Fix missing rope_freqs with convert_hf_to_gguf (#402)

* lora : fix llama conversion script with ROPE_FREQS

* convert : refactor rope_freqs generation

This should also fix vocab-only conversion for Phi-3.

* convert : adapt MiniCPM3 to separate rope_freqs insertion

MiniCPM3's tokenizer is treated as a SentencePiece tokenizer to avoid
having to run its custom Python code which mixes tokenization
in the same file as tool calls.

gguf-py : add long and short RoPE factors to tensor mappings

Empty, but the key names are used to populate the mappings.

---------

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: Francis Couture-Harpin <git@compilade.net>

1 files changed, 4 insertions, 0 deletions

diff --git a/convert_lora_to_gguf.py b/convert_lora_to_gguf.py
index a88d0d4a..ef088034 100755
--- a/convert_lora_to_gguf.py
+++ b/convert_lora_to_gguf.py
@@ -331,6 +331,10 @@ if __name__ == '__main__':
                 self.gguf_writer.add_float32(gguf.Keys.Adapter.LORA_ALPHA, self.lora_alpha)
                 super().set_gguf_parameters()
 
+            def generate_extra_tensors(self) -> Iterable[tuple[str, Tensor]]:
+                # Never add extra tensors (e.g. rope_freqs) for LoRA adapters
+                return ()
+
             def get_tensors(self) -> Iterator[tuple[str, Tensor]]:
                 tensor_map: dict[str, PartialLoraTensor] = {}