Skip to content

Use F16 for memory_k and memory_v (as suggested in #146) #154

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

Use F16 for memory_k and memory_v (as suggested in #146) #154

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions main.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -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

Expand Down Expand Up @@ -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);

Expand Down