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 consistently picks out occurrences of the substring “ion” (as in the “-ion” suffix in chemical/technical terms).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    =re
    -0.07
     paul
    -0.07
    	diff
    -0.07
    (ids
    -0.06
     pink
    -0.06
     examinations
    -0.06
    ーチ
    -0.06
     hosts
    -0.06
     appreciated
    -0.06
    .EXIT
    -0.06
    POSITIVE LOGITS
    tdown
    0.07
     Answer
    0.07
    یدن
    0.07
     громадян
    0.07
    urchased
    0.06
    ธรรม
    0.06
    asive
    0.06
     Accounts
    0.06
    ूचन
    0.06
    But
    0.06
    Act Density 0.002%

    No Known Activations