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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    delimiters and formatting markers for code or markup in chat transcripts (e.g., code fences, HTML tags, and chat meta tokens).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Expo
    -0.08
    温馨
    -0.07
     средне
    -0.07
    넗
    -0.07
     traveler
    -0.07
     snow
    -0.07
     ellipse
    -0.07
     tea
    -0.07
    וידאו
    -0.07
     crosses
    -0.07
    POSITIVE LOGITS
     �
    0.07
    ",$
    0.07
    productive
    0.07
    ɥ
    0.07
    0.07
    つもり
    0.07
     resemblance
    0.07
     האחר
    0.06
     �
    0.06
    歉
    0.06
    Act Density 0.314%

    No Known Activations