Update README.md
Browse files
README.md
CHANGED
|
@@ -8,6 +8,17 @@ tags:
|
|
| 8 |
- sillytavern
|
| 9 |
- idol
|
| 10 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
|
| 13 |
|
|
|
|
| 8 |
- sillytavern
|
| 9 |
- idol
|
| 10 |
---
|
| 11 |
+
### This version was created during the 3.1 release and has several issues:
|
| 12 |
+
- The unreviewed technology was not mature at the time, leading to numerous rejection problems.
|
| 13 |
+
- The tokenizer.json used the initial LLama 3.1 version, which Meta later modified. I tested the new tokenizer.json and found it incompatible, so I did not make any changes.
|
| 14 |
+
### This is the latest iteration in the series, which has undergone extensive modifications.
|
| 15 |
+
https://huggingface.co/aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored
|
| 16 |
+
|
| 17 |
+
### DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored Test File:
|
| 18 |
+
https://huggingface.co/aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored/blob/main/Uncensored_Test/harmful_behaviors.csv
|
| 19 |
+
|
| 20 |
+
### DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored Test results:
|
| 21 |
+
Out of 520 test questions, only one was rejected, resulting in a pass rate of 99.81%.
|
| 22 |
|
| 23 |
|
| 24 |
|