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

remove from llama_context_params

Browse source
This commit is contained in:
netrunnereve 2023-08-06 00:44:29 -04:00
parent ce6d86ec41
commit 0480362f12
3 changed files with 0 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

1
examples/common.cpp
View file

@ -668,7 +668,6 @@ struct llama_context_params llama_context_params_from_gpt_params(const gpt_param
lparams.embedding = params.embedding;
lparams.rope_freq_base = params.rope_freq_base;
lparams.rope_freq_scale = params.rope_freq_scale;
lparams.pp_threads = params.pp_threads;
return lparams;
}

1
llama.cpp
View file

@ -895,7 +895,6 @@ struct llama_context_params llama_context_default_params() {
/*.rms_norm_eps =*/ LLAMA_DEFAULT_RMS_EPS,
/*.gpu_layers =*/ 0,
/*.main_gpu =*/ 0,
/*.pp_threads =*/ GGML_DEFAULT_N_THREADS,
/*.tensor_split =*/ nullptr,
/*.rope_freq_base =*/ 10000.0f,
/*.rope_freq_scale =*/ 1.0f,

1
llama.h
View file

@ -94,7 +94,6 @@ extern "C" {
float rms_norm_eps; // rms norm epsilon (TEMP - will be moved to model hparams)
int32_t n_gpu_layers; // number of layers to store in VRAM
int32_t main_gpu; // the GPU that is used for scratch and small tensors
int32_t pp_threads; // number of threads used for prompt processing only
const float * tensor_split; // how to split layers across multiple GPUs (size: LLAMA_MAX_DEVICES)