1000B

README.md

@@ -10,5 +10,5 @@ Please refer to that repository's model card for details.
 
 The current revision is a quantization of the 700B token checkpoint.
 
-The conversion was done with [llama.cpp](https://github.com/ggerganov/llama.cpp) version b1641 (6744dbe924a317e3e2a5a2a4a2037061b2223449)
+The conversion was done with [llama.cpp](https://github.com/ggerganov/llama.cpp) version b2354 (e25fb4b18fcedb9bed6be4585cf842e9a669b28b)
 on a Google Compute machine generously sponsored by [Valohai](https://valohai.com/).

ggml-model-Q3_K.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05854a5ef2b53e2174b6d90e8f4d459ada8ab03213a153eef95ee3b893b5f3ad
-size 18477373600
+oid sha256:50f23c03e012f0e3434e5691dc9bbde86bc861517e192669ab294aa05c3c19d9
+size 18580134048

ggml-model-Q4_K.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1cd5c3ef547904f513855864e2ba088d0b7631abc3cce7a7dbf90b8e8976dea6
+oid sha256:02d1b0dbcfe2226947b5e53f05474cfa9c562dfc425f114abdf7feb044c070b6
 size 22439901344

ggml-model-Q5_K.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1496c6841aa49b8731e4b0bcc91bac3a8bf1c2f03468b234715468de0b29088d
+oid sha256:fdebb377b7d727cb0a74f0a0e8b994fed4f91a0d3feff7c3f289051409e1f2f2
 size 26113472672