vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-11 01:24:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

You'll never take us alive

Browse source 
We swore that death will do us part
They'll call our crimes a work of art
This commit is contained in:
Concedo 2025-01-09 11:27:06 +08:00
commit e788b8289a
19 changed files with 1866 additions and 1704 deletions
Download patch file Download diff file Expand all files Collapse all files

3
src/llama-quant.cpp
View file

@ -875,7 +875,8 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
// update the gguf meta data as we go
gguf_set_tensor_type(ctx_outs[cur_split].get(), name.c_str(), new_type);
gguf_set_tensor_data(ctx_outs[cur_split].get(), name.c_str(), new_data, new_size);
GGML_ASSERT(gguf_get_tensor_size(ctx_outs[cur_split].get(), gguf_find_tensor(ctx_outs[cur_split].get(), name.c_str())) == new_size);
gguf_set_tensor_data(ctx_outs[cur_split].get(), name.c_str(), new_data);
// write tensor data + padding
fout.write((const char *) new_data, new_size);