root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-30 13:24:35 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

llama : adapt new models to F16 KQ_mask

Browse Source
This commit is contained in:
Georgi Gerganov 2024-03-03 13:50:54 +02:00
parent 02a645e7b7
commit 6aefd11204
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
llama.cpp
View File

@ -7362,7 +7362,7 @@ struct llm_build_context {
cb(inp_pos, "inp_pos", -1);
// KQ_mask (mask for 1 head, it will be broadcasted to all heads)
struct ggml_tensor * KQ_mask = ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, n_tokens, n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0);
struct ggml_tensor * KQ_mask = ggml_cast(ctx0, ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, GGML_PAD(n_tokens, GGML_KQ_MASK_PAD), n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0), GGML_TYPE_F16);
cb(KQ_mask, "KQ_mask", -1);
for (int il = 0; il < n_layer; ++il) {
@ -7489,7 +7489,7 @@ struct llm_build_context {
cb(inp_pos, "inp_pos", -1);
// KQ_mask (mask for 1 head, it will be broadcasted to all heads)
struct ggml_tensor * KQ_mask = ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, n_tokens, n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0);
struct ggml_tensor * KQ_mask = ggml_cast(ctx0, ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, GGML_PAD(n_tokens, GGML_KQ_MASK_PAD), n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0), GGML_TYPE_F16);
cb(KQ_mask, "KQ_mask", -1);
for (int il = 0; il < n_layer; ++il) {
@ -7724,7 +7724,7 @@ struct llm_build_context {
cb(inp_pos, "inp_pos", -1);
// KQ_mask (mask for 1 head, it will be broadcasted to all heads)
struct ggml_tensor * KQ_mask = ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, n_tokens, n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0);
struct ggml_tensor * KQ_mask = ggml_cast(ctx0, ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, GGML_PAD(n_tokens, GGML_KQ_MASK_PAD), n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0), GGML_TYPE_F16);
cb(KQ_mask, "KQ_mask", -1);
for (int il = 0; il < n_layer; ++il) {