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

fix typos

Browse source
This commit is contained in:
Concedo 2023-07-03 21:36:42 +08:00
parent 3d2907d208
commit bfeb3471d7
3 changed files with 6 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

4
otherarch/gptj_v3.cpp
View file

@ -348,7 +348,7 @@ ModelLoadResult gptj_model_load(const std::string & fname, gptj_model & model, g
const auto & hparams = model.hparams; const auto & hparams = model.hparams;
size_t vram_total = 0; size_t vram_total = 0;
const int n_gpu = std::min(gpulayers, int(hparams.n_layer)); const int n_gpu = std::min(gpulayers, int(hparams.n_layer));
fprintf(stderr, "%s: [opencl] offloading %d layers to GPU\n", __func__, n_gpu); fprintf(stderr, "%s: [GPU] offloading %d layers to GPU\n", __func__, n_gpu);
for (int i = 0; i < n_gpu; ++i) { for (int i = 0; i < n_gpu; ++i) {
const auto & layer = model.layers[i]; const auto & layer = model.layers[i];
layer.c_attn_q_proj_w->backend = GGML_BACKEND_GPU; layer.c_attn_q_proj_w->backend = GGML_BACKEND_GPU;
@ -373,7 +373,7 @@ ModelLoadResult gptj_model_load(const std::string & fname, gptj_model & model, g
ggml_cuda_transform_tensor(layer.c_mlp_proj_w->data,layer.c_mlp_proj_w); vram_total += ggml_nbytes(layer.c_mlp_proj_w); ggml_cuda_transform_tensor(layer.c_mlp_proj_w->data,layer.c_mlp_proj_w); vram_total += ggml_nbytes(layer.c_mlp_proj_w);
#endif #endif
} }
fprintf(stderr, "%s: [opencl] total VRAM used: %zu MB\n", __func__, vram_total / 1024 / 1024); fprintf(stderr, "%s: [GPU] total VRAM used: %zu MB\n", __func__, vram_total / 1024 / 1024);
} }
#endif #endif

4
otherarch/mpt_v3.cpp
View file

@ -301,7 +301,7 @@ bool mpt_model_load(const std::string & fname, mpt_model & model, gpt_vocab & vo
const auto & hparams = model.hparams; const auto & hparams = model.hparams;
size_t vram_total = 0; size_t vram_total = 0;
const int n_gpu = std::min(gpulayers, int(hparams.n_layers)); const int n_gpu = std::min(gpulayers, int(hparams.n_layers));
fprintf(stderr, "%s: [opencl] offloading %d layers to GPU\n", __func__, n_gpu); fprintf(stderr, "%s: [GPU] offloading %d layers to GPU\n", __func__, n_gpu);
for (int i = 0; i < n_gpu; ++i) { for (int i = 0; i < n_gpu; ++i) {
const auto & layer = model.layers[i]; const auto & layer = model.layers[i];
layer.ffn_up_proj->backend = GGML_BACKEND_GPU; layer.ffn_up_proj->backend = GGML_BACKEND_GPU;
@ -320,7 +320,7 @@ bool mpt_model_load(const std::string & fname, mpt_model & model, gpt_vocab & vo
ggml_cuda_transform_tensor(layer.c_attn_out_proj_weight->data,layer.c_attn_out_proj_weight); vram_total += ggml_nbytes(layer.c_attn_out_proj_weight); ggml_cuda_transform_tensor(layer.c_attn_out_proj_weight->data,layer.c_attn_out_proj_weight); vram_total += ggml_nbytes(layer.c_attn_out_proj_weight);
#endif #endif
} }
fprintf(stderr, "%s: [opencl] total VRAM used: %zu MB\n", __func__, vram_total / 1024 / 1024); fprintf(stderr, "%s: [GPU] total VRAM used: %zu MB\n", __func__, vram_total / 1024 / 1024);
} }
#endif #endif

4
otherarch/neox_v3.cpp
View file

@ -335,7 +335,7 @@ ModelLoadResult gpt_neox_model_load(const std::string & fname, gpt_neox_model &
const auto & hparams = model.hparams; const auto & hparams = model.hparams;
size_t vram_total = 0; size_t vram_total = 0;
const int n_gpu = std::min(gpulayers, int(hparams.n_layer)); const int n_gpu = std::min(gpulayers, int(hparams.n_layer));
fprintf(stderr, "%s: [opencl] offloading %d layers to GPU\n", __func__, n_gpu); fprintf(stderr, "%s: [GPU] offloading %d layers to GPU\n", __func__, n_gpu);
for (int i = 0; i < n_gpu; ++i) { for (int i = 0; i < n_gpu; ++i) {
const auto & layer = model.layers[i]; const auto & layer = model.layers[i];
layer.c_attn_attn_w->backend = GGML_BACKEND_GPU; layer.c_attn_attn_w->backend = GGML_BACKEND_GPU;
@ -354,7 +354,7 @@ ModelLoadResult gpt_neox_model_load(const std::string & fname, gpt_neox_model &
ggml_cuda_transform_tensor(layer.c_mlp_proj_w->data,layer.c_mlp_proj_w); vram_total += ggml_nbytes(layer.c_mlp_proj_w); ggml_cuda_transform_tensor(layer.c_mlp_proj_w->data,layer.c_mlp_proj_w); vram_total += ggml_nbytes(layer.c_mlp_proj_w);
#endif #endif
} }
fprintf(stderr, "%s: [opencl] total VRAM used: %zu MB\n", __func__, vram_total / 1024 / 1024); fprintf(stderr, "%s: [GPU] total VRAM used: %zu MB\n", __func__, vram_total / 1024 / 1024);
} }
#endif #endif