Summary of A Comprehensive Evaluation Of Quantized Instruction-tuned Large Language Models: An Experimental Analysis Up to 405b, by Jemin Lee et al.
A Comprehensive Evaluation of Quantized Instruction-Tuned Large Language Models: An Experimental Analysis up to 405B
by Jemin Lee, Sihyeong Park, Jinse Kwon, Jihun Oh, Yongin Kwon
First submitted to arxiv on: 17 Sep 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper evaluates the performance of instruction-tuned large language models (LLMs) across various quantization methods on models ranging from 7B to 405B. The study uses 13 benchmarks and six task types, including commonsense Q&A, knowledge and language understanding, instruction following, hallucination detection, mathematics, and dialogue. Key findings reveal that quantizing a larger LLM to a similar size as a smaller FP16 LLM generally performs better across most benchmarks, except for hallucination detection and instruction following. The study also finds that performance varies significantly with different quantization methods, model size, and bit-width, with weight-only methods often yielding better results in larger models. Additionally, the study concludes that task difficulty does not significantly impact accuracy degradation due to quantization. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research paper looks at how well large language models work when they’re “quantized” (made smaller). The scientists tested different ways of making these models smaller and found that some methods work better than others. They also looked at models of different sizes and found that bigger models can sometimes perform worse after being made smaller. This study helps us understand how to make the most of these powerful language models. |
Keywords
» Artificial intelligence » Hallucination » Language understanding » Quantization
