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

    Neuronpedia

    INDEX
    Explanations

    quotations with attributions

    oai_token-act-pair · gpt-3.5-turbo

    sentences that conclude with a period

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     purse
    -0.95
     mosqu
    -0.85
     fermented
    -0.82
     metic
    -0.76
     submerged
    -0.75
     buggy
    -0.73
     extinguished
    -0.72
     swim
    -0.70
     brief
    -0.70
     fleeting
    -0.70
    POSITIVE LOGITS
    Downloadha
    1.03
     There
    0.93
     That
    0.93
     Then
    0.92
     Meanwhile
    0.90
     His
    0.90
     Those
    0.89
     But
    0.89
     Even
    0.88
     They
    0.88
    Act Density 0.694%

    No Known Activations