Update README.md
Browse files
README.md
CHANGED
|
@@ -16,7 +16,7 @@ library_name: transformers
|
|
| 16 |
|
| 17 |
The **Mistral-7B-v0.1-Adapted** collection of large language models (LLMs), is a collection of adapted generative models in 7B (text in/text out), adapted models from **Mistral-7B-Base-v0.1**.
|
| 18 |
|
| 19 |
-
|
| 20 |
|
| 21 |
**Model developer:** SapienzaNLP, ISTI-CNR, ILC-CNR
|
| 22 |
|
|
@@ -24,7 +24,7 @@ The **Mistral-7B-v0.1-Adapted** collection of large language models (LLMs), is a
|
|
| 24 |
|
| 25 |
## Data used for the adaptation
|
| 26 |
|
| 27 |
-
The **Mistral-7B-v0.1-Adapted**
|
| 28 |
The data are extracted to be skewed toward Italian language with a ration of one over four. Extracting the first 9B tokens from Italian part of CulturaX and the first 3B tokens from English part of CulturaX.
|
| 29 |
|
| 30 |
|
|
@@ -32,7 +32,7 @@ The data are extracted to be skewed toward Italian language with a ration of one
|
|
| 32 |
|
| 33 |
You can run conversational inference using the Transformers pipeline abstraction or by leveraging the Auto classes with the generate() function.
|
| 34 |
|
| 35 |
-
Make sure to update your transformers installation via pip install --upgrade transformers
|
| 36 |
|
| 37 |
```python
|
| 38 |
import transformers
|
|
@@ -47,6 +47,8 @@ pipeline = transformers.pipeline(
|
|
| 47 |
pipeline("Cosa si può fare in una bella giornata di sole?")
|
| 48 |
```
|
| 49 |
|
|
|
|
|
|
|
| 50 |
## Citation
|
| 51 |
|
| 52 |
If you use any part of this work, please consider citing the paper as follows:
|
|
|
|
| 16 |
|
| 17 |
The **Mistral-7B-v0.1-Adapted** collection of large language models (LLMs), is a collection of adapted generative models in 7B (text in/text out), adapted models from **Mistral-7B-Base-v0.1**.
|
| 18 |
|
| 19 |
+
Mistral-v0.1-Italian-FVT is a continually trained Mistral model.
|
| 20 |
|
| 21 |
**Model developer:** SapienzaNLP, ISTI-CNR, ILC-CNR
|
| 22 |
|
|
|
|
| 24 |
|
| 25 |
## Data used for the adaptation
|
| 26 |
|
| 27 |
+
The **Mistral-7B-v0.1-Adapted** models are trained on a collection of Italian and English data extracted from [CulturaX](https://huggingface.co/datasets/uonlp/CulturaX).
|
| 28 |
The data are extracted to be skewed toward Italian language with a ration of one over four. Extracting the first 9B tokens from Italian part of CulturaX and the first 3B tokens from English part of CulturaX.
|
| 29 |
|
| 30 |
|
|
|
|
| 32 |
|
| 33 |
You can run conversational inference using the Transformers pipeline abstraction or by leveraging the Auto classes with the generate() function.
|
| 34 |
|
| 35 |
+
Make sure to update your transformers installation via `pip install --upgrade transformers`.
|
| 36 |
|
| 37 |
```python
|
| 38 |
import transformers
|
|
|
|
| 47 |
pipeline("Cosa si può fare in una bella giornata di sole?")
|
| 48 |
```
|
| 49 |
|
| 50 |
+
Code: https://github.com/SapienzaNLP/sava
|
| 51 |
+
|
| 52 |
## Citation
|
| 53 |
|
| 54 |
If you use any part of this work, please consider citing the paper as follows:
|