github/gpt4all
1
0
Fork 0
mirror of https://github.com/nomic-ai/gpt4all.git synced 2025-09-06 11:00:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

remove old llama.cpp submodules

Browse Source
This commit is contained in:
Cebtenzzre
2023-10-05 11:59:23 -04:00
committed by Adam Treat
parent cc6db61c93
commit d87573ea75
5 changed files with 0 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
gpt4all-backend/llama.cpp-230511

Submodule gpt4all-backend/llama.cpp-230511 deleted from f826aac617

1
gpt4all-backend/llama.cpp-230519

Submodule gpt4all-backend/llama.cpp-230519 deleted from 5ea4339273

10
gpt4all-backend/llamamodel.cpp
View File

@@ -39,15 +39,10 @@ const char *modelType_ = "LLaMA";
struct gpt_params {
int32_t seed = -1; // RNG seed
int32_t n_keep = 0; // number of tokens to keep from initial prompt
#if LLAMA_DATE <= 230511
int32_t n_parts = -1; // amount of model parts (-1 = determine from model dimensions)
#endif
#if LLAMA_DATE >= 230519
// sampling parameters
float tfs_z = 1.0f; // 1.0 = disabled
float typical_p = 1.0f; // 1.0 = disabled
#endif
std::string prompt = "";
@@ -57,7 +52,6 @@ struct gpt_params {
bool use_mlock = false; // use mlock to keep model in memory
};
#if LLAMA_DATE >= 230519
static int llama_sample_top_p_top_k(
llama_context *ctx,
const llama_token *last_n_tokens_data,
@@ -85,7 +79,6 @@ static int llama_sample_top_p_top_k(
llama_sample_temperature(ctx, &candidates_p, temp);
return llama_sample_token(ctx, &candidates_p);
}
#endif
struct LLamaPrivate {
const std::string modelPath;
@@ -150,9 +143,6 @@ bool LLamaModel::loadModel(const std::string &modelPath)
#else
d_ptr->params.use_mlock = params.use_mlock;
#endif
#if LLAMA_DATE <= 230511
d_ptr->params.n_parts = params.n_parts;
#endif
#ifdef GGML_USE_METAL
std::cerr << "llama.cpp: using Metal" << std::endl;
// metal always runs the whole model if n_gpu_layers is not 0, at least

2
gpt4all-backend/llmodel_shared.h
View File

@@ -80,7 +80,6 @@ struct llm_kv_cache {
}
};
#if LLAMA_DATE >= 230519
inline void ggml_graph_compute_g4a(llm_buffer& buf, ggml_cgraph * graph, int n_threads) {
struct ggml_cplan plan = ggml_graph_plan(graph, n_threads);
if (plan.work_size > 0) {
@@ -89,4 +88,3 @@ inline void ggml_graph_compute_g4a(llm_buffer& buf, ggml_cgraph * graph, int n_t
}
ggml_graph_compute(graph, &plan);
}
#endif