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

    Neuronpedia

    INDEX
    Explanations

    Code snippets

    np_max-act · gemini-2.0-flash

    programming conditional/branching constructs (if/then/else style tokens and related directive markers).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    .imwrite
    -0.06
    /material
    -0.06
    .just
    -0.06
     reput
    -0.06
     roofs
    -0.06
    atasets
    -0.06
     disse
    -0.06
    _STACK
    -0.06
    groupBy
    -0.06
     steadfast
    -0.06
    POSITIVE LOGITS
    (sub
    0.06
    ancode
    0.06
     उद
    0.06
     blinked
    0.06
    [V
    0.06
    [:,
    0.06
     choke
    0.06
    　ゝ
    0.06
    าษฎ
    0.06
     sub
    0.06
    Act Density 0.024%

    No Known Activations