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

    Neuronpedia

    INDEX
    Explanations

    verbs in conditional sentences

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "were to" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     dynamics
    -0.71
     nets
    -0.70
     handles
    -0.69
     equations
    -0.67
     transmissions
    -0.66
     handling
    -0.65
     tcp
    -0.63
     deficits
    -0.62
     Unreal
    -0.61
     VIDEOS
    -0.61
    POSITIVE LOGITS
    wered
    1.28
    asted
    1.15
    lled
    1.12
    pless
    1.09
    othy
    0.99
    tted
    0.96
    ying
    0.96
    asting
    0.96
    iling
    0.93
    dden
    0.92
    Act Density 0.091%

    No Known Activations