├── .gitignore ├── LICENSE ├── README.md ├── assets ├── 2403.04643.pdf └── QAQ_logo.png ├── requirements.txt └── src ├── config.py ├── evaluator.py ├── experiments ├── __init__.py ├── attention_insight.py ├── base.py ├── grid_search.py ├── key_value_difference.py ├── kvcache_distribution.py └── test.py ├── main.py ├── models.py ├── qa_dataset.py └── quantizer.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/.gitignore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/README.md -------------------------------------------------------------------------------- /assets/2403.04643.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/assets/2403.04643.pdf -------------------------------------------------------------------------------- /assets/QAQ_logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/assets/QAQ_logo.png -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/requirements.txt -------------------------------------------------------------------------------- /src/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/config.py -------------------------------------------------------------------------------- /src/evaluator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/evaluator.py -------------------------------------------------------------------------------- /src/experiments/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/__init__.py -------------------------------------------------------------------------------- /src/experiments/attention_insight.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/attention_insight.py -------------------------------------------------------------------------------- /src/experiments/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/base.py -------------------------------------------------------------------------------- /src/experiments/grid_search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/grid_search.py -------------------------------------------------------------------------------- /src/experiments/key_value_difference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/key_value_difference.py -------------------------------------------------------------------------------- /src/experiments/kvcache_distribution.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/kvcache_distribution.py -------------------------------------------------------------------------------- /src/experiments/test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/experiments/test.py -------------------------------------------------------------------------------- /src/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/main.py -------------------------------------------------------------------------------- /src/models.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/models.py -------------------------------------------------------------------------------- /src/qa_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/qa_dataset.py -------------------------------------------------------------------------------- /src/quantizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ClubieDong/QAQ-KVCacheQuantization/HEAD/src/quantizer.py --------------------------------------------------------------------------------