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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash

    The neuron predominantly activates on numeric metadata tokens—e.g. dates, timestamps, comment counts, and other number sequences.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Năm
    -0.07
    .yy
    -0.06
     sizi
    -0.06
    ुभव
    -0.06
    BUG
    -0.06
    ству
    -0.06
     BX
    -0.06
     Inches
    -0.06
     Buk
    -0.06
     अन
    -0.06
    POSITIVE LOGITS
    hydration
    0.07
    hydrate
    0.07
    _unc
    0.07
     paras
    0.06
     exporters
    0.06
     Microsystems
    0.06
     induce
    0.06
     затвердж
    0.06
    periment
    0.06
     representatives
    0.06
    Act Density 0.004%

    No Known Activations