Update README.md
Browse files
README.md
CHANGED
@@ -30,7 +30,8 @@ Please view either repo for details on the remaster's results, and other importa
|
|
30 |
These are "final" result files of the full precision rebuild (including end result merge(s)) minus
|
31 |
GGUF and Imatrix level upscaling / adjustments which occuring during "GGUFing" processes.
|
32 |
|
33 |
-
If you use these to create your own GGUFs, please use "outfile" at F32 for best results.
|
|
|
34 |
|
35 |
Imatrix processes should use a stable dataset(s) of at least 500 "chunks" or more.
|
36 |
If smaller dataset(s) are used this may corrupt or reduce the quality of the Imatrix builds.
|
@@ -53,6 +54,6 @@ of prompts in usage. This is a factor due to the precision upscale.
|
|
53 |
|
54 |
Doing this will ensure the quality of the upscale is maximized in the GGUFs.
|
55 |
|
56 |
-
Happy GGUFing, EXL2ing, GPTQing, AWQing, HQQing.
|
57 |
|
58 |
Long live open source!
|
|
|
30 |
These are "final" result files of the full precision rebuild (including end result merge(s)) minus
|
31 |
GGUF and Imatrix level upscaling / adjustments which occuring during "GGUFing" processes.
|
32 |
|
33 |
+
If you use these to create your own GGUFs, please use "outfile" at F32 for best results. If
|
34 |
+
you use F16 this will reduce the quality by a factor of 2 or higher.
|
35 |
|
36 |
Imatrix processes should use a stable dataset(s) of at least 500 "chunks" or more.
|
37 |
If smaller dataset(s) are used this may corrupt or reduce the quality of the Imatrix builds.
|
|
|
54 |
|
55 |
Doing this will ensure the quality of the upscale is maximized in the GGUFs.
|
56 |
|
57 |
+
Happy GGUFing, EXL2ing, GPTQing, AWQing, HQQing and of course "Merging".
|
58 |
|
59 |
Long live open source!
|