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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a restriction or limitation to specific conditions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "only" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    git
    -0.65
    idon
    -0.63
    ysis
    -0.62
    ffen
    -0.58
    liner
    -0.58
    cel
    -0.57
    insula
    -0.57
    claimer
    -0.56
     understatement
    -0.56
    duino
    -0.55
    POSITIVE LOGITS
     marginally
    0.97
     insofar
    0.96
     spor
    0.90
    onse
    0.85
     intermitt
    0.82
     incidentally
    0.82
     occasionally
    0.80
    ones
    0.75
     temporarily
    0.75
     partially
    0.74
    Act Density 0.062%

    No Known Activations