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

    Neuronpedia

    INDEX
    Explanations

    and Canada

    np_acts-logits-general · gemini-2.5-flash-lite

    references to suicide/crisis/self‑harm hotlines and emergency mental‑health support contact information.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    立方
    0.39
    Valent
    0.39
    bandit
    0.38
    ヴィトン
    0.38
    パッド
    0.38
    丁
    0.37
     מק
    0.37
    瑯
    0.36
     longitudinally
    0.35
     гидро
    0.35
    POSITIVE LOGITS
    Su
    0.37
    dech
    0.35
    кет
    0.35
     Su
    0.35
    kan
    0.33
     আশে
    0.33
     su
    0.33
    Kan
    0.33
    nica
    0.32
     IEEE
    0.32
    Act Density 0.011%

    No Known Activations