lora : fix llama conversion script with model having ROPE_FREQS

[ngxson]

Resolve #9114

---

• I have read the contributing guidelines
• Self-reported review complexity:
  • Low
  • Medium
  • High

[Ujjawal-K-Panchal]

This fixes the issue #9114 I raised. Please look at the final output log here.

[compilade]

For Phi-3, vocab-only conversion is also affected by these rope_freqs tensors, because this is in set_gguf_parameters. (which makes vocab-only Phi-3-128k models produce invalid GGUF files (this is already a problem on master))

A more general solution to both LoRA and vocab-only conversions should be possible.

Maybe some kind of self.generate_extra_tensors() which would be called by self.prepare_tensors() before it calls self.get_tensors(). And LoraModel could simply override generate_extra_tensors() to a no-op (and vocab-only conversion does not call prepare_tensors). It can be done in a follow-up PR, though.

[ngxson]

OK I'll merge this now and will let you refactor this further in a follow-up PR.
Thank you for the help!