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

    Neuronpedia

    INDEX
    Explanations

    technical, code, and numbers

    np_max-act · gemini-2.0-flash

    This neuron detects Russian word fragments of "возведение" (and its inflected forms) in Cyrillic text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Fern
    -0.07
    .topic
    -0.06
     �
    -0.06
    �
    -0.06
     slic
    -0.06
    	W
    -0.06
     Kn
    -0.06
     reminis
    -0.06
    ches
    -0.06
    _inverse
    -0.06
    POSITIVE LOGITS
    =value
    0.08
    .intValue
    0.07
     تحص
    0.07
    ิม
    0.07
     dhcp
    0.06
     tribunal
    0.06
     ثابت
    0.06
    triangle
    0.06
    erah
    0.06
    hx
    0.06
    Act Density 0.000%

    No Known Activations