sdgoij/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-05-11 11:34:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into gg/flash-attn

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2024-04-24 14:00:32 +03:00
parent 751591d520 3fe847b574
commit 8937ec5307
57 changed files with 1576 additions and 8245 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
examples/server/server.cpp
View File

@@ -854,7 +854,7 @@ struct server_context {
slot.sparams.penalize_nl = json_value(data, "penalize_nl", default_sparams.penalize_nl);
slot.params.n_keep = json_value(data, "n_keep", slot.params.n_keep);
slot.params.n_discard = json_value(data, "n_discard", default_params.n_discard);
slot.params.seed = json_value(data, "seed", default_params.seed);
slot.sparams.seed = json_value(data, "seed", default_sparams.seed);
slot.sparams.n_probs = json_value(data, "n_probs", default_sparams.n_probs);
slot.sparams.min_keep = json_value(data, "min_keep", default_sparams.min_keep);
@@ -1028,7 +1028,6 @@ struct server_context {
send_error(task, "Failed to parse grammar", ERROR_TYPE_INVALID_REQUEST);
return false;
}
llama_set_rng_seed(ctx, slot.params.seed);
}
slot.command = SLOT_COMMAND_LOAD_PROMPT;
@@ -1118,7 +1117,7 @@ struct server_context {
bool process_token(completion_token_output & result, server_slot & slot) {
// remember which tokens were sampled - used for repetition penalties during sampling
const std::string token_str = llama_token_to_piece(ctx, result.tok);
const std::string token_str = llama_token_to_piece(ctx, result.tok, false);
slot.sampled = result.tok;
// search stop word and delete it
@@ -1201,7 +1200,7 @@ struct server_context {
});
}
if (result.tok == llama_token_eos(model)) {
if (llama_token_is_eog(model, result.tok)) {
slot.stopped_eos = true;
slot.has_next_token = false;