mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-29 04:44:34 +00:00
speculative : clean-up and add comments and TODOs [no ci]
This commit is contained in:
parent
71fc16bb6c
commit
fe043ff1ff
@ -60,6 +60,17 @@ void common_perf_print(const struct llama_context * ctx, const struct common_sam
|
|||||||
//
|
//
|
||||||
llama_token common_sampler_sample(struct common_sampler * gsmpl, struct llama_context * ctx, int idx, bool grammar_first = false);
|
llama_token common_sampler_sample(struct common_sampler * gsmpl, struct llama_context * ctx, int idx, bool grammar_first = false);
|
||||||
|
|
||||||
|
// generalized version of common_sampler_sample
|
||||||
|
//
|
||||||
|
// will cross-reference the sampled tokens with a batch of draft tokens
|
||||||
|
// if the sampler disagrees at some point, we stop and return the sampled tokens up to now
|
||||||
|
//
|
||||||
|
// `common_sampler_sample_n(gsmpl, ctx, { idx }, {})` is equivalent to `common_sampler_sample(gsmpl, ctx, idx)`
|
||||||
|
//
|
||||||
|
// requires: idxs.size() == draft.size() + 1
|
||||||
|
//
|
||||||
|
// returns at least 1 token, up to idxs.size()
|
||||||
|
//
|
||||||
std::vector<llama_token> common_sampler_sample_n(struct common_sampler * gsmpl, struct llama_context * ctx, const std::vector<int> & idxs, const std::vector<llama_token> & draft, bool grammar_first = false);
|
std::vector<llama_token> common_sampler_sample_n(struct common_sampler * gsmpl, struct llama_context * ctx, const std::vector<int> & idxs, const std::vector<llama_token> & draft, bool grammar_first = false);
|
||||||
|
|
||||||
uint32_t common_sampler_get_seed(const struct common_sampler * gsmpl);
|
uint32_t common_sampler_get_seed(const struct common_sampler * gsmpl);
|
||||||
|
@ -4,11 +4,6 @@
|
|||||||
#include "common.h"
|
#include "common.h"
|
||||||
#include "sampling.h"
|
#include "sampling.h"
|
||||||
|
|
||||||
#include <vector>
|
|
||||||
|
|
||||||
struct seq_draft {
|
|
||||||
};
|
|
||||||
|
|
||||||
struct common_speculative {
|
struct common_speculative {
|
||||||
struct common_speculative_params params;
|
struct common_speculative_params params;
|
||||||
|
|
||||||
@ -140,7 +135,7 @@ void common_speculative_add_draft(
|
|||||||
}
|
}
|
||||||
|
|
||||||
// don't waste time on small batches
|
// don't waste time on small batches
|
||||||
// TODO: do not evaluate the draft model for tha many rounds
|
// TODO: do not evaluate the draft model for that many rounds
|
||||||
if (batch_tgt.n_tokens < spec->params.n_min) {
|
if (batch_tgt.n_tokens < spec->params.n_min) {
|
||||||
batch_tgt.n_tokens = 1;
|
batch_tgt.n_tokens = 1;
|
||||||
spec->tokens.resize(0);
|
spec->tokens.resize(0);
|
||||||
|
@ -19,8 +19,21 @@ struct common_speculative * common_speculative_init(struct common_speculative_pa
|
|||||||
|
|
||||||
void common_speculative_free(struct common_speculative * spec);
|
void common_speculative_free(struct common_speculative * spec);
|
||||||
|
|
||||||
|
// TODO: remove
|
||||||
void common_speculative_set_prompt(struct common_speculative * spec, llama_token * tokens, int32_t n_tokens);
|
void common_speculative_set_prompt(struct common_speculative * spec, llama_token * tokens, int32_t n_tokens);
|
||||||
|
|
||||||
|
// sample up to n_draft tokens and add them to the batch using the draft model
|
||||||
|
//
|
||||||
|
// TODO: change to:
|
||||||
|
//
|
||||||
|
// void common_speculative_add_draft(
|
||||||
|
// struct common_speculative * spec,
|
||||||
|
// struct llama_batch & batch_tgt,
|
||||||
|
// llama_token * tokens,
|
||||||
|
// int32_t n_tokens);
|
||||||
|
//
|
||||||
|
// and update the internal logic to compute only the new tokens
|
||||||
|
//
|
||||||
void common_speculative_add_draft(
|
void common_speculative_add_draft(
|
||||||
struct common_speculative * spec,
|
struct common_speculative * spec,
|
||||||
struct llama_batch & batch_tgt,
|
struct llama_batch & batch_tgt,
|
||||||
|
@ -120,7 +120,6 @@ int main(int argc, char ** argv) {
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
// Tokenize the prompt
|
// Tokenize the prompt
