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

    Neuronpedia

    INDEX
    Explanations

    (v

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens—especially version numbers and other digit sequences in documentation URLs and identifiers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     馬
    -0.07
    _MONITOR
    -0.07
    Here
    -0.06
     Marie
    -0.06
    Near
    -0.06
    COUNT
    -0.06
    ToRemove
    -0.06
     Crate
    -0.06
     ARC
    -0.06
    arters
    -0.06
    POSITIVE LOGITS
    (norm
    0.07
    resources
    0.07
     Základní
    0.07
    .vx
    0.07
    (sk
    0.06
     wym
    0.06
    stup
    0.06
    yy
    0.06
     сок
    0.06
    τηκε
    0.06
    Act Density 0.001%

    No Known Activations