ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-03-02 10:00:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Revert delta net 3 (#1339)

Browse Source 
* Revert "Simplify delta-net (#1335)"

This reverts commit e5fc30244c.

* Revert "Fused delta net 3 (#1333)"

This reverts commit 7b68353e09.
This commit is contained in:
Kawrakow
2026-02-28 13:12:08 +01:00
committed by GitHub
parent e5fc30244c
commit 1922449b2c
9 changed files with 693 additions and 101 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
common/common.cpp
View File

@@ -1533,7 +1533,7 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
}
if (arg == "-fdn" || arg == "--fused-delta-net") {
CHECK_ARG
fprintf(stderr, "=================== %s has been deprecated\n", arg.c_str());
params.fused_delta_net = std::stoi(argv[i]);
return true;
}
if (arg == "-smf16" || arg == "--split-mode-f16") {
@@ -2276,6 +2276,7 @@ void gpt_params_print_usage(int /*argc*/, char ** argv, const gpt_params & param
options.push_back({ "*", "-grt, --graph-reduce-type", "Type for data exchange between GPUs (default: %s)", "f32"});
options.push_back({ "*", "-smgs, --split-mode-graph-scheduling,", "Force Split Mode Graph Scheduling (default: %d)", params.split_mode_graph_scheduling});
options.push_back({ "*", "-sas, --scheduler_async,", "Async evaluation of compute graphs: %d)", params.scheduler_async});
options.push_back({ "*", "-fdn, --fused-delta-net N", "Use fused delta-net when batch size is <= N with recurrent models: %d)", params.fused_delta_net});
options.push_back({ "*", "-vq, --validate-quants", "validate quantized data while loading the model (default: %d)", params.validate_quants});
options.push_back({ "*", "-p, --prompt PROMPT", "prompt to start generation with\n"
"in conversation mode, this will be used as system prompt\n"
@@ -3354,6 +3355,7 @@ struct llama_context_params common_context_params_to_llama(const gpt_params & pa
cparams.split_mode_graph_scheduling = params.split_mode_graph_scheduling;
//cparams.split_mode_f16 = params.split_mode_f16;
cparams.scheduler_async = params.scheduler_async;
cparams.fused_delta_net = params.fused_delta_net;
cparams.min_experts = params.min_experts;
cparams.thresh_experts = params.thresh_experts;
cparams.only_active_experts = params.only_active_exps;
@@ -4364,6 +4366,7 @@ void yaml_dump_non_result_info(FILE * stream, const gpt_params & params, const l
//fprintf(stream, "split_mode_f16: %s # default: true\n", params.split_mode_f16 ? "true" : "false");
fprintf(stream, "reduce_type: %s # default f16\n", params.reduce_type.c_str());
fprintf(stream, "scheduler_async: %s # default: false\n", params.scheduler_async ? "true" : "false");
fprintf(stream, "fused_delta_net: %d # default: 0\n", params.fused_delta_net );
fprintf(stream, "ser: %d,%g # defaulr: -1,0\n", params.min_experts, params.thresh_experts);
fprintf(stream, "temp: %f # default: 0.8\n", sparams.temp);