Upload README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,63 @@
|
|
1 |
-
---
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model:
|
3 |
+
- ruslanmv/ai-medical-model-32bit
|
4 |
+
- Locutusque/Llama-3-Hercules-5.0-8B
|
5 |
+
- refuelai/Llama-3-Refueled
|
6 |
+
library_name: transformers
|
7 |
+
tags:
|
8 |
+
- mergekit
|
9 |
+
- merge
|
10 |
+
- medical
|
11 |
+
- custom
|
12 |
+
license: mit
|
13 |
+
language:
|
14 |
+
- en
|
15 |
+
datasets:
|
16 |
+
- ruslanmv/ai-medical-chatbot
|
17 |
+
- Locutusque/hercules-v5.0
|
18 |
+
---
|
19 |
+
# Medichat-V2-Llama3-8B
|
20 |
+
|
21 |
+
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
22 |
+
|
23 |
+
This model is particularly effective in **structuring** the unstructured clinical texts.
|
24 |
+
|
25 |
+
### Model Composition and Features:
|
26 |
+
1. **Base Model**: The foundation of this model is based on "refuelai/Llama-3-Refueled," which itself is a refined version of the Llama3-8B model, renowned for its instruction-following capabilities and adaptability across various domains.
|
27 |
+
|
28 |
+
2. **Merged Models**:
|
29 |
+
- **ruslanmv/ai-medical-model-32bit**: A model fine-tuned specifically for answering technical medical questions, providing a solid base of medical knowledge.
|
30 |
+
- **Locutusque/Llama-3-Hercules-5.0-8B**: Known for its ability to follow complex instructions and handle conversational interactions effectively, especially in scientific and technical contexts.
|
31 |
+
|
32 |
+
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [refuelai/Llama-3-Refueled](https://huggingface.co/refuelai/Llama-3-Refueled) as a base.
|
33 |
+
|
34 |
+
### Models Merged
|
35 |
+
|
36 |
+
The following models were included in the merge:
|
37 |
+
* [ruslanmv/ai-medical-model-32bit](https://huggingface.co/ruslanmv/ai-medical-model-32bit)
|
38 |
+
* [Locutusque/Llama-3-Hercules-5.0-8B](https://huggingface.co/Locutusque/Llama-3-Hercules-5.0-8B)
|
39 |
+
|
40 |
+
### Configuration
|
41 |
+
|
42 |
+
The following YAML configuration was used to produce this model:
|
43 |
+
|
44 |
+
```yaml
|
45 |
+
|
46 |
+
|
47 |
+
models:
|
48 |
+
- model: Locutusque/Llama-3-Hercules-5.0-8B
|
49 |
+
parameters:
|
50 |
+
weight: [0.25, 0.35, 0.45, 0.35, 0.25]
|
51 |
+
density: [0.1, 0.25, 0.5, 0.25, 0.1]
|
52 |
+
- model: refuelai/Llama-3-Refueled
|
53 |
+
- model: ruslanmv/ai-medical-model-32bit
|
54 |
+
parameters:
|
55 |
+
weight: [0.55, 0.45, 0.35, 0.45, 0.55]
|
56 |
+
density: [0.1, 0.25, 0.5, 0.25, 0.1]
|
57 |
+
merge_method: dare_ties
|
58 |
+
base_model: refuelai/Llama-3-Refueled
|
59 |
+
parameters:
|
60 |
+
int8_mask: true
|
61 |
+
dtype: bfloat16
|
62 |
+
|
63 |
+
```
|