WIP split mode attn

Works for LlaMA models, but not for GLM-4.5.
Doesn't seem to improve performance, so I guess no point in trying to
fix it.
This commit is contained in:
Kawrakow
2025-12-01 09:34:14 +00:00
parent a8cb1860b3
commit 63d0389e18
6 changed files with 88 additions and 58 deletions

View File

@@ -1276,6 +1276,9 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
else if (arg_next == "layer") {
params.split_mode = LLAMA_SPLIT_MODE_LAYER;
}
else if (arg_next == "attn") {
params.split_mode = LLAMA_SPLIT_MODE_ATTN;
}
else if (arg_next == "graph") {
params.split_mode = LLAMA_SPLIT_MODE_GRAPH;
}