Update README.md
Browse files
README.md
CHANGED
@@ -1 +1,80 @@
|
|
1 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
datasets:
|
4 |
+
- Intel/orca_dpo_pairs
|
5 |
+
language:
|
6 |
+
- en
|
7 |
+
pipeline_tag: text-generation
|
8 |
+
---
|
9 |
+
|
10 |
+
|
11 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/63a259d0f30c46422789d38d/vO3iATjO8ulfcakTltE4k.png)
|
12 |
+
|
13 |
+
# Go Bruins V2 - A Fine-tuned Language Model
|
14 |
+
|
15 |
+
## Updates
|
16 |
+
|
17 |
+
## Overview
|
18 |
+
|
19 |
+
**Go Bruins-V2** is a language model fine-tuned on the rwitz/go-bruins architecture. It's designed to push the boundaries of NLP applications, offering unparalleled performance in generating human-like text.
|
20 |
+
|
21 |
+
## Model Details
|
22 |
+
|
23 |
+
- **Developer:** Ryan Witzman
|
24 |
+
- **Base Model:** [rwitz/go-bruins](https://huggingface.co/rwitz/go-bruins)
|
25 |
+
- **Fine-tuning Method:** Direct Preference Optimization (DPO)
|
26 |
+
- **Training Steps:** 642
|
27 |
+
- **Language:** English
|
28 |
+
- **License:** MIT
|
29 |
+
|
30 |
+
## Capabilities
|
31 |
+
|
32 |
+
Go Bruins excels in a variety of NLP tasks, including but not limited to:
|
33 |
+
- Text generation
|
34 |
+
- Language understanding
|
35 |
+
- Sentiment analysis
|
36 |
+
|
37 |
+
## Usage
|
38 |
+
|
39 |
+
**Warning:** This model may output NSFW or illegal content. Use with caution and at your own risk.
|
40 |
+
|
41 |
+
### For Direct Use:
|
42 |
+
|
43 |
+
```python
|
44 |
+
from transformers import pipeline
|
45 |
+
|
46 |
+
model_name = "rwitz/go-bruins-v2"
|
47 |
+
inference_pipeline = pipeline('text-generation', model=model_name)
|
48 |
+
|
49 |
+
input_text = "Your input text goes here"
|
50 |
+
output = inference_pipeline(input_text)
|
51 |
+
|
52 |
+
print(output)
|
53 |
+
```
|
54 |
+
|
55 |
+
|
56 |
+
### Not Recommended For:
|
57 |
+
|
58 |
+
- Illegal activities
|
59 |
+
- Harassment
|
60 |
+
- Professional advice or crisis situations
|
61 |
+
|
62 |
+
## Training and Evaluation
|
63 |
+
|
64 |
+
Trained on a dataset from [athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW](https://huggingface.co/datasets/athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW), Go Bruins V2 has shown promising improvements over its predecessor, Go Bruins.
|
65 |
+
|
66 |
+
# Evaluations
|
67 |
+
|
68 |
+
| Metric | Average | Arc Challenge | Hella Swag | MMLU | Truthful Q&A | Winogrande | GSM8k |
|
69 |
+
|---------------|---------|---------------|------------|------|--------------|------------|-------|
|
70 |
+
| **Score** | ??? | 69.97 | Coming Soon| Coming Soon | 60.37 | 81.61 | 69.90 |
|
71 |
+
|
72 |
+
Note: The original MMLU evaluation has been corrected to include 5-shot data rather than 1-shot data.
|
73 |
+
## Contact
|
74 |
+
|
75 |
+
For any inquiries or feedback, reach out to Ryan Witzman on Discord: `rwitz_`.
|
76 |
+
|
77 |
+
---
|
78 |
+
|
79 |
+
*This model card was created with care by Ryan Witzman.*
|
80 |
+
rewrite this model card for new version called go-bruins-v2 that is finetuned on dpo on the original go-bruins model on athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW
|