Skip to content

Commit 4a58b99

Browse files
committed
llama : move llama_expert_gating_func_type to llama-hparams.h
1 parent a48c3df commit 4a58b99

File tree

2 files changed

+6
-6
lines changed

2 files changed

+6
-6
lines changed

include/llama.h

Lines changed: 0 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -116,12 +116,6 @@ extern "C" {
116116
LLAMA_ROPE_TYPE_VISION = GGML_ROPE_TYPE_VISION,
117117
};
118118

119-
enum llama_expert_gating_func_type {
120-
LLAMA_EXPERT_GATING_FUNC_TYPE_NONE = 0,
121-
LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX = 1,
122-
LLAMA_EXPERT_GATING_FUNC_TYPE_SIGMOID = 2,
123-
};
124-
125119
enum llama_token_type { //TODO: remove, required until per token attributes are available from GGUF file
126120
LLAMA_TOKEN_TYPE_UNDEFINED = 0,
127121
LLAMA_TOKEN_TYPE_NORMAL = 1,

src/llama-hparams.h

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -8,6 +8,12 @@
88
#define LLAMA_MAX_LAYERS 512
99
#define LLAMA_MAX_EXPERTS 256 // DeepSeekV3
1010

11+
enum llama_expert_gating_func_type {
12+
LLAMA_EXPERT_GATING_FUNC_TYPE_NONE = 0,
13+
LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX = 1,
14+
LLAMA_EXPERT_GATING_FUNC_TYPE_SIGMOID = 2,
15+
};
16+
1117
struct llama_hparams_posnet {
1218
uint32_t n_embd;
1319
uint32_t n_layer;

0 commit comments

Comments
 (0)