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

server: rename template mistral to llama2

Browse source
This commit is contained in:
ngxson 2024-02-09 09:55:20 +01:00
parent 27976c31b6
commit 269437e4eb
3 changed files with 5 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

4
examples/server/oai.hpp
View file

@ -21,8 +21,8 @@ inline static json oaicompat_completion_params_parse(
json llama_params; json llama_params;
bool using_chatml = chat_template == "chatml"; bool using_chatml = chat_template == "chatml";
std::string formatted_prompt = using_chatml std::string formatted_prompt = using_chatml
? format_chatml(body["messages"]) // OpenAI 'messages' to chatml ? format_chatml(body["messages"]) // OpenAI 'messages' to chatml (with <|im_start|>,...)
: format_mistral(body["messages"]); // OpenAI 'messages' to mistral format : format_llama2(body["messages"]); // OpenAI 'messages' to llama2 (with [INST],...)
llama_params["__oaicompat"] = true; llama_params["__oaicompat"] = true;

2
examples/server/server.cpp
View file

@ -1861,7 +1861,7 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
printf(" -gan N, --grp-attn-n N set the group attention factor to extend context size through self-extend(default: 1=disabled), used together with group attention width `--grp-attn-w`"); printf(" -gan N, --grp-attn-n N set the group attention factor to extend context size through self-extend(default: 1=disabled), used together with group attention width `--grp-attn-w`");
printf(" -gaw N, --grp-attn-w N set the group attention width to extend context size through self-extend(default: 512), used together with group attention factor `--grp-attn-n`"); printf(" -gaw N, --grp-attn-w N set the group attention width to extend context size through self-extend(default: 512), used together with group attention factor `--grp-attn-n`");
printf(" --chat-template FORMAT_NAME"); printf(" --chat-template FORMAT_NAME");
printf(" set chat template, possible valus is: mistral, chatml (default %s)", sparams.chat_template.c_str()); printf(" set chat template, possible valus is: llama2, chatml (default %s)", sparams.chat_template.c_str());
printf("\n"); printf("\n");
} }

4
examples/server/utils.hpp
View file

@ -167,7 +167,7 @@ static T json_value(const json &body, const std::string &key, const T &default_v
: default_value; : default_value;
} }
inline std::string format_mistral(std::vector<json> messages) inline std::string format_llama2(std::vector<json> messages)
{ {
std::ostringstream output; std::ostringstream output;
bool is_inside_turn = false; bool is_inside_turn = false;
@ -190,7 +190,7 @@ inline std::string format_mistral(std::vector<json> messages)
} }
} }
LOG_VERBOSE("format_mistral", {{"text", output.str()}}); LOG_VERBOSE("format_llama2", {{"text", output.str()}});
return output.str(); return output.str();
} }