more text
Browse files
README.md
CHANGED
@@ -7,12 +7,35 @@ license: apache-2.0
|
|
7 |
|
8 |
# Model Yaml
|
9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
```
|
11 |
-
|
12 |
-
|
|
|
13 |
```
|
14 |
|
15 |
-
|
16 |
|
17 |
-
|
18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
7 |
|
8 |
# Model Yaml
|
9 |
|
10 |
+
The Mistral-7B-v0.1 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters.
|
11 |
+
Mistral-7B-v0.1 outperforms Llama 2 13B on all benchmarks we tested.
|
12 |
+
|
13 |
+
For full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/announcing-mistral-7b/).
|
14 |
+
|
15 |
+
## Model Architecture
|
16 |
+
|
17 |
+
Mistral-7B-v0.1 is a transformer model, with the following architecture choices:
|
18 |
+
- Grouped-Query Attention
|
19 |
+
- Sliding-Window Attention
|
20 |
+
- Byte-fallback BPE tokenizer
|
21 |
+
|
22 |
+
## Troubleshooting
|
23 |
+
|
24 |
+
- If you see the following error:
|
25 |
+
```
|
26 |
+
KeyError: 'mistral'
|
27 |
```
|
28 |
+
- Or:
|
29 |
+
```
|
30 |
+
NotImplementedError: Cannot copy out of meta tensor; no data!
|
31 |
```
|
32 |
|
33 |
+
Ensure you are utilizing a stable version of Transformers, 4.34.0 or newer.
|
34 |
|
35 |
+
## Notice
|
36 |
+
|
37 |
+
Mistral 7B is a pretrained base model and therefore does not have any moderation mechanisms.
|
38 |
+
|
39 |
+
## The Mistral AI Team
|
40 |
+
|
41 |
+
Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
|