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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "count" and its variations, indicating a focus on numerical references or counting concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    maal
    -0.18
    urette
    -0.16
    /latest
    -0.15
    ãĤī
    -0.15
    ated
    -0.15
    uces
    -0.15
    339
    -0.15
    ooks
    -0.15
    .Actions
    -0.14
    controls
    -0.14
    POSITIVE LOGITS
    enance
    0.36
    =count
    0.29
    /count
    0.28
    DownLatch
    0.28
     noses
    0.22
    erc
    0.21
    INUE
    0.20
    -count
    0.20
    inue
    0.19
    Vectorizer
    0.19
    Act Density 0.033%

    No Known Activations