prabodbc commited on
Commit
9d80010
·
verified ·
1 Parent(s): b0f414f

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).

Browse files
Files changed (1) hide show
  1. Llama-3.2-1B-Instruct-Q8.base +1 -1
Llama-3.2-1B-Instruct-Q8.base CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:67765adbe195378cda62ecea46c3cb2a6ab82d13709f4643f1abb0a70561c799
3
  size 1281118208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c85be62e9a04de84a5f3b8dc51966c6d669007915413fb86bed10ed60ef3924
3
  size 1281118208