Update README.md

README.md

@@ -9,7 +9,7 @@ metrics:
 - accuracy
 pipeline_tag: text-generation
 model-index:
-- name: Latxa-Llama-3.1-70B-Instruct
+- name: Latxa-Llama-3.1-70B-Instruct-FP8
   results:
   - task:
       type: multiple-choice
@@ -68,7 +68,7 @@ model-index:
 
 license: llama3.1
 base_model:
-- meta-llama/Llama-3.1-70B-Instruct
+- HiTZ/Latxa-Llama-3.1-70B-Instruct
 
 co2_eq_emissions:
   emissions: 1900800
@@ -80,12 +80,15 @@ co2_eq_emissions:
 quantized_by: HiTZ
 ---
 
-# Model Card for HiTZ/Latxa-Llama-3.1-70B-Instruct
+# Model Card for HiTZ/Latxa-Llama-3.1-70B-Instruct-FP8
 
 <p align="center">
   <img src="https://github.com/hitz-zentroa/latxa/blob/b9aa705f60ee2cc03c9ed62fda82a685abb31b07/assets/latxa_round.png?raw=true" style="height: 350px;">
 </p>
 
+> [!IMPORTANT]
+> This is a FP8 quantized version of the original Latxa 3.1 70B Instruct.
+
 We introduce Latxa 3.1 70B Instruct, an instructed version of [Latxa](https://aclanthology.org/2024.acl-long.799/). This new Latxa is based on Llama-3.1 (Instruct), which we trained on our Basque corpus (Etxaniz et al., 2024) comprising 4.3M documents and 4.2B tokens using language adaptation techniques (paper in preparation). 
 > [!WARNING]
 > DISCLAIMER
@@ -123,7 +126,7 @@ Use the code below to get started with the model.
 ```python
 from transformers import pipeline
 
-pipe = pipeline('text-generation', model='HiTZ/Latxa-Llama-3.1-70B-Instruct')
+pipe = pipeline('text-generation', model='HiTZ/Latxa-Llama-3.1-70B-Instruct-FP8')
 
 messages = [
 	{'role': 'user', 'content': 'Kaixo!'},