RFC-0030: Incorporate comments

Author: australopitek

RFC-0030-native-fp8-dtype.md

@@ -10,7 +10,7 @@ More and more companies working on Deep Learning accelerators are experimenting
 Since fp8 data type seems to be a natural evolution of currently used fp16/bf16, to reduce computation of big DL models, it’s worth to standardize this type. Few attempts of this were done recently:
 
 * Nvidia, Arm and Intel - https://arxiv.org/pdf/2209.05433.pdf
-* GraphCore and AMD - https://arxiv.org/pdf/2206.02915.pdf
+* GraphCore, AMD and Qualcomm - https://arxiv.org/pdf/2206.02915.pdf
 * Tesla - https://tesla-cdn.thron.com/static/MXMU3S_tesla-dojo-technology_1WDVZN.pdf
 
 This RFC proposes adding two 8-bit floating point data types variants to PyTorch, based on the Nvidia/Arm/Intel paper. It’s important to consider these two variants, because they’re already known to be used by Nvidia H100 and Intel Gaudi2 accelerators.