huggingface/optimum
1
0
Fork 0
mirror of https://github.com/huggingface/optimum.git synced 2026-10-12 02:15:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: fix float16 accumulation data type (#2477)

Browse Source
This commit is contained in:
Rohan170603
2026-08-18 16:42:28 +02:00
committed by GitHub
parent 787038e023
commit 88862e1c24
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/concept_guides/quantization.mdx
+1 -1
View File
@@ -15,7 +15,7 @@ The basic idea behind quantization is quite easy: going from high-precision repr
floating-point) for weights and activations to a lower precision data type. The most common lower precision data types
are:
- `float16`, accumulation data type `float16`
- `float16`, accumulation data type `float32`
- `bfloat16`, accumulation data type `float32`
- `int16`, accumulation data type `int32`
- `int8`, accumulation data type `int32`