MarsupialAI
commited on
Commit
•
4c0b98f
1
Parent(s):
380c2bf
Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,20 @@ tags:
|
|
7 |
- ERP
|
8 |
- chat
|
9 |
- storywriting
|
10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
- ERP
|
8 |
- chat
|
9 |
- storywriting
|
10 |
+
---
|
11 |
+
# Kitchen Sink 103b
|
12 |
+
|
13 |
+
|
14 |
+
This model is a rotating-stack merge of three 70b models, each of which is a merge of multiple merges and finetunes. The result is a large model that contains a little bit of everything - including the kitchen sink.
|
15 |
+
|
16 |
+
Component models for this stack are
|
17 |
+
- royallab/Aetheria-L2-70B
|
18 |
+
- lizpreciatior/lzlv_70b_fp16_hf
|
19 |
+
- Sao10K/WinterGoddess-1.4x-70B-L2
|
20 |
+
|
21 |
+
Components of those models include Nous-Hermes-Llama2-70b, Xwin-LM-7B-V0.1, Mythospice-70b, Euryale-1.3-L2-70B, tulu-2-dpo-70b, GOAT-70B-Storytelling, Platypus2-70B-instruct, Lila-70B, SunsetBoulevard, and some private LoRAs.
|
22 |
+
|
23 |
+
|
24 |
+
# WTF is a rotating-stack merge?
|
25 |
+
|
26 |
+
Jeb Carter found that the performance of a stacked merge could be significantly improved by reversing the model order in the stack, and then doing a linear merge between the original and reversed stacks. That is what I did here, creating three stacked merges with the three source models, and then doing a 1:1:1 linear merge of all three stacks. The exact merge configs can be found in the recipe.txt file.
|