Skip to content

RWKV v6: Add time_mix_decay_w1/w2 in quant exclusion list #9387

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 10, 2024
Merged

RWKV v6: Add time_mix_decay_w1/w2 in quant exclusion list #9387

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions convert_hf_to_gguf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -302,6 +302,8 @@ def prepare_tensors(self):
gguf.MODEL_TENSOR.TIME_MIX_FIRST,
gguf.MODEL_TENSOR.TIME_MIX_W1,
gguf.MODEL_TENSOR.TIME_MIX_W2,
gguf.MODEL_TENSOR.TIME_MIX_DECAY_W1,
gguf.MODEL_TENSOR.TIME_MIX_DECAY_W2,
)
)
or not new_name.endswith(".weight")
Expand Down
2 changes: 2 additions & 0 deletions src/llama.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17530,6 +17530,8 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s
quantize &= name.find("time_mix_first.weight") == std::string::npos;
quantize &= name.find("time_mix_w1.weight") == std::string::npos;
quantize &= name.find("time_mix_w2.weight") == std::string::npos;
quantize &= name.find("time_mix_decay_w1.weight") == std::string::npos;
quantize &= name.find("time_mix_decay_w2.weight") == std::string::npos;

// do not quantize relative position bias (T5)
quantize &= name.find("attn_rel_b.weight") == std::string::npos;
Expand Down
Loading