huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-10 20:04:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Removed internal implementation details and clarified

Browse Source
This commit is contained in:
Merve Noyan 2023-08-23 17:00:40 +03:00 committed by GitHub
parent 98afdbbc1d
commit 2ec5436f9c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/source/conceptual/paged_attention.md
View File

@ -1,9 +1,9 @@
# Paged Attention
LLMs struggle with memory limitations during generation. In the decoding part of generation, all input tokens generated keys and values are stored in GPU memory, also referred as _KV cache_. KV cache is exhaustive for memory which causes inefficiencies in LLM serving.
LLMs struggle with memory limitations during generation. In the decoding part of generation, all input tokens generated keys and values are stored in GPU memory, also referred to as _KV cache_. KV cache is exhaustive for memory, which causes inefficiencies in LLM serving.
PagedAttention addresses the memory waste by partitioning the KV cache into blocks, allowing keys and values to be stored in non-contiguous memory. This approach improves GPU utilization and throughput.
PagedAttention also enables memory sharing, useful for parallel sampling. PagedAttention keeps track of shared memory through a block table and implements the Copy-on-Write mechanism to ensure safe sharing.
PagedAttention keeps a block table for memory sharing. This enables e.g. parallel sampling, where for a given prompt, multiple outputs are generated, and the computation and memory are shared between the outputs.
You can learn more about PagedAttention by reading the documentation [here](https://vllm.ai/).
You can learn more about PagedAttention by reading the documentation [here](https://vllm.ai/).