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

    Neuronpedia

    INDEX
    Explanations

    phrases related to quality assessment or approval

    oai_token-act-pair · gpt-3.5-turbo

    expressions of unfavorable evaluations or criticisms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    beit
    -0.68
     irresist
    -0.65
     Neuroscience
    -0.63
    lessness
    -0.62
    iland
    -0.62
    ription
    -0.60
     generously
    -0.60
     Heath
    -0.59
    onomy
    -0.58
    rather
    -0.58
    POSITIVE LOGITS
     anymore
    1.43
     nor
    1.08
    ãĤ®
    0.83
    Ú
    0.82
    ÙIJ
    0.81
     yet
    0.74
    yet
    0.72
     either
    0.70
    vier
    0.69
     unless
    0.69
    Act Density 0.362%

    No Known Activations