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

    Neuronpedia

    INDEX
    Explanations

    us

    np_max-act · gemini-2.0-flash

    This neuron fires on subword chunks that contain tight consonant clusters—especially double letters (bb, ss) or the “ck” digraph.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    persist
    -0.07
    ，我
    -0.07
    money
    -0.07
    around
    -0.07
    Henry
    -0.06
    Rate
    -0.06
     imagination
    -0.06
    луб
    -0.06
    vědom
    -0.06
    ',{
    -0.06
    POSITIVE LOGITS
     glean
    0.08
     kız
    0.07
     břez
    0.07
     guardians
    0.07
     Kath
    0.07
     Böl
    0.07
     ErrorCode
    0.07
     vyšší
    0.07
     Mutable
    0.07
    .SOCK
    0.07
    Act Density 0.439%

    No Known Activations