vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update ggml-phi-knc-dot_q5_K_q8_K.c

Browse source 
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
This commit is contained in:
Julia Longtin 2024-04-04 21:24:08 +01:00 • committed by GitHub
parent 8422686d0a
commit 2e86228a9b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 0 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

1
ggml-phi-knc-dot_q5_K_q8_K.c
View file

@ -157,7 +157,6 @@ inline static void GGML_5bit_Unpack (const uint8x16_t * q4, const uint8_t * q1,
: "zmm0", "zmm1", "zmm2", "zmm4", "zmm5", "zmm6", "zmm7", "zmm8", "zmm9", "zmm10", "zmm11", "cc", "ecx", "k1", "k2", "r12", "r8", "memory"
);
}
// A function for getting the dot product of two vectors, one of 5 bit resolution, and one of 8.
// Used during inference, if your model prints "llama_model_loader: - type q5_K: XXX tensors", and XXX is not zero. :)
void ggml_vec_dot_q5_K_q8_K(int n, float * restrict s, size_t bs, const void * restrict vx, size_t bx, const void * restrict vy, size_t by, int nrc) {