happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

arch: refactor LLM_TENSOR_NAMES

This commit is contained in:
Xuan Son Nguyen 2025-12-15 10:02:17 +01:00
parent 4aced7a631
commit d2bed05bee
2 changed files with 1861 additions and 2250 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4101
src/llama-arch.cpp
View File

File diff suppressed because it is too large Load Diff

10
src/llama-arch.h
View File

@ -3,6 +3,7 @@
#include "ggml.h" // ggml_op
#include <string>
#include <set>
//
// gguf constants (sync with gguf.py)
@ -315,6 +316,7 @@ enum llm_tensor {
LLM_TENSOR_DENSE_3_OUT,
LLM_TENSOR_OUTPUT,
LLM_TENSOR_OUTPUT_NORM,
LLM_TENSOR_OUTPUT_NORM_LFM2, // fix for wrong tensor name
LLM_TENSOR_ROPE_FREQS,
LLM_TENSOR_ROPE_FACTORS_LONG,
LLM_TENSOR_ROPE_FACTORS_SHORT,
@ -525,6 +527,10 @@ struct LLM_TN_IMPL {
const int bid;
const int xid;
const std::set<llm_tensor> model_tensors;
LLM_TN_IMPL(llm_arch arch, llm_tensor tensor, const char * suffix, int bid, int xid);
std::string str() const;
operator std::string() const {
@ -546,11 +552,11 @@ struct LLM_TN {
llm_arch arch;
LLM_TN_IMPL operator()(llm_tensor tensor, const char * suffix, int bid = -1, int xid = -1) const {
return { arch, tensor, suffix, bid, xid };
return LLM_TN_IMPL(arch, tensor, suffix, bid, xid);
}
LLM_TN_IMPL operator()(llm_tensor tensor, int bid = -1, int xid = -1) const {
return { arch, tensor, nullptr, bid, xid };
return LLM_TN_IMPL(arch, tensor, nullptr, bid, xid);
}
};