Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-29 07:34:18 +01:00
Code Issues Packages Projects Releases Wiki Activity

llama : store non-RoPEd K cache (WIP)

Browse Source
This commit is contained in:
Georgi Gerganov 2023-09-17 23:12:28 +03:00
parent fad56936d4
commit 784d14ed31
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 31 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

42
llama.cpp
View File

@ -2428,16 +2428,25 @@ static struct ggml_cgraph * llm_build_llama(
} }
} }
// KQ_pos - contains the positions // Q_pos - contains the positions
struct ggml_tensor * KQ_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, N); struct ggml_tensor * Q_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, N);
ggml_allocr_alloc(lctx.alloc, KQ_pos); ggml_allocr_alloc(lctx.alloc, Q_pos);
if (!ggml_allocr_is_measure(lctx.alloc)) { if (!ggml_allocr_is_measure(lctx.alloc)) {
int * data = (int *) KQ_pos->data; int * data = (int *) Q_pos->data;
for (int i = 0; i < N; ++i) { for (int i = 0; i < N; ++i) {
data[i] = n_past + i; data[i] = n_past + i;
} }
} }
struct ggml_tensor * K_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_past + N);
ggml_allocr_alloc(lctx.alloc, K_pos);
if (!ggml_allocr_is_measure(lctx.alloc)) {
int * data = (int *) K_pos->data;
for (int i = 0; i < n_past + N; ++i) {
data[i] = i;
}
}
for (int il = 0; il < n_layer; ++il) { for (int il = 0; il < n_layer; ++il) {
ggml_format_name(inpL, "layer_inp_%d", il); ggml_format_name(inpL, "layer_inp_%d", il);
@ -2474,14 +2483,18 @@ static struct ggml_cgraph * llm_build_llama(
offload_func_kq(tmpq); offload_func_kq(tmpq);
ggml_set_name(tmpq, "tmpq"); ggml_set_name(tmpq, "tmpq");
struct ggml_tensor * Kcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpk, n_embd_head, n_head_kv, N), KQ_pos, n_embd_head, 0, 0, freq_base, freq_scale); // Note: we are not RoPE-ing K here
struct ggml_tensor * Kcur = tmpk;
offload_func_kq(Kcur); offload_func_kq(Kcur);
ggml_set_name(Kcur, "Kcur"); ggml_set_name(Kcur, "Kcur");
struct ggml_tensor * Qcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpq, n_embd_head, n_head, N), KQ_pos, n_embd_head, 0, 0, freq_base, freq_scale); struct ggml_tensor * Qcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpq, n_embd_head, n_head, N), Q_pos, n_embd_head, 0, 0, freq_base, freq_scale);
offload_func_kq(Qcur); offload_func_kq(Qcur);
ggml_set_name(Qcur, "Qcur"); ggml_set_name(Qcur, "Qcur");
struct ggml_tensor * ck;
struct ggml_tensor * cv;
// store key and value to memory // store key and value to memory
{ {
// compute the transposed [N, n_embd] V matrix // compute the transposed [N, n_embd] V matrix
@ -2504,9 +2517,11 @@ static struct ggml_cgraph * llm_build_llama(
offload_func_v(v); offload_func_v(v);
ggml_set_name(v, "v"); ggml_set_name(v, "v");
// important: storing RoPE-ed version of K in the KV cache! ck = ggml_cpy(ctx0, Kcur, k);
ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); cv = ggml_cpy(ctx0, Vcur, v);
ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v));
ggml_build_forward_expand(gf, ck);
ggml_build_forward_expand(gf, cv);
} }
struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3);
@ -2515,13 +2530,18 @@ static struct ggml_cgraph * llm_build_llama(
struct ggml_tensor * K = struct ggml_tensor * K =
ggml_view_3d(ctx0, kv_self.k, ggml_view_3d(ctx0, kv_self.k,
n_embd_head, n_past + N, n_head_kv, n_embd_head, n_head_kv, n_past + N,
ggml_element_size(kv_self.k)*n_embd_gqa,
ggml_element_size(kv_self.k)*n_embd_head, ggml_element_size(kv_self.k)*n_embd_head,
ggml_element_size(kv_self.k)*n_embd_gqa,
ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il);
offload_func_kq(K); offload_func_kq(K);
ggml_set_name(K, "K"); ggml_set_name(K, "K");
// RoPE the K cache
K->src[1] = ck; // TODO: HACK!!
K = ggml_rope_custom(ctx0, K, K_pos, n_embd_head, 0, 0, freq_base, freq_scale);
K = ggml_permute(ctx0, K, 0, 2, 1, 3);
// K * Q // K * Q
struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q);
offload_func_kq(KQ); offload_func_kq(KQ);