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

    Neuronpedia

    INDEX
    Explanations

    calculation

    np_max-act · gemini-2.0-flash

    discussions on mathematical transformations, particularly Fourier transforms.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on tokens where the author is talking about doing or checking calculations (e.g. “doing,” “calculations,” “made,” “understand”), so it’s detecting meta-comments about performing or verifying mathematical work.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     customary
    -0.07
    andır
    -0.07
    forcing
    -0.07
    業務
    -0.06
    month
    -0.06
     Punk
    -0.06
    .mkdirs
    -0.06
     spectra
    -0.06
    Bubble
    -0.06
     Bucket
    -0.06
    POSITIVE LOGITS
    lug
    0.06
    /ag
    0.06
     unleashed
    0.06
     predis
    0.06
     ID
    0.06
    NavigationBar
    0.06
     tung
    0.06
    wp
    0.06
     özellikleri
    0.06
     มห
    0.06
    Act Density 0.039%

    No Known Activations