zmlcc 98172d46fa Fix some errors about BlockQ8_1 (#776) ...

* use int8 type instead of uint8 for BlockQ8_1.qs

The uint8 type of BlockQ8_1.qs causes great loss for negative weights
Ref: ebc96086af/ggml.c (L904)

Signed-off-by: Zhang Miaolei <zmlcc@outlook.com>

* fix sum error in vec_dot of BlockQ4_1

Ref: ebc96086af/ggml.c (L2840)

Signed-off-by: Zhang Miaolei <zmlcc@outlook.com>

* fix sum error in vec_dot of BlockQ5_1

Ref: ebc96086af/ggml.c (L3490)

Signed-off-by: Zhang Miaolei <zmlcc@outlook.com>

---------

Signed-off-by: Zhang Miaolei <zmlcc@outlook.com>

2023-09-08 13:29:40 +01:00

..

avx.rs

Small cleanups (avoid some possible mutations) (#670)

2023-08-30 08:54:00 +01:00

ggml_file.rs

Tensor -> QTensor conversion (#496)

2023-08-18 08:19:20 +01:00

gguf_file.rs

Add ggufv2 support (#725)

2023-09-03 14:41:57 +01:00

k_quants.rs

Fix some errors about BlockQ8_1 (#776)

2023-09-08 13:29:40 +01:00

mod.rs

Support for quantized tensors in the python api. (#706)

2023-09-01 15:53:42 +01:00

neon.rs

Neon optimized vecdot (#666)

2023-08-29 22:28:46 +01:00

utils.rs

Avoid some mutable variables (take 2). (#554)

2023-08-22 18:51:20 +01:00