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

    Neuronpedia

    INDEX
    Explanations

    links and references

    np_max-act · gemini-2.0-flash

    This neuron fires on in-text references to a book (e.g. “referred to a book by …”), detecting book‐citation phrases.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     astore
    -0.06
    competition
    -0.06
     ύ
    -0.06
     lent
    -0.06
    _FD
    -0.06
     δε
    -0.06
     Tray
    -0.06
    栏
    -0.06
    lexible
    -0.06
     withheld
    -0.06
    POSITIVE LOGITS
     attenu
    0.07
     NE
    0.07
    SY
    0.07
    "urls
    0.07
     almost
    0.07
     realidad
    0.06
     THEY
    0.06
    "These
    0.06
    рг
    0.06
     گ
    0.06
    Act Density 0.133%

    No Known Activations