py : fix persimmon `n_rot` conversion (#5460)

* convert : fix persimmon offical weight conversion to write correct n_rot.

* Update convert-persimmon-to-gguf.py

---------

Co-authored-by: Georgi Gerganov <redacted>

diff --git a/convert-persimmon-to-gguf.py b/convert-persimmon-to-gguf.py
index d2be805d12ed0b3eb9a7d6243c33ed63e053c007..def210531e27b1794597573790b799eff81a4d34 100755 (executable)
--- a/convert-persimmon-to-gguf.py
+++ b/convert-persimmon-to-gguf.py
@@ -88,7 +88,8 @@ def main():
     gguf_writer.add_embedding_length(hidden_size)
     gguf_writer.add_block_count(block_count)
     gguf_writer.add_feed_forward_length(hparams.ffn_hidden_size)
-    gguf_writer.add_rope_dimension_count(hidden_size // head_count)
+    # ref: https://github.com/ggerganov/llama.cpp/pull/4889/commits/eea19039fc52ea2dbd1aab45b59ab4e3e29a3443
+    gguf_writer.add_rope_dimension_count(hidden_size // head_count // 2)
     gguf_writer.add_head_count(head_count)
     gguf_writer.add_head_count_kv(head_count_kv)
     gguf_writer.add_rope_freq_base(hparams.rotary_emb_base)