From 7a6f681daf919aefdfc34b5ce36272a7a1de5b6c Mon Sep 17 00:00:00 2001 From: Saood Karim Date: Tue, 25 Mar 2025 09:15:19 -0500 Subject: [PATCH] Fix README.md --- examples/sweep-bench/README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/examples/sweep-bench/README.md b/examples/sweep-bench/README.md index 608fd104..d92740de 100644 --- a/examples/sweep-bench/README.md +++ b/examples/sweep-bench/README.md @@ -7,6 +7,7 @@ in each ubatch-sized window. Only a single token sequence is used. The benchmark steps are: for each ubatch-sized window in context: + 1. generate ubatch/4 tokens (not the whole window to save some time) 2. measure generation performance 3. remove generated tokens from KV cache