Pangu Ultra: Pushing the Limits of Dense Large Language Models on Ascend NPUs
This work addresses the optimization and system challenges of training dense LLMs for commercial applications, though it is incremental in scaling existing methods.
The authors tackled the challenge of training large-scale dense language models by developing Pangu Ultra, a 135-billion-parameter model trained on Ascend NPUs, which achieved state-of-the-art results on benchmarks, outperforming models like Llama 405B and Mistral Large 2 and competing with DeepSeek-R1.
We present Pangu Ultra, a Large Language Model (LLM) with 135 billion parameters and dense Transformer modules trained on Ascend Neural Processing Units (NPUs). Although the field of LLM has been witnessing unprecedented advances in pushing the scale and capability of LLM in recent years, training such a large-scale model still involves significant optimization and system challenges. To stabilize the training process, we propose depth-scaled sandwich normalization, which effectively eliminates loss spikes during the training process of deep models. We pre-train our model on 13.2 trillion diverse and high-quality tokens and further enhance its reasoning capabilities during post-training. To perform such large-scale training efficiently, we utilize 8,192 Ascend NPUs with a series of system optimizations. Evaluations on multiple diverse benchmarks indicate that Pangu Ultra significantly advances the state-of-the-art capabilities of dense LLMs such as Llama 405B and Mistral Large 2, and even achieves competitive results with DeepSeek-R1, whose sparse model structure contains much more parameters. Our exploration demonstrates that Ascend NPUs are capable of efficiently and effectively training dense models with more than 100 billion parameters. Our model and system will be available for our commercial customers.