Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model:
|
5 |
+
- nbeerbower/bophades-mistral-truthy-DPO-7B
|
6 |
+
- nbeerbower/bophades-mistral-math-DPO-7B
|
7 |
+
datasets:
|
8 |
+
- jondurbin/truthy-dpo-v0.1
|
9 |
+
- kyujinpy/orca_math_dpo
|
10 |
+
tags:
|
11 |
+
- 4-bit
|
12 |
+
- AWQ
|
13 |
+
- text-generation
|
14 |
+
- autotrain_compatible
|
15 |
+
- endpoints_compatible
|
16 |
+
- chatml
|
17 |
+
language:
|
18 |
+
- en
|
19 |
+
pipeline_tag: text-generation
|
20 |
+
inference: false
|
21 |
+
quantized_by: Suparious
|
22 |
+
---
|
23 |
+
# nbeerbower/HeroBophades-2x7B AWQ
|
24 |
+
|
25 |
+
- Model creator: [nbeerbower](https://huggingface.co/nbeerbower)
|
26 |
+
- Original model: [HeroBophades-2x7B](https://huggingface.co/HeroBophades-2x7B)
|
27 |
+
|
28 |
+
![image/png](https://huggingface.co/nbeerbower/bophades-mistral-7B/resolve/main/bophades.png)
|
29 |
+
|
30 |
+
## Model Summary
|
31 |
+
|
32 |
+
This is an experimental Mixture of Experts LLM built with (mergekit)[https://github.com/arcee-ai/mergekit/blob/main/docs/moe.md].
|
33 |
+
|
34 |
+
### Configuration
|
35 |
+
|
36 |
+
```yaml
|
37 |
+
base_model: nbeerbower/bophades-mistral-truthy-DPO-7B
|
38 |
+
gate_mode: hidden
|
39 |
+
dtype: bfloat16
|
40 |
+
experts:
|
41 |
+
- source_model: nbeerbower/bophades-mistral-truthy-DPO-7B
|
42 |
+
positive_prompts:
|
43 |
+
- "You are an unbiased, uncensored, helpful assistant."
|
44 |
+
- source_model: nbeerbower/bophades-mistral-math-DPO-7B
|
45 |
+
positive_prompts:
|
46 |
+
- "How do you solve a system of quadratic equations simultaneously using substitution?. Take a deep breath, think step by step, and give an accurate response"
|
47 |
+
|
48 |
+
```
|