From 7ac48c6b9842d1fe11afed7413841887391fc0ff Mon Sep 17 00:00:00 2001 From: Jake VanderPlas Date: Wed, 25 Sep 2024 14:36:15 -0700 Subject: [PATCH] Update README with improved listing of supported types. --- README.md | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 070e163c..2e9c5601 100644 --- a/README.md +++ b/README.md @@ -8,8 +8,9 @@ - [`bfloat16`](https://en.wikipedia.org/wiki/Bfloat16_floating-point_format): an alternative to the standard [`float16`](https://en.wikipedia.org/wiki/Half-precision_floating-point_format) format -- `float8_*`: several experimental 8-bit floating point representations - including: +- 8-bit floating point representations, parameterized by number of exponent and + mantissa bits, as well as the bias (if any) and representability of infinity, + NaN, and signed zero. * `float8_e3m4` * `float8_e4m3` * `float8_e4m3b11fnuz` @@ -17,11 +18,16 @@ * `float8_e4m3fnuz` * `float8_e5m2` * `float8_e5m2fnuz` -- Microscaling (MX) sub-byte floating point representations including: + * `float8_e8m0fnu` +- Microscaling (MX) sub-byte floating point representations: * `float4_e2m1fn` * `float6_e2m3fn` * `float6_e3m2fn` -- `int2`, `int4`, `uint2` and `uint4`: low precision integer types. +- Narrow integer encodings: + * `int2` + * `int4` + * `uint2` + * `uint4` See below for specifications of these number formats.