Drop measurement-rig detail from the no-baseline note

README.md

@@ -38,7 +38,7 @@ NVIDIA RTX PRO 4000 Blackwell, settings as in the Run snippet below, one warmup
 measured runs, everything resident on the GPU with no offload. Latency covers
 the full pipeline call; VRAM is peak device usage.
 
-There is no dense bf16 row because a dense bf16 *latency* row cannot be measured on this 24 GiB GPU: the dense transformer alone is 26.3 GiB, so only the quantized builds run resident. The dense reference image below still came from the same weights under offload, which changes speed but not numerics. The NVFP4 build is 1.78x faster **on this GPU**, because NVIDIA RTX PRO 4000 Blackwell has native FP4 tensor cores that only NVFP4 can use. **INT4 is the build for Turing through Ada, where NVFP4 does not run at all**, and has not been benchmarked there.
+There is no dense bf16 row because it does not fit on a 24 GiB card — the transformer alone is 26.3 GiB — while both quantized builds run resident. The NVFP4 build is 1.78x faster **on this GPU**, because NVIDIA RTX PRO 4000 Blackwell has native FP4 tensor cores that only NVFP4 can use. **INT4 is the build for Turing through Ada, where NVFP4 does not run at all**, and has not been benchmarked there.
 
 ## Output Comparison