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

Add additional debug info and increased ctx sizes, fixed a bug loading vulkan config

Browse source
This commit is contained in:
Concedo 2024-04-25 23:07:37 +08:00
parent b56805a2ba
commit 0871c7cbd1
2 changed files with 12 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

10
gpttype_adapter.cpp
View file

@ -1876,6 +1876,14 @@ generation_outputs gpttype_generate(const generation_inputs inputs)
std::fill(last_n_tokens.begin(), last_n_tokens.end(), 0);
n_past = 0;
if (debugmode==1)
{
std::string outstr = "";
printf("\n\n[Debug: Dump Raw Input Tokens, format: %d]\n", file_format);
outstr += get_tok_vec_str(embd_inp);
printf("%s\n", RemoveBell(outstr).c_str());
}
bool is_mamba = (file_format == FileFormat::GGUF_GENERIC && file_format_meta.model_architecture==GGUFArch::ARCH_MAMBA);
if (file_format == FileFormat::RWKV_1 || file_format==FileFormat::RWKV_2 || is_mamba)
@ -2011,7 +2019,7 @@ generation_outputs gpttype_generate(const generation_inputs inputs)
if (debugmode==1)
{
std::string outstr = "";
printf("\n[Debug: Dump Input Tokens, format: %d]\n", file_format);
printf("\n[Debug: Dump Forwarded Input Tokens, format: %d]\n", file_format);
outstr += get_tok_vec_str(embd_inp);
outstr += "\n\n[Debug: n_past="+std::to_string(n_past)+" Context Size = " + std::to_string(current_context_tokens.size()) + "]\n";
outstr += get_tok_vec_str(current_context_tokens);