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

    Neuronpedia

    INDEX
    Explanations

    numbered steps in instructions

    np_max-act · gemini-2.0-flash

    The neuron activates on the numbered step or stage labels (e.g. “Step 1:”, “Stage 2:”) in procedural outlines.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    d
    -0.07
    .v
    -0.07
    _m
    -0.06
    _be
    -0.06
    _end
    -0.06
     uninstall
    -0.06
    chimp
    -0.06
    cribed
    -0.06
     resend
    -0.06
    &amp
    -0.06
    POSITIVE LOGITS
     Bundy
    0.06
     monk
    0.06
     ergonomic
    0.06
     Ele
    0.06
     StringBuffer
    0.06
     gathering
    0.06
    伙
    0.06
    NONE
    0.06
    ("#{
    0.06
     joking
    0.06
    Act Density 0.026%

    No Known Activations