vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : remove obsolete KQ_scale

Browse source
This commit is contained in:
Georgi Gerganov 2023-12-24 15:30:12 +02:00
parent 307481f28d
commit eedd434575
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
1 changed files with 0 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama.cpp
View file

@ -5643,10 +5643,6 @@ struct llm_build_context {
struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens);
cb(inp_pos, "inp_pos", -1); cb(inp_pos, "inp_pos", -1);
// KQ_scale
struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1);
cb(KQ_scale, "KQ_scale", -1);
// KQ_mask (mask for 1 head, it will be broadcasted to all heads) // KQ_mask (mask for 1 head, it will be broadcasted to all heads)
struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1);
cb(KQ_mask, "KQ_mask", -1); cb(KQ_mask, "KQ_mask", -1);