Skip to content

speculative: give the draft context the target's recurrent-state slots #174

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
bri-prism wants to merge 1 commit into from
Closed

speculative: give the draft context the target's recurrent-state slots #174

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion common/common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -386,7 +386,7 @@ struct common_params_speculative {

uint32_t need_n_rs_seq() const {
bool needs_rs_seq = std::any_of(types.begin(), types.end(), [&](auto t) {
return t == COMMON_SPECULATIVE_TYPE_DRAFT_MTP || t == COMMON_SPECULATIVE_TYPE_DRAFT_EAGLE3 || t == COMMON_SPECULATIVE_TYPE_DRAFT_DFLASH || t == COMMON_SPECULATIVE_TYPE_DRAFT_DSPARK;
return t == COMMON_SPECULATIVE_TYPE_DRAFT_SIMPLE || t == COMMON_SPECULATIVE_TYPE_DRAFT_MTP || t == COMMON_SPECULATIVE_TYPE_DRAFT_EAGLE3 || t == COMMON_SPECULATIVE_TYPE_DRAFT_DFLASH || t == COMMON_SPECULATIVE_TYPE_DRAFT_DSPARK;
});

return needs_rs_seq ? draft.n_max : 0u;
Expand Down
7 changes: 4 additions & 3 deletions common/speculative.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2416,9 +2416,10 @@ common_speculative_init_result::common_speculative_init_result(
// the draft context holds as many tokens per sequence as the target context
cparams.n_ctx = llama_n_ctx(ctx_tgt);

// note: for small models maybe we can set this to the maximum possible draft from all speculative types
// the extra memory for small models is likely negligible?
cparams.n_rs_seq = 0;
// the draft context must reserve the same per-sequence recurrent-state slots as the target.
// leaving this at 0 makes common_context_can_seq_rm() classify the context as FULL, so
// llama_memory_seq_rm() silently fails on it and stale positions are retained.
cparams.n_rs_seq = params.speculative.need_n_rs_seq();
cparams.ctx_other = ctx_tgt;

std::string model_path;
Expand Down