Re-convert with fixed rotary q/k permutation (baseRT-internal#114); NIAH-validated at 2k/8k. Q4 now uses the default-q4 quality profile (embeddings/head at higher precision).

Llama-3.2-1B-Instruct-Q4.base

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:292fe9db91c3a0da29e4d7e049eecc6fc30b0b9e009df3775b0f4894bac1a65e
-size 701861888
+oid sha256:9558edb638113e8c702d530d29e677177b365859a8977df1a487b78ba5c9ba8f
+size 1079447552