vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-14 02:49:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	docs/backend/SYCL.md
#	docs/build.md
#	ggml/CMakeLists.txt
#	ggml/src/ggml-cpu/CMakeLists.txt
#	ggml/src/ggml-cpu/amx/mmq.cpp
#	ggml/src/ggml-cpu/ggml-cpu.c
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-sycl/common.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/sycl_hw.cpp
#	ggml/src/ggml-sycl/sycl_hw.hpp
#	ggml/src/ggml-vulkan/CMakeLists.txt
#	tests/test-backend-ops.cpp
This commit is contained in:
Concedo 2025-06-28 08:56:29 +08:00
commit 0ac20e30b5
45 changed files with 2186 additions and 907 deletions
Download patch file Download diff file Expand all files Collapse all files

9
src/llama-quant.cpp
View file

@ -223,7 +223,7 @@ static ggml_type llama_tensor_get_type(quantize_state_impl & qs, ggml_type new_t
new_type = GGML_TYPE_Q6_K;
}
}
} else if (name == "token_embd.weight") {
} else if (name == "token_embd.weight" || name == "per_layer_token_embd.weight") {
if (qs.params->token_embedding_type < GGML_TYPE_COUNT) {
new_type = qs.params->token_embedding_type;
} else {
@ -833,6 +833,13 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
// NOTE: can't use LLM_TN here because the layer number is not known
quantize &= name.find("ffn_gate_inp.weight") == std::string::npos;
// these are very small (e.g. 4x4)
quantize &= name.find("altup") == std::string::npos;
quantize &= name.find("laurel") == std::string::npos;
// these are not too big so keep them as it is
quantize &= name.find("per_layer_model_proj") == std::string::npos;
// do not quantize positional embeddings and token types (BERT)
quantize &= name != LLM_TN(model.arch)(LLM_TENSOR_POS_EMBD, "weight");
quantize &= name != LLM_TN(model.arch)(LLM_TENSOR_TOKEN_TYPES, "weight");