参考文献

本文档中关于大语言模型的精度调优经验主要来自于以下文献：

• GLM-130B: AN OPEN BILINGUAL PRE-TRAINEDMODEL
• OPT: Open Pre-trained Transformer Language Models
• PaLM: Scaling Language Modeling with Pathways
• OLMo: Accelerating the Science of Language Models
• Train With Mixed Precision-NVIDIA DOCS HUB
• OLMo-7B wandb training metrics
• A Theory on Adam Instability in Large-Scale Machine Learning