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

Convert vector to f16 for dequantize mul mat vec (#1913)

Browse source 
* Convert vector to f16 for dmmv

* compile option

* Added compilation option description to README

* Changed cmake CUDA_ARCHITECTURES from "OFF" to "native"
This commit is contained in:
Johannes Gäßler 2023-06-19 10:23:56 +02:00 • committed by GitHub
parent b24c3049d9
commit 16b9cd1939
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 158 additions and 68 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama.cpp
View file

@ -1620,7 +1620,7 @@ static bool llama_eval_internal(
model.layers[il].w1,
cur);
offload_func(cur);
ggml_set_name(cur, "result_w2");
ggml_set_name(cur, "result_w1");
// SILU activation
cur = ggml_silu(ctx0, cur);