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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "subscribe" and related phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eger
    -0.18
    wards
    -0.17
    umat
    -0.17
    sten
    -0.16
    ern
    -0.16
    c
    -0.15
    ward
    -0.15
    139
    -0.15
    á»ī
    -0.14
    akis
    -0.14
    POSITIVE LOGITS
    'gc
    0.17
    iglia
    0.17
    .unsubscribe
    0.15
    ìłĿ
    0.15
    _globals
    0.15
    ylül
    0.15
    /un
    0.15
    vsp
    0.15
    bable
    0.15
    .nano
    0.14
    Act Density 0.011%

    No Known Activations