root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-26 11:24:35 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

ggml: cache sin/cos for RoPE (#4908)

Browse Source
This commit is contained in:
Johannes Gäßler 2024-01-13 21:41:37 +01:00 committed by GitHub
parent 4be5ef556d
commit c71d608ce7
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 32 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
ggml.c
View File

@ -11638,6 +11638,21 @@ static float ggml_rope_yarn_corr_dim(int n_dims, int n_orig_ctx, float n_rot, fl
return n_dims * logf(n_orig_ctx / (n_rot * 2 * (float)M_PI)) / (2 * logf(base)); return n_dims * logf(n_orig_ctx / (n_rot * 2 * (float)M_PI)) / (2 * logf(base));
} }
static void ggml_rope_cache_init(
float theta_base, float freq_scale, float corr_dims[2], int64_t ne0, float ext_factor, float mscale,
float * cache, float sin_sign, float theta_scale
) {
float theta = theta_base;
for (int64_t i0 = 0; i0 < ne0; i0 += 2) {
rope_yarn(
theta, freq_scale, corr_dims, i0, ext_factor, mscale, &cache[i0 + 0], &cache[i0 + 1]
);
cache[i0 + 1] *= sin_sign;
theta *= theta_scale;
}
}
void ggml_rope_yarn_corr_dims( void ggml_rope_yarn_corr_dims(
int n_dims, int n_orig_ctx, float freq_base, float beta_fast, float beta_slow, float dims[2] int n_dims, int n_orig_ctx, float freq_base, float beta_fast, float beta_slow, float dims[2]
) { ) {
@ -11720,6 +11735,12 @@ static void ggml_compute_forward_rope_f32(
for (int64_t i3 = 0; i3 < ne3; i3++) { for (int64_t i3 = 0; i3 < ne3; i3++) {
for (int64_t i2 = 0; i2 < ne2; i2++) { for (int64_t i2 = 0; i2 < ne2; i2++) {
const int64_t p = pos[i2]; const int64_t p = pos[i2];
float * cache = (float *) params->wdata + (ne0 + CACHE_LINE_SIZE_F32)*ith;
if (!is_glm && !is_neox) { // TODO: cache sin/cos for glm, neox
ggml_rope_cache_init(p, freq_scale, corr_dims, ne0, ext_factor, attn_factor, cache, sin_sign, theta_scale);
}
for (int64_t i1 = 0; i1 < ne1; i1++) { for (int64_t i1 = 0; i1 < ne1; i1++) {
if (ir++ < ir0) continue; if (ir++ < ir0) continue;
if (ir > ir1) break; if (ir > ir1) break;
@ -11753,18 +11774,13 @@ static void ggml_compute_forward_rope_f32(
} }
} else if (!is_neox) { } else if (!is_neox) {
for (int64_t i0 = 0; i0 < ne0; i0 += 2) { for (int64_t i0 = 0; i0 < ne0; i0 += 2) {
float cos_theta, sin_theta; const float cos_theta = cache[i0 + 0];
rope_yarn( const float sin_theta = cache[i0 + 1];
theta_base, freq_scale, corr_dims, i0, ext_factor, attn_factor, &cos_theta, &sin_theta
);
sin_theta *= sin_sign;
// zeta scaling for xPos only: // zeta scaling for xPos only:
float zeta = xpos_base != 0.0f ? powf((i0 + 0.4f * ne0) / (1.4f * ne0), p / xpos_base) : 1.0f; float zeta = xpos_base != 0.0f ? powf((i0 + 0.4f * ne0) / (1.4f * ne0), p / xpos_base) : 1.0f;
if (xpos_down) zeta = 1.0f / zeta; if (xpos_down) zeta = 1.0f / zeta;
theta_base *= theta_scale;
const float * const src = (float *)((char *) src0->data + i3*nb03 + i2*nb02 + i1*nb01 + i0*nb00); const float * const src = (float *)((char *) src0->data + i3*nb03 + i2*nb02 + i1*nb01 + i0*nb00);
float * dst_data = (float *)((char *) dst->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0); float * dst_data = (float *)((char *) dst->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0);
@ -11888,6 +11904,12 @@ static void ggml_compute_forward_rope_f16(
for (int64_t i3 = 0; i3 < ne3; i3++) { for (int64_t i3 = 0; i3 < ne3; i3++) {
for (int64_t i2 = 0; i2 < ne2; i2++) { for (int64_t i2 = 0; i2 < ne2; i2++) {
const int64_t p = pos[i2]; const int64_t p = pos[i2];
float * cache = (float *) params->wdata + (ne0 + CACHE_LINE_SIZE_F32)*ith;
if (!is_glm && !is_neox) { // TODO: cache sin/cos for glm, neox
ggml_rope_cache_init(p, freq_scale, corr_dims, ne0, ext_factor, attn_factor, cache, sin_sign, theta_scale);
}
for (int64_t i1 = 0; i1 < ne1; i1++) { for (int64_t i1 = 0; i1 < ne1; i1++) {
if (ir++ < ir0) continue; if (ir++ < ir0) continue;
if (ir > ir1) break; if (ir > ir1) break;
@ -11921,13 +11943,8 @@ static void ggml_compute_forward_rope_f16(
} }
} else if (!is_neox) { } else if (!is_neox) {
for (int64_t i0 = 0; i0 < ne0; i0 += 2) { for (int64_t i0 = 0; i0 < ne0; i0 += 2) {
float cos_theta, sin_theta; const float cos_theta = cache[i0 + 0];
rope_yarn( const float sin_theta = cache[i0 + 1];
theta_base, freq_scale, corr_dims, i0, ext_factor, attn_factor, &cos_theta, &sin_theta
);
sin_theta *= sin_sign;
theta_base *= theta_scale;
const ggml_fp16_t * const src = (ggml_fp16_t *)((char *) src0->data + i3*nb03 + i2*nb02 + i1*nb01 + i0*nb00); const ggml_fp16_t * const src = (ggml_fp16_t *)((char *) src0->data + i3*nb03 + i2*nb02 + i1*nb01 + i0*nb00);
ggml_fp16_t * dst_data = (ggml_fp16_t *)((char *) dst->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0); ggml_fp16_t * dst_data = (ggml_fp16_t *)((char *) dst->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0);
@ -16722,6 +16739,7 @@ struct ggml_cplan ggml_graph_plan(const struct ggml_cgraph * cgraph, int n_threa
} }
} break; } break;
case GGML_OP_SOFT_MAX: case GGML_OP_SOFT_MAX:
case GGML_OP_ROPE:
{ {
cur = ggml_type_size(GGML_TYPE_F32) * node->ne[0] * n_tasks; cur = ggml_type_size(GGML_TYPE_F32) * node->ne[0] * n_tasks;
} break; } break;