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

    Neuronpedia

    INDEX
    Explanations

    phrases related to competitions or series

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to rankings or ratings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Arabian
    -0.66
     Aval
    -0.66
     revel
    -0.65
     veins
    -0.64
     brim
    -0.63
     CSI
    -0.61
     eclipse
    -0.60
     Carib
    -0.59
     Rouge
    -0.58
     needles
    -0.58
    POSITIVE LOGITS
    sized
    1.16
    selling
    1.16
    sounding
    1.12
    nat
    1.10
    looking
    1.06
    enough
    1.05
    fit
    1.04
    luck
    1.03
    eff
    1.00
    known
    1.00
    Act Density 0.040%

    No Known Activations