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

    Neuronpedia

    INDEX
    Explanations

    with

    np_max-act · gemini-2.0-flash

    The neuron spikes on the special marker that ends a header block (the “<|end_header_id|>” token).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    aghetti
    -0.06
     paddingTop
    -0.06
    app
    -0.06
     predictions
    -0.06
    ┃
    -0.06
    :\
    -0.05
    cwd
    -0.05
     heroine
    -0.05
    @[
    -0.05
    Website
    -0.05
    POSITIVE LOGITS
     bom
    0.08
     Bee
    0.07
     Kuzey
    0.07
     Iv
    0.07
     IPAddress
    0.07
     gece
    0.07
    ικής
    0.07
     juste
    0.07
     Monaco
    0.07
     вз
    0.07
    Act Density 0.050%

    No Known Activations