tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-27 03:55:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

examples : remove infill (#13283)

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-05-07 10:28:02 +03:00
committed by GitHub
parent 6c7fd67b64
commit 4773d7a02f
7 changed files with 6 additions and 655 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
common/arg.cpp
View File

@ -1283,7 +1283,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params) {
params.use_color = true;
}
).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_INFILL, LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP}));
).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP}));
add_opt(common_arg(
{"-t", "--threads"}, "N",
string_format("number of threads to use during generation (default: %d)", params.cpuparams.n_threads),
@ -1416,7 +1416,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
add_opt(common_arg(
{"-n", "--predict", "--n-predict"}, "N",
string_format(
ex == LLAMA_EXAMPLE_MAIN || ex == LLAMA_EXAMPLE_INFILL
ex == LLAMA_EXAMPLE_MAIN
? "number of tokens to predict (default: %d, -1 = infinity, -2 = until context filled)"
: "number of tokens to predict (default: %d, -1 = infinity)",
params.n_predict),
@ -1655,7 +1655,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.input_prefix = value;
params.enable_chat_template = false;
}
).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_INFILL}));
).set_examples({LLAMA_EXAMPLE_MAIN}));
add_opt(common_arg(
{"--in-suffix"}, "STRING",
"string to suffix after user inputs with (default: empty)",
@ -1663,7 +1663,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.input_suffix = value;
params.enable_chat_template = false;
}
).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_INFILL}));
).set_examples({LLAMA_EXAMPLE_MAIN}));
add_opt(common_arg(
{"--no-warmup"},
"skip warming up the model with an empty run",
@ -1680,7 +1680,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params) {
params.spm_infill = true;
}
).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_INFILL}));
).set_examples({LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"--samplers"}, "SAMPLERS",
string_format("samplers that will be used for generation in the order, separated by \';\'\n(default: %s)", sampler_type_names.c_str()),
@ -2892,7 +2892,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params) {
params.simple_io = true;
}
).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_INFILL}));
).set_examples({LLAMA_EXAMPLE_MAIN}));
add_opt(common_arg(
{"--positive-file"}, "FNAME",
string_format("positive prompts file, one prompt per line (default: '%s')", params.cvector_positive_file.c_str()),

1
common/common.h
View File

@ -66,7 +66,6 @@ enum llama_example {
LLAMA_EXAMPLE_COMMON,
LLAMA_EXAMPLE_SPECULATIVE,
LLAMA_EXAMPLE_MAIN,
LLAMA_EXAMPLE_INFILL,
LLAMA_EXAMPLE_EMBEDDING,
LLAMA_EXAMPLE_PERPLEXITY,
LLAMA_EXAMPLE_RETRIEVAL,