Update README.md
Browse files
README.md
CHANGED
|
@@ -239,4 +239,17 @@ The model is an instruction-tuned version of [**Meta-Llama-3-8b-instruct**](http
|
|
| 239 |
This model aims to be the **multilingual base-model** to further fine-tune in the Italian environment.
|
| 240 |
|
| 241 |
|
| 242 |
-
The 🌟**ANITA project**🌟 *(**A**dvanced **N**atural-based interaction for the **ITA**lian language)* wants to provide Italian NLP researchers with an improved model the for Italian Language 🇮🇹 use cases.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
This model aims to be the **multilingual base-model** to further fine-tune in the Italian environment.
|
| 240 |
|
| 241 |
|
| 242 |
+
The 🌟**ANITA project**🌟 *(**A**dvanced **N**atural-based interaction for the **ITA**lian language)* wants to provide Italian NLP researchers with an improved model the for Italian Language 🇮🇹 use cases.
|
| 243 |
+
|
| 244 |
+
<hr>
|
| 245 |
+
|
| 246 |
+
**Model developers** Marco Polignano - University of Bari Aldo Moro, Italy
|
| 247 |
+
|
| 248 |
+
**Variations** The model release has been **supervised fine-tuning (SFT)** using **QLoRA** in the 4bit version, on a long list of instruction-based datasets. **ORPO** approach over the *mlabonne/orpo-dpo-mix-40k* dataset is used to align with human preferences for helpfulness and safety.
|
| 249 |
+
|
| 250 |
+
**Input** Models input text only.
|
| 251 |
+
|
| 252 |
+
**Output** Models generate text and code only.
|
| 253 |
+
|
| 254 |
+
**Model Architecture** *Llama 3 architecture*.
|
| 255 |
+
|