Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,51 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
datasets:
|
4 |
+
- CreitinGameplays/merged-data-v2
|
5 |
+
base_model:
|
6 |
+
- mistralai/Mistral-7B-v0.3
|
7 |
+
- mistralai/Mistral-7B-Instruct-v0.3
|
8 |
+
language:
|
9 |
+
- en
|
10 |
+
---
|
11 |
+
# **ConvAI-9b v2: A Conversational AI Model**
|
12 |
+

|
13 |
+
## **1. Model Details**
|
14 |
+
|
15 |
+
* **Model Name:** ConvAI-9b v2
|
16 |
+
* **Authors:** CreitinGameplays
|
17 |
+
* **Date:** May 29th, 2024
|
18 |
+
|
19 |
+
## **2. Model Description**
|
20 |
+
|
21 |
+
ConvAI-9b v2 is a fine-tuned conversational AI model with 9 billion parameters. It is based on the following models:
|
22 |
+
|
23 |
+
* **Base Model:** [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3)
|
24 |
+
* **Merged Model:** [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3)
|
25 |
+
|
26 |
+
## **3. Training Data**
|
27 |
+
|
28 |
+
The model was fine-tuned on a custom dataset of conversations between an AI assistant and a user. The dataset format followed a specific structure:
|
29 |
+
|
30 |
+
```
|
31 |
+
<|system|> (system prompt, e.g.: You are a helpful AI language model called ChatGPT, your goal is helping users with their questions) </s> <|user|> (user prompt) </s>
|
32 |
+
```
|
33 |
+
|
34 |
+
|
35 |
+
## **4. Intended Uses**
|
36 |
+
|
37 |
+
ConvAI-9b is intended for use in conversational AI applications, such as:
|
38 |
+
|
39 |
+
* Chatbots
|
40 |
+
* Virtual assistants
|
41 |
+
* Interactive storytelling
|
42 |
+
* Educational tools
|
43 |
+
|
44 |
+
## **5. Limitations**
|
45 |
+
|
46 |
+
* Like any other language model, ConvAI-9b v2 may generate incorrect or misleading responses.
|
47 |
+
* It may exhibit biases present in the training data.
|
48 |
+
* The model's performance can be affected by the quality and format of the input text.
|
49 |
+
|
50 |
+
## **6. Evaluation**
|
51 |
+
~ soon
|