|
||||||
std::vector<llama_token> inp;
|
std::vector<llama_token> inp;
|
||||||
inp = common_tokenize(ctx_tgt, params.prompt, true, true);
|
inp = common_tokenize(ctx_tgt, params.prompt, true, true);
|
||||||
@ -139,18 +138,6 @@ int main(int argc, char ** argv) {
|
|||||||
LOG("%s", common_token_to_piece(ctx_tgt, id).c_str());
|
LOG("%s", common_token_to_piece(ctx_tgt, id).c_str());
|
||||||
}
|
}
|
||||||
|
|
||||||
const int n_input = inp.size();
|
|
||||||
|
|
||||||
const auto t_enc_start = ggml_time_us();
|
|
||||||
|
|
||||||
// eval the prompt
|
|
||||||
llama_decode(ctx_tgt, llama_batch_get_one(inp.data(), n_input - 1));
|
|
||||||
|
|
||||||
// note: keep the last token separate!
|
|
||||||
llama_token id_last = inp.back();
|
|
||||||
|
|
||||||
int n_past = inp.size() - 1;
|
|
||||||
|
|
||||||
// how many tokens to draft each time
|
// how many tokens to draft each time
|
||||||
int n_draft = params.n_draft;
|
int n_draft = params.n_draft;
|
||||||
|
|
||||||
@ -161,9 +148,25 @@ int main(int argc, char ** argv) {
|
|||||||
// used to determine end of generation
|
// used to determine end of generation
|
||||||
bool has_eos = false;
|
bool has_eos = false;
|
||||||
|
|
||||||
|
// ================================================
|
||||||
|
// everything until here is standard initialization
|
||||||
|
// the relevant stuff for speculative decoding starts here
|
||||||
|
|
||||||
|
const int n_input = inp.size();
|
||||||
|
|
||||||
|
const auto t_enc_start = ggml_time_us();
|
||||||
|
|
||||||
// target model sampling context
|
// target model sampling context
|
||||||
struct common_sampler * smpl = common_sampler_init(model_tgt, params.sparams);
|
struct common_sampler * smpl = common_sampler_init(model_tgt, params.sparams);
|
||||||
|
|
||||||
|
// eval the prompt
|
||||||
|
llama_decode(ctx_tgt, llama_batch_get_one(inp.data(), n_input - 1));
|
||||||
|
|
||||||
|
// note: keep the last token separate!
|
||||||
|
llama_token id_last = inp.back();
|
||||||
|
|
||||||
|
int n_past = inp.size() - 1;
|
||||||
|
|
||||||
// init the speculator
|
// init the speculator
|
||||||
struct common_speculative_params params_spec;
|
struct common_speculative_params params_spec;
|
||||||
params_spec.n_draft = n_draft;
|
params_spec.n_draft = n_draft;
|
||||||
@ -174,6 +177,13 @@ int main(int argc, char ** argv) {
|
|||||||
struct common_speculative * spec = common_speculative_init(params_spec);
|
struct common_speculative * spec = common_speculative_init(params_spec);
|
||||||
|
|
||||||
// feed the prompt to the speculator
|
// feed the prompt to the speculator
|
||||||
|
//
|
||||||
|
// this has to be kept synchronized with the target context
|
||||||
|
//
|
||||||
|
// TODO: simplify this by moving the context management logic in the common_speculative instance
|
||||||
|
// for example, the common_speculative_add_draft can pass the entire context (or part of it) and the
|
||||||
|
// speculator will automatically compute any new tokens that are not present in its context
|
||||||
|
//
|
||||||
common_speculative_set_prompt(spec, inp.data(), n_input - 1);
|
common_speculative_set_prompt(spec, inp.data(), n_input - 1);
|
||||||
|
|
||||||
llama_batch batch_tgt = llama_batch_init(llama_n_batch(ctx_tgt), 0, 1);
|
llama_batch batch_tgt = llama_batch_init(llama_n_batch(ctx_tgt), 0, 1);
|
||||||
@ -188,23 +198,41 @@ int main(int argc, char ** argv) {
|
|||||||
common_batch_add (batch_tgt, id_last, n_past, { 0 }, true);
|
common_batch_add (batch_tgt, id_last, n_past, { 0 }, true);
|
||||||
|
|
||||||
// optionally, append draft tokens to the target batch
|
// optionally, append draft tokens to the target batch
|
||||||
|
//
|
||||||
|
// this is the most important part of the speculation. the more probable tokens that are provided here
|
||||||
|
// the better the performance will be. in theory, this computation can be performed asynchronously and even
|
||||||
|
// offloaded to a remote device. it doesn't even have to be based on an LLM. instead, it can provide tokens
|
||||||
|
// from a cache or lookup tables.
