Update README.md
Browse files
README.md
CHANGED
@@ -9,21 +9,19 @@ tags:
|
|
9 |
---
|
10 |
# miqurelian-120b
|
11 |
|
12 |
-
This is
|
13 |
|
14 |
## Model Details
|
15 |
|
16 |
- Max Context: 32768 tokens
|
17 |
- Layers: 140
|
18 |
|
19 |
-
### Prompt template
|
20 |
|
21 |
```
|
22 |
<s>[INST] {prompt} [/INST]
|
23 |
```
|
24 |
|
25 |
-
## Merge Details
|
26 |
-
|
27 |
### Merge Method
|
28 |
|
29 |
This model was merged using the [linear](https://arxiv.org/abs/2203.05482) merge method.
|
|
|
9 |
---
|
10 |
# miqurelian-120b
|
11 |
|
12 |
+
This is a 120b merge created by interleaving layers of [miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf) with [Aurelian](https://huggingface.co/grimulkan/aurelian-v0.5-70b-rope8-32K-fp16), a creative writing model, using [mergekit](https://github.com/cg123/mergekit). It performs approximtely SOTA for long-context creative writing tasks that require strong semantic coherence.
|
13 |
|
14 |
## Model Details
|
15 |
|
16 |
- Max Context: 32768 tokens
|
17 |
- Layers: 140
|
18 |
|
19 |
+
### Prompt template
|
20 |
|
21 |
```
|
22 |
<s>[INST] {prompt} [/INST]
|
23 |
```
|
24 |
|
|
|
|
|
25 |
### Merge Method
|
26 |
|
27 |
This model was merged using the [linear](https://arxiv.org/abs/2203.05482) merge method.
|