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

    Neuronpedia

    INDEX
    Explanations

    Code syntax

    np_max-act · gemini-2.0-flash

    The neuron fires primarily on standalone numerals used as step or list markers (e.g. “1.”, “2.”, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     director
    -0.07
    elyn
    -0.07
     Accountability
    -0.07
    licant
    -0.06
    ite
    -0.06
    machine
    -0.06
    twitter
    -0.06
    ْر
    -0.06
     ни
    -0.06
    !,
    -0.06
    POSITIVE LOGITS
     Dias
    0.07
     etkisi
    0.07
    Permanent
    0.07
     maximal
    0.06
    'nde
    0.06
     بشكل
    0.06
    ños
    0.06
     lẫn
    0.06
    ves
    0.06
    §Ã
    0.06
    Act Density 0.055%

    No Known Activations