|
||||||
|
//
|
||||||
common_speculative_add_draft(spec, batch_tgt, id_last, n_past);
|
common_speculative_add_draft(spec, batch_tgt, id_last, n_past);
|
||||||
|
|
||||||
// evaluate the target model on the drafted tokens
|
// evaluate the target model on [id_last, draft0, draft1, ..., draftN-1]
|
||||||
{
|
{
|
||||||
//LOG_DBG("target batch: %s\n", string_from(ctx_tgt, batch_tgt).c_str());
|
//LOG_DBG("target batch: %s\n", string_from(ctx_tgt, batch_tgt).c_str());
|
||||||
|
|
||||||
llama_decode(ctx_tgt, batch_tgt);
|
llama_decode(ctx_tgt, batch_tgt);
|
||||||
}
|
}
|
||||||
|
|
||||||
// process the full target batch and return the accepted token based on the target sampler
|
// sample from the full target batch and return the accepted tokens based on the target sampler
|
||||||
|
//
|
||||||
|
// for each token to be accepted, the sampler would have to sample that same token
|
||||||
|
// in such cases, instead of decoding the sampled token as we normally do, we simply continue with the
|
||||||
|
// available logits from the batch and sample the next token until we run out of logits or the sampler
|
||||||
|
// disagrees with the draft
|
||||||
|
//
|
||||||
const auto ids = common_speculative_sample(spec, smpl, ctx_tgt);
|
const auto ids = common_speculative_sample(spec, smpl, ctx_tgt);
|
||||||
|
|
||||||
|
GGML_ASSERT(ids.size() > 0); // there will always be at least one accepted token
|
||||||
|
|
||||||
n_past += ids.size();
|
n_past += ids.size();
|
||||||
n_drafted += batch_tgt.n_tokens - 1;
|
n_drafted += batch_tgt.n_tokens - 1;
|
||||||
n_accept += ids.size() - 1;
|
n_accept += ids.size() - 1;
|
||||||
|
|
||||||
// process the accepted tokens and update contexts
|
// process the accepted tokens and update contexts
|
||||||
|
//
|
||||||
|
// this is the standard token post-processing that we normally do
|
||||||
|
// in this case, we do it for a group of accepted tokens at once
|
||||||
|
//
|
||||||
{
|
{
|
||||||
llama_token id;
|
llama_token id;
|
||||||
std::string token_str;
|
std::string token_str;
|
||||||
@ -232,7 +260,7 @@ int main(int argc, char ** argv) {
|
|||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
|
|
||||||
LOG_DBG("the sampled target token (%d, '%s') did not match, or we ran out of drafted tokens\n", id, token_str.c_str());
|
LOG_DBG("accepted %d draft tokens, the last target token is: (%d, '%s')\n", (int) ids.size() - 1, id, token_str.c_str());
|
||||||
|
|
||||||
{
|
{
|
||||||
LOG_DBG("clear kv cache from any extra tokens, n_past = %d\n", n_past);
|
LOG_DBG("clear kv cache from any extra tokens, n_past = %d\n", n_past);
|
||||||
@ -241,6 +269,7 @@ int main(int argc, char ** argv) {
|
|||||||
llama_kv_cache_seq_rm(ctx_dft, 0, n_past, -1);
|
llama_kv_cache_seq_rm(ctx_dft, 0, n_past, -1);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
// remember the last accepted token for the next iteration
|
||||||
id_last = id;
|
id_last = id;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
Loading…
Reference in New Issue
Block a user