81
    few
    -0.71
    idav
    -0.70
    seek
    -0.70
    merce
    -0.70
    faced
    -0.70
    hai
    -0.70
    lied
    -0.69
    icators
    -0.68
    threat
    -0.67
    POSITIVE LOGITS
     afar
    1.37
     whence
    1.09
     thence
    0.86
     anywhere
    0.80
     abroad
    0.80
     Uni
    0.79
     somewhere
    0.78
     Epic
    0.75
     scratch
    0.74
     underneath
    0.74
    Act Density 0.140%

    No Known Activations