tristan/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

server: fix reported top tokens for temperature 0 (#7203)

Browse source
This commit is contained in:
Johannes Gäßler 2024-05-11 10:11:28 +02:00 • committed by GitHub
parent b83cc3f5b3
commit 5ae3426b0b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 7 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
common/sampling.cpp
View file

@ -35,7 +35,7 @@ struct llama_sampling_context * llama_sampling_init(const struct llama_sampling_
result->prev.resize(params.n_prev); result->prev.resize(params.n_prev);
result->n_considered = 0; result->n_valid = 0;
llama_sampling_set_rng_seed(result, params.seed); llama_sampling_set_rng_seed(result, params.seed);
@ -66,7 +66,7 @@ void llama_sampling_reset(llama_sampling_context * ctx) {
std::fill(ctx->prev.begin(), ctx->prev.end(), 0); std::fill(ctx->prev.begin(), ctx->prev.end(), 0);
ctx->cur.clear(); ctx->cur.clear();
ctx->n_considered = 0; ctx->n_valid = 0;
} }
void llama_sampling_set_rng_seed(struct llama_sampling_context * ctx, uint32_t seed) { void llama_sampling_set_rng_seed(struct llama_sampling_context * ctx, uint32_t seed) {
@ -256,7 +256,7 @@ static llama_token llama_sampling_sample_impl(
} }
} }
ctx_sampling->n_considered = cur_p.size; ctx_sampling->n_valid = temp == 0.0f ? 0 : cur_p.size;
return id; return id;
} }

2
common/sampling.h
View file

@ -81,7 +81,7 @@ struct llama_sampling_context {
// TODO: replace with ring-buffer // TODO: replace with ring-buffer
std::vector<llama_token> prev; std::vector<llama_token> prev;
std::vector<llama_token_data> cur; std::vector<llama_token_data> cur;
size_t n_considered; size_t n_valid; // Number of correct top tokens with correct probabilities.
std::mt19937 rng; std::mt19937 rng;
}; };

6
examples/server/server.cpp
View file

@ -2270,10 +2270,10 @@ struct server_context {
const size_t n_probs = std::min(cur_p.size, (size_t) slot.sparams.n_probs); const size_t n_probs = std::min(cur_p.size, (size_t) slot.sparams.n_probs);
if (n_probs > 0) { if (n_probs > 0) {
const size_t n_considered = slot.ctx_sampling->n_considered; const size_t n_valid = slot.ctx_sampling->n_valid;
// Make sure at least n_probs top tokens are at the front of the vector: // Make sure at least n_probs top tokens are at the front of the vector:
if (slot.sparams.temp == 0.0f && n_probs > n_considered) { if (slot.sparams.temp == 0.0f && n_probs > n_valid) {
llama_sample_top_k(ctx, &cur_p, n_probs, 0); llama_sample_top_k(ctx, &cur_p, n_probs, 0);
} }
@ -2289,7 +2289,7 @@ struct server_context {
for (size_t i = 0; i < n_probs; ++i) { for (size_t i = 0; i < n_probs; ++i) {
result.probs.push_back({ result.probs.push_back({
cur_p.data[i].id, cur_p.data[i].id,
i >= n_considered ? 0.0f : cur_p.data[i].p // Tokens filtered out due to e.g. top_k have 0 probability. i >= n_valid ? 0.0f : cur_p.data[i].p // Tokens filtered out due to e.g. top_k have 0 probability.
}); });
} }
} }