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

Use F16 for memory_k and memory_v

Browse source
This commit is contained in:
Ty Everett 2023-03-14 23:10:12 -07:00
parent 47857e564c
commit 1b73521840
1 changed files with 4 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

8
main.cpp
View file

@ -215,8 +215,8 @@ bool llama_model_load(const std::string & fname, llama_model & model, gpt_vocab
ctx_size += n_layer*(n_ff*n_embd*ggml_type_sizef(wtype)); // w2
ctx_size += n_layer*(n_ff*n_embd*ggml_type_sizef(wtype)); // w3
ctx_size += n_ctx*n_layer*n_embd*ggml_type_sizef(GGML_TYPE_F32); // memory_k
ctx_size += n_ctx*n_layer*n_embd*ggml_type_sizef(GGML_TYPE_F32); // memory_v
ctx_size += n_ctx*n_layer*n_embd*ggml_type_sizef(GGML_TYPE_F16); // memory_k
ctx_size += n_ctx*n_layer*n_embd*ggml_type_sizef(GGML_TYPE_F16); // memory_v
ctx_size += (5 + 10*n_layer)*256; // object overhead
@ -302,8 +302,8 @@ bool llama_model_load(const std::string & fname, llama_model & model, gpt_vocab
const int n_mem = n_layer*n_ctx;
const int n_elements = n_embd*n_mem;
model.memory_k = ggml_new_tensor_1d(ctx, GGML_TYPE_F32, n_elements);
model.memory_v = ggml_new_tensor_1d(ctx, GGML_TYPE_F32, n_elements);
model.memory_k = ggml_new_tensor_1d(ctx, GGML_TYPE_F16, n_elements);
model.memory_v = ggml_new_tensor_1d(ctx, GGML_TYPE_F16, n_elements);
const size_t memory_size = ggml_nbytes(model.memory_k) + ggml_nbytes(model.memory_v);