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

    Neuronpedia

    INDEX
    Explanations

    phrases related to hitting or reaching success or a positive outcome

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express success or achievement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iterator
    -0.69
    urdue
    -0.67
    umn
    -0.67
    sbm
    -0.65
    rique
    -0.65
    href
    -0.65
    lege
    -0.63
    tif
    -0.63
    igation
    -0.62
    lance
    -0.62
    POSITIVE LOGITS
     snag
    1.14
     stride
    1.13
     brakes
    1.10
     nail
    0.95
     pause
    0.93
     button
    0.93
     mark
    0.93
     pavement
    0.85
     reset
    0.84
     jack
    0.84
    Act Density 0.100%

    No Known Activations