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

    Neuronpedia

    INDEX
    Explanations

    jailbreak prompts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Contribution
    0.47
    UpSync
    0.42
    푎
    0.42
     Anybody
    0.41
    ᶤ
    0.41
     contribution
    0.41
    ੋਰ
    0.40
    нк
    0.40
     صفحات
    0.40
    Anybody
    0.40
    POSITIVE LOGITS
     OpenAI
    0.74
    回答
    0.72
     GPT
    0.65
    openai
    0.61
     답변
    0.61
     openai
    0.60
    応答
    0.60
     response
    0.59
    response
    0.57
     refusing
    0.57
    Act Density 0.050%

    No Known Activations