Update README.md
Browse files
README.md
CHANGED
|
@@ -8,6 +8,26 @@ tags:
|
|
| 8 |
- mergekit
|
| 9 |
- merge
|
| 10 |
license: llama3.3
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
# merge
|
| 13 |
|
|
|
|
| 8 |
- mergekit
|
| 9 |
- merge
|
| 10 |
license: llama3.3
|
| 11 |
+
metrics:
|
| 12 |
+
- perplexity
|
| 13 |
+
- accuracy
|
| 14 |
+
---
|
| 15 |
+
# About.
|
| 16 |
+
|
| 17 |
+
A solid model. Not too fancy, only serious stuff for a smart model.
|
| 18 |
+
|
| 19 |
+
Most of the best 70b finetunes packed into one multi-echelon merge, with the excellent Negative Llama as a Lead.
|
| 20 |
+
- Sidekick A : Smarteaz is my L3.3 70b smart base with the best models abliterated by Huihui-AI, with Deepseek R1 Distill, Tulu and Nemotron (3.1) mixed within.
|
| 21 |
+
- Sidekick B : Hermes is a very qualitative L3.1 70b model, lorablated by Maxime Labonne.
|
| 22 |
+
|
| 23 |
+
Both models empower Negative Llama and make it smarter than ever.
|
| 24 |
+
Nemesis V1.1 can be used as a standalone, or a smart-base for merges like Smarteaz, but I fear that it might dilute its qualities.
|
| 25 |
+
Smart roleplay is also possible, of course. It's not the funniest model, though.
|
| 26 |
+
|
| 27 |
+
- Perplexity Wikitext eng 512 : 3.35. (member of my best class)
|
| 28 |
+
- Arc-C : 62 (same)
|
| 29 |
+
- ARC-E : 83 (same)
|
| 30 |
+
|
| 31 |
---
|
| 32 |
# merge
|
| 33 |
|