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

    Neuronpedia

    INDEX
    Explanations

    blog comments

    np_max-act · gemini-2.0-flash

    The neuron detects short name-like or handle-like tokens (author names, usernames, or blog/site handles).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    	AT
    -0.06
    .Column
    -0.06
    	GL
    -0.06
    (SE
    -0.06
     BDS
    -0.06
     terse
    -0.06
     ağaç
    -0.06
    Drupal
    -0.06
     находится
    -0.06
    esub
    -0.06
    POSITIVE LOGITS
    Var
    0.07
     Backpack
    0.07
     judging
    0.07
    Persist
    0.07
    至
    0.06
    مع
    0.06
     이러
    0.06
    δη
    0.06
    如何
    0.06
     intellectually
    0.06
    Act Density 0.030%

    No Known Activations