Update README.md
Browse files
README.md
CHANGED
@@ -1,12 +1,30 @@
|
|
1 |
---
|
2 |
-
library_name: transformers
|
3 |
-
tags:
|
4 |
-
- text-generation-inference
|
5 |
language:
|
6 |
-
-
|
7 |
pipeline_tag: text-generation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
---
|
9 |
|
10 |
# Model Card for Model ID
|
11 |
|
12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
|
|
|
|
|
|
2 |
language:
|
3 |
+
- multilingual
|
4 |
pipeline_tag: text-generation
|
5 |
+
tags:
|
6 |
+
- nlp
|
7 |
+
- code
|
8 |
+
inference:
|
9 |
+
parameters:
|
10 |
+
temperature: 0.7
|
11 |
+
widget:
|
12 |
+
- messages:
|
13 |
+
- role: user
|
14 |
+
content: Can you provide ways to prepare ramen dishes?
|
15 |
---
|
16 |
|
17 |
# Model Card for Model ID
|
18 |
|
19 |
+
## Overview:
|
20 |
+
raaec/Phi-3-mini-4k-instruct-shy is a language model that exhibits introverted behavior, using orthogonalization to ablate extroverted tendencies.
|
21 |
+
|
22 |
+
## !! When using the model make sure to use `tokenizer = AutoTokenizer.from_pretrained("microsft/Phi-3-mini-4k-instruct")`
|
23 |
+
|
24 |
+
## Methodology:
|
25 |
+
|
26 |
+
### Base Model: microsoft/Phi-3-medium-4k-instruct
|
27 |
+
### Orthogonalization: Applied to ablate extroverted behaviors.
|
28 |
+
### Ablation Technique: Utilizes minimal data to inhibit refusal and enhance introversion without altering other behaviors.
|
29 |
+
### Purpose:
|
30 |
+
This model is ideal for applications requiring concise, reserved responses.(sometimes a bit funny)
|