Longskywork: A Training Recipe For Efficiently Extending Context Length In Large Language Models

Zhao Liang, Wei Tianwen, Zeng Liang, Cheng Cheng, Yang Liu, Cheng Peng, Wang Lijie, Li Chenxia, Wu Xuejie, Zhu Bo, Gan Yimeng, Hu Rui, Yan Shuicheng, Fang Han, Zhou Yahui. Arxiv 2024

[Paper]    
Efficiency And Optimization Fine Tuning Language Modeling Pretraining Methods Training Techniques

We introduce LongSkywork, a long-context Large Language Model (LLM) capable of processing up to 200,000 tokens. We provide a training recipe for efficiently extending context length of LLMs. We identify that the critical element in enhancing long-context processing capability is to incorporate a long-context SFT stage following the standard SFT stage. A mere 200 iterations can convert the standard SFT model into a long-context model. To reduce the effort in collecting and annotating data for long-context language modeling, we develop two novel methods for creating synthetic data. These methods are applied during the continual pretraining phase as well as the Supervised Fine-Tuning (SFT) phase, greatly enhancing the training efficiency of our long-context LLMs. Our findings suggest that synthetic long-context SFT data can surpass the performance of data curated by humans to some extent. LongSkywork achieves outstanding performance on a variety of long-context benchmarks. In the Needle test, a benchmark for long-context information retrieval, our models achieved perfect accuracy across multiple context spans. Moreover, in realistic application scenarios, LongSkywork-13B demonstrates performance on par with Claude2.1, the leading long-context model, underscoring the effectiveness of our proposed methods.

Similar Work