EchoAtt: Attend, Copy, then Adjust for More Efficient Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915575134945280 |
|---|---|
| author | Rajabzadeh, Hossein Jafari, Aref Sharma, Aman Jami, Benyamin Kwon, Hyock Ju Ghodsi, Ali Chen, Boxing Rezagholizadeh, Mehdi |
| author_facet | Rajabzadeh, Hossein Jafari, Aref Sharma, Aman Jami, Benyamin Kwon, Hyock Ju Ghodsi, Ali Chen, Boxing Rezagholizadeh, Mehdi |
| contents | Large Language Models (LLMs), with their increasing depth and number of parameters, have demonstrated outstanding performance across a variety of natural language processing tasks. However, this growth in scale leads to increased computational demands, particularly during inference and fine-tuning. To address these challenges, we introduce EchoAtt, a novel framework aimed at optimizing transformer-based models by analyzing and leveraging the similarity of attention patterns across layers. Our analysis reveals that many inner layers in LLMs, especially larger ones, exhibit highly similar attention matrices. By exploiting this similarity, EchoAtt enables the sharing of attention matrices in less critical layers, significantly reducing computational requirements without compromising performance. We incorporate this approach within a knowledge distillation setup, where a pre-trained teacher model guides the training of a smaller student model. The student model selectively shares attention matrices in layers with high similarity while inheriting key parameters from the teacher. Our best results with TinyLLaMA-1.1B demonstrate that EchoAtt improves inference speed by 15\%, training speed by 25\%, and reduces the number of parameters by approximately 4\%, all while improving zero-shot performance. These findings highlight the potential of attention matrix sharing to enhance the efficiency of LLMs, making them more practical for real-time and resource-limited applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_14595 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | EchoAtt: Attend, Copy, then Adjust for More Efficient Large Language Models Rajabzadeh, Hossein Jafari, Aref Sharma, Aman Jami, Benyamin Kwon, Hyock Ju Ghodsi, Ali Chen, Boxing Rezagholizadeh, Mehdi Computation and Language Machine Learning Large Language Models (LLMs), with their increasing depth and number of parameters, have demonstrated outstanding performance across a variety of natural language processing tasks. However, this growth in scale leads to increased computational demands, particularly during inference and fine-tuning. To address these challenges, we introduce EchoAtt, a novel framework aimed at optimizing transformer-based models by analyzing and leveraging the similarity of attention patterns across layers. Our analysis reveals that many inner layers in LLMs, especially larger ones, exhibit highly similar attention matrices. By exploiting this similarity, EchoAtt enables the sharing of attention matrices in less critical layers, significantly reducing computational requirements without compromising performance. We incorporate this approach within a knowledge distillation setup, where a pre-trained teacher model guides the training of a smaller student model. The student model selectively shares attention matrices in layers with high similarity while inheriting key parameters from the teacher. Our best results with TinyLLaMA-1.1B demonstrate that EchoAtt improves inference speed by 15\%, training speed by 25\%, and reduces the number of parameters by approximately 4\%, all while improving zero-shot performance. These findings highlight the potential of attention matrix sharing to enhance the efficiency of LLMs, making them more practical for real-time and resource-limited applications. |
| title | EchoAtt: Attend, Copy, then Adjust for More Efficient Large Language Models |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2409.14595 |