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

    Neuronpedia

    INDEX
    Explanations

    underwater-related terms and phrases

    oai_token-act-pair · gpt-3.5-turbo

    references to underwater environments and conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ========
    -1.04
    ============
    -0.90
    aign
    -0.88
    ====
    -0.86
    ======
    -0.81
    present
    -0.81
    MQ
    -0.79
    Brand
    -0.79
    ================
    -0.79
    xx
    -0.78
    POSITIVE LOGITS
     underwater
    1.24
     diving
    1.01
     submerged
    0.93
     crocod
    0.90
     landsl
    0.88
     aquarium
    0.86
    melon
    0.83
    eatures
    0.83
     eleph
    0.82
     cliffs
    0.82
    Act Density 0.011%

    No Known Activations