Print Email Facebook Twitter CodeGPT on XTC Title CodeGPT on XTC: Compressing a CodeGPT Model Using Hybrid Layer Reduction and Extreme Quantisation through Knowledge Distillation Author de Moor, Aral (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Izadi, M. (mentor) Al-Kaswan, A. (mentor) van Deursen, A. (mentor) Anand, A. (graduation committee) Degree granting institution Delft University of Technology Programme Computer Science and Engineering Project CSE3000 Research Project Date 2023-06-27 Abstract Large language models are powerful because of their state-of-the-art language processing abilities. But, they come at the cost of being extremely resource-intensive, and are steadily growing in size. As a result, compressing such models for resource- constrained devices is an active and promising re- search area. In spite of their current popular- ity, many novel compression techniques lack im- plementation for GPT models. We apply the XTC pipeline, consisting of layer-reduction and quantisation through knowledge distillation, to a CodeGPT generative model. The resulting mod- els are evaluated on the CodeXGLUE line-level code-completion benchmark. Based on this, we demonstrate that (1) XTC can be adapted to GPT- like models, translating many of the findings of the original study; (2) a 6-layer reduction with 1-bit weight and 8-bit activation quantisation is able to reduce model size 15×, in addition to almost dou- bling inference speed, with minimal performance degradation. The resulting compressed models show promise for use in local code generation. By showing that a novel compression technique can be adapted to GPT-like models, we hope to further in- spire research in this field. Subject Large Language ModelsCompressionKnowledge DistillationLayer ReductionQuantisationGPT To reference this document use: http://resolver.tudelft.nl/uuid:f37924fc-ecac-4bd4-b923-7d4c73f74a72 Bibliographical note https://github.com/AISE-TUDelft/LLM4CodeCompression Replication Code Repository Part of collection Student theses Document type bachelor thesis Rights © 2023 Aral de Moor Files PDF CSE3000_XTC_Aral_Final_.pdf 269.15 KB Close viewer /islandora/object/uuid:f37924fc-ecac-4bd4-b923-7d4c73f74a72/datastream/OBJ/view