Update README.md
Browse files
README.md
CHANGED
|
@@ -22,14 +22,20 @@ Highest quality will be Q6/Q8.
|
|
| 22 |
This merge was an experiment to test already established Roleplay, Fiction and Story
|
| 23 |
generation of "Tiefighter" with a some of "Orca 2"'s qualities.
|
| 24 |
|
| 25 |
-
For Imatrix plus this was a test of high precision in specific areas of the model leading to a slightly larger
|
| 26 |
In addition the Imatrix process itself used a larger "calibration" file than standard to further enhance quality.
|
| 27 |
|
|
|
|
|
|
|
| 28 |
A blank or standard Alpaca Template for text generation will work.
|
| 29 |
Currently "CHATML" is untested.
|
| 30 |
|
| 31 |
Context length: 4096.
|
| 32 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
# merge
|
| 34 |
|
| 35 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
|
|
|
| 22 |
This merge was an experiment to test already established Roleplay, Fiction and Story
|
| 23 |
generation of "Tiefighter" with a some of "Orca 2"'s qualities.
|
| 24 |
|
| 25 |
+
For Imatrix plus this was a test of high precision in specific areas of the model leading to a slightly larger compressed file.
|
| 26 |
In addition the Imatrix process itself used a larger "calibration" file than standard to further enhance quality.
|
| 27 |
|
| 28 |
+
The process added appoximately 310 MB to each compressed file.
|
| 29 |
+
|
| 30 |
A blank or standard Alpaca Template for text generation will work.
|
| 31 |
Currently "CHATML" is untested.
|
| 32 |
|
| 33 |
Context length: 4096.
|
| 34 |
|
| 35 |
+
Please see the orginal model card for specific details of use, additional credits and tips:
|
| 36 |
+
|
| 37 |
+
[KoboldAI/LLaMA2-13B-Tiefighter](https://huggingface.co/KoboldAI/LLaMA2-13B-Tiefighter)
|
| 38 |
+
|
| 39 |
# merge
|
| 40 |
|
| 41 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|