|
--- |
|
language: |
|
- en |
|
- it |
|
license: apache-2.0 |
|
tags: |
|
- Conversational |
|
base_model: mistralai/Mistral-7B-v0.1 |
|
datasets: |
|
- Antonio88/TaliStran-DataSet |
|
--- |
|
|
|
|
|
 |
|
|
|
|
|
|
|
TaliML-7B-ITA-V.1.0.FINAL: The Italian Language Model |
|
|
|
TaliStran is a model developed leveraging the power of Mistral 7B. This model has been specifically trained to understand and interact in Italian, providing coherent responses to questions asked in this language. |
|
|
|
Main Features: |
|
Training: TaliStran was trained on a selected small dataset of about 500 questions and answers, ensuring targeted preparation for understanding and generating text in Italian. |
|
|
|
Training Efficiency: The training process was carried out using Google Colab, utilizing an A100 graphics card, which allowed the preparation to be completed efficiently and with limited resources. |
|
|
|
Model in Test: Continuous improvements will be made. |
|
|
|
Please note that the responses obtained from the model should not be considered as absolute truths. |
|
|
