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

    Neuronpedia

    INDEX
    Explanations

    ellite

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word “satellite” (or its subword “ellite”) across diverse contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     room
    -0.08
     armed
    -0.08
     ROOM
    -0.07
     rooms
    -0.07
    Format
    -0.07
     Room
    -0.07
     woods
    -0.07
     cough
    -0.07
     Commands
    -0.07
    849
    -0.07
    POSITIVE LOGITS
     satellite
    0.13
     satellites
    0.11
     Satellite
    0.08
    eli
    0.07
    ΕΤ
    0.07
    uelle
    0.07
    ετ
    0.07
    _PCIE
    0.07
     dato
    0.07
    sat
    0.07
    Act Density 0.003%

    No Known Activations