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

    Neuronpedia

    INDEX
    Explanations

    :

    np_max-act · gemini-2.0-flash

    This neuron detects the hyphen markers that introduce list items (leading “-” tokens) in structured text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    agit
    -0.06
     дисцип
    -0.06
     reflection
    -0.06
    _ISO
    -0.06
     transitional
    -0.06
    preview
    -0.06
    Ack
    -0.06
     multiprocessing
    -0.06
     lớn
    -0.06
     подоб
    -0.06
    POSITIVE LOGITS
     sells
    0.08
     hvordan
    0.07
    .getEntity
    0.07
     onCancel
    0.07
     cams
    0.06
     OTP
    0.06
    WHO
    0.06
     hose
    0.06
     anon
    0.06
    /meta
    0.06
    Act Density 0.002%

    No Known Activations