Alexandre-Numind
commited on
Commit
•
9e926fd
1
Parent(s):
cbb8243
Update README.md
Browse files
README.md
CHANGED
@@ -3,13 +3,34 @@ license: mit
|
|
3 |
language:
|
4 |
- en
|
5 |
widget:
|
6 |
-
- text: '
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
---
|
14 |
# Structure Extraction Model by NuMind 🔥
|
15 |
|
|
|
3 |
language:
|
4 |
- en
|
5 |
widget:
|
6 |
+
- text: '<|input|>
|
7 |
+
### Template:
|
8 |
+
{
|
9 |
+
"Model": {
|
10 |
+
"Name": "",
|
11 |
+
"Number of parameters": "",
|
12 |
+
"Number of max token": "",
|
13 |
+
"Architecture": []
|
14 |
+
},
|
15 |
+
"Usage": {
|
16 |
+
"Use case": [],
|
17 |
+
"Licence": ""
|
18 |
+
}
|
19 |
+
}
|
20 |
+
### Text:
|
21 |
+
We introduce Mistral 7B, a 7–billion-parameter language model engineered for
|
22 |
+
superior performance and efficiency. Mistral 7B outperforms the best open 13B
|
23 |
+
model (Llama 2) across all evaluated benchmarks, and the best released 34B
|
24 |
+
model (Llama 1) in reasoning, mathematics, and code generation. Our model
|
25 |
+
leverages grouped-query attention (GQA) for faster inference, coupled with sliding
|
26 |
+
window attention (SWA) to effectively handle sequences of arbitrary length with a
|
27 |
+
reduced inference cost. We also provide a model fine-tuned to follow instructions,
|
28 |
+
Mistral 7B – Instruct, that surpasses Llama 2 13B – chat model both on human and
|
29 |
+
automated benchmarks. Our models are released under the Apache 2.0 license.
|
30 |
+
Code https://github.com/mistralai/mistral-src
|
31 |
+
Webpage https://mistral.ai/news/announcing-mistral-7b/
|
32 |
+
<|output|>
|
33 |
+
'
|
34 |
---
|
35 |
# Structure Extraction Model by NuMind 🔥
|
36 |
|