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

    Neuronpedia

    INDEX
    Explanations

    ion

    np_max-act · gemini-2.0-flash

    This neuron activates on noun forms ending in “-ion,” i.e. words with the “ion” suffix.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     conoc
    -0.06
     belt
    -0.06
     آمده
    -0.06
    Hop
    -0.06
    experience
    -0.06
     updates
    -0.06
    	message
    -0.06
     northeastern
    -0.06
     deriving
    -0.06
     вір
    -0.06
    POSITIVE LOGITS
    .Split
    0.07
     customerId
    0.06
     Communications
    0.06
                                  ↵
    0.06
     Phot
    0.06
    '].$
    0.06
    .VarChar
    0.06
     teknik
    0.06
    )||(
    0.06
     leth
    0.06
    Act Density 0.002%

    No Known Activations