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

    Neuronpedia

    INDEX
    Explanations

    short phrases or sayings

    oai_token-act-pair · gpt-3.5-turbo

    references to well-known sayings or adages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     subcommittee
    -0.63
    von
    -0.62
    ega
    -0.61
    edited
    -0.60
    ensibly
    -0.59
    addons
    -0.59
    ADS
    -0.59
    changes
    -0.59
     finalized
    -0.59
     pending
    -0.59
    POSITIVE LOGITS
     Thou
    1.03
    Always
    0.99
     Always
    0.95
     Beware
    0.95
     beware
    0.85
    Never
    0.84
     shalt
    0.82
     '[
    0.80
     persever
    0.80
     Never
    0.78
    Act Density 0.439%

    No Known Activations