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

    Neuronpedia

    INDEX
    Explanations

    Cross

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word “Cross” (often as the “cross-” prefix) at the start of compound terms.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Alman
    -0.07
     alım
    -0.07
     Leonard
    -0.07
    (len
    -0.07
     Ya
    -0.07
     Hein
    -0.07
     Samuel
    -0.07
    unifu
    -0.07
     Emit
    -0.06
     emit
    -0.06
    POSITIVE LOGITS
     Cross
    0.17
     cross
    0.14
    Cross
    0.12
     CROSS
    0.11
    cross
    0.11
     crossed
    0.10
    .Cross
    0.09
    _cross
    0.09
     crossings
    0.09
     CrossRef
    0.09
    Act Density 0.016%

    No Known Activations