Skip to content

Distribute and run LLMs with a single file.

Introduce --special flag

Justine Tunney72fb8ca7cd30
diff --git a/llama.cpp/common.cpp b/llama.cpp/common.cpp
index 7a04f9cf..5022a565 100644
--- a/llama.cpp/common.cpp
+++ b/llama.cpp/common.cpp
@@ -843,8 +843,8 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
params.interactive = true;
return true;
}
- if (arg == "--interactive-specials") {
- params.interactive_specials = true;
+ if (arg == "-sp" || arg == "--special") {
+ params.special = true;
return true;
}
if (arg == "--embedding") {
@@ -1363,8 +1363,8 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
printf(" -h, --help show this help message and exit\n");
printf(" --version show version and build info\n");
printf(" -i, --interactive run in interactive mode\n");
- printf(" --interactive-specials allow special tokens in user text, in interactive mode\n");
printf(" --interactive-first run in interactive mode and wait for input right away\n");
+ printf(" -sp, --special special tokens output enabled\n");
printf(" -cnv, --conversation run in conversation mode (does not print special tokens and suffix/prefix)\n");
printf(" -ins, --instruct run in instruction mode (use with Alpaca models)\n");
printf(" -cml, --chatml run in chatml mode (use with ChatML-compatible models)\n");
@@ -2620,7 +2620,7 @@ void dump_non_result_info_yaml(FILE * stream, const gpt_params & params, const l
dump_string_yaml_multiline(stream, "in_suffix", params.input_prefix.c_str());
fprintf(stream, "instruct: %s # default: false\n", params.instruct ? "true" : "false");
fprintf(stream, "interactive: %s # default: false\n", params.interactive ? "true" : "false");
- fprintf(stream, "interactive_specials: %s # default: false\n", params.interactive_specials ? "true" : "false");
+ fprintf(stream, "specials: %s # default: false\n", params.special ? "true" : "false");
fprintf(stream, "interactive_first: %s # default: false\n", params.interactive_first ? "true" : "false");
fprintf(stream, "keep: %d # default: 0\n", params.n_keep);
fprintf(stream, "logdir: %s # default: unset (no logging)\n", params.logdir.c_str());
diff --git a/llama.cpp/common.h b/llama.cpp/common.h
index 1fdbc7d3..465e1b8c 100644
--- a/llama.cpp/common.h
+++ b/llama.cpp/common.h
@@ -146,7 +146,7 @@ struct gpt_params {
bool random_prompt = false; // do not randomize prompt if none provided
bool use_color = false; // use color to distinguish generations and inputs
bool interactive = false; // interactive mode
- bool interactive_specials = false; // whether to allow special tokens from user, during interactive mode
+ bool special = false; // enable special token output
bool conversation = false; // conversation mode (does not print special tokens and suffix/prefix)
bool chatml = false; // chatml mode (used for models trained on chatml syntax)
bool prompt_cache_all = false; // save user input and generations to prompt cache
diff --git a/llama.cpp/main/main.cpp b/llama.cpp/main/main.cpp
index 700bfbcd..27ddb551 100644
--- a/llama.cpp/main/main.cpp
+++ b/llama.cpp/main/main.cpp
@@ -815,7 +815,7 @@ int main(int argc, char ** argv) {
if (input_echo && display) {
for (auto id : embd) {
// [jart] don't print special tokens until its design stabilizes
- const std::string token_str = llama_token_to_piece(ctx, id, false);
+ const std::string token_str = llama_token_to_piece(ctx, id, params.special);
printf("%s", token_str.c_str());
if (embd.size() > 1) {
@@ -958,7 +958,7 @@ int main(int argc, char ** argv) {
}
const auto line_pfx = ::llama_tokenize(ctx, params.input_prefix, false, true);
- const auto line_inp = ::llama_tokenize(ctx, buffer, false, params.interactive_specials);
+ const auto line_inp = ::llama_tokenize(ctx, buffer, false, params.conversation);
const auto line_sfx = ::llama_tokenize(ctx, params.input_suffix, false, true);
LOG("input tokens: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, line_inp).c_str());