OpenAI解释Astra计算图深度与GPT-4的关系，强调思维链监控的重要性与挑战。
OpenAI表示Astra等前沿模型的计算图深度与GPT-4相差不超过两倍。该公司致力于保留和使用思维链监控技术，以观察模型对齐如何从训练分布中泛化。思维链监控技术被认为脆弱且呈负面趋势，但OpenAI正通过研究加强这一技术。
What exactly does it it mean for monitorability that "The depth of the computation graph for our pre...
What exactly does it it mean for monitorability that "The depth of the computation graph for our present frontier models, including Astra, is within a factor of two of GPT-4" @merettm ? (And did you mean GPT-4 or was that a typo?) For example, does that mean Astra is half as monitorable as GPT-4? Jakub Pachocki @merettm I want to prevent a race into unmonitorability kicked off by confused reporting. The depth of the computation graph for our present frontier models, including Astra, is within a factor of two of GPT-4. OpenAI has worked to preserve and utilize chain-of-thought monitoring since our very first reasoning models. We deeply care about this technique, as it can give us a view into how model alignment generalizes from its training distribution. I do think it is fragile and unfortunately trending in a negative direction, for reasons not contingent on architecture changes that I will write about soon. But there are things we can do to strengthen it, and it's a core goal of our current research program. 🔗 View Quoted Tweet 💬 3 🔄 0 ❤️ 12 👀 8675 📊 4 ⚡