Update README.md
Browse files
README.md
CHANGED
|
@@ -10,9 +10,19 @@ tags:
|
|
| 10 |
---
|
| 11 |
|
| 12 |
28/04/2024- UPDATE:
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
# merge
|
| 17 |
|
| 18 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
28/04/2024- UPDATE:
|
| 13 |
+
|
| 14 |
+
Fixed tokenizer / vocab issues.
|
| 15 |
+
Verified operation, conversion to GGUF now works too.
|
| 16 |
+
GGUF uploaded, with Imatrix Plus GGUFs to follow shortly.
|
| 17 |
+
|
| 18 |
+
Imatrix Plus GGUFs are [here](https://huggingface.co/DavidAU/D_AU-Orac-13B-Tiefighter-slerp-imat-plus-GGUF)
|
| 19 |
+
This includes all Imatrix compressions as well as regular "Qs" which have also been "Imatrixed" too.
|
| 20 |
+
|
| 21 |
+
"Imatrix Plus" is an upgraded form of Imatrix which using full precision for specific parts of the compression.
|
| 22 |
+
|
| 23 |
+
This results in slightly higher quality, especially at lower compressions.
|
| 24 |
+
This method is applied across all compressions from IQ1 to Q8.
|
| 25 |
+
|
| 26 |
# merge
|
| 27 |
|
| 28 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|