SanjiWatsuki
commited on
Commit
•
5405223
1
Parent(s):
fb14c30
Update README.md
Browse files
README.md
CHANGED
@@ -13,7 +13,23 @@ tags:
|
|
13 |
|
14 |
This repository hosts quantized GGUF files for **Loyal-Macaroni-Maid-7B**, a 7B model aimed at having engaging RP with solid character card adherence and being a smart cookie at the same time.
|
15 |
|
16 |
-
In my limited testing, it's a great RP model suitable for RP/ERP with sharp reasoning skills for a 7B.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
<!-- description end -->
|
19 |
<!-- prompt-template start -->
|
|
|
13 |
|
14 |
This repository hosts quantized GGUF files for **Loyal-Macaroni-Maid-7B**, a 7B model aimed at having engaging RP with solid character card adherence and being a smart cookie at the same time.
|
15 |
|
16 |
+
In my limited testing, it's a great RP model suitable for RP/ERP with sharp reasoning skills for a 7B.
|
17 |
+
|
18 |
+
On benchmarks like MT-Bench, this RP model scores shockingly well.
|
19 |
+
|
20 |
+
| Model | MT Bench | MMLU |
|
21 |
+
|------------------------|----------|------|
|
22 |
+
| GPT-4-Turbo | 9.32 | |
|
23 |
+
| GPT-4 | 8.99 | 86.4 |
|
24 |
+
| Starling-7B | 8.09 | 63.9 |
|
25 |
+
| Claude-2 | 8.06 | 78.5 |
|
26 |
+
| Loyal-Macaroni-Maid-7B | 7.95 | ~64.9|
|
27 |
+
| GPT-3.5-Turbo | 7.94 | 70 |
|
28 |
+
| Claude-1 | 7.9 | 77 |
|
29 |
+
| Openchat-3.5 | 7.81 | 64.3 |
|
30 |
+
| Zephyr-7B-beta | 7.34 | 61.4 |
|
31 |
+
| Llama-2-70b-chat-hf | 6.86 | 63 |
|
32 |
+
| Neural-chat-7b-v3-1 | 6.84 | 62.4 |
|
33 |
|
34 |
<!-- description end -->
|
35 |
<!-- prompt-template start -->
|