Update README.md
Browse files
README.md
CHANGED
@@ -91,13 +91,13 @@ pipeline_tag: text-generation
|
|
91 |
library_name: transformers
|
92 |
---
|
93 |
|
94 |
-
#
|
95 |
|
96 |
## Table of Contents
|
97 |
<details>
|
98 |
<summary>Click to expand</summary>
|
99 |
|
100 |
-
- [
|
101 |
- [Table of Contents](#table-of-contents)
|
102 |
- [Model description](#model-description)
|
103 |
- [Intended uses and limitations](#intended-uses-and-limitations)
|
@@ -117,12 +117,13 @@ library_name: transformers
|
|
117 |
|
118 |
## Model description
|
119 |
|
120 |
-
**
|
121 |
It is the result of a continual pretraining of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) with a multilingual corpus of almost 20B tokens, with an emphasis of Galician texts.
|
122 |
|
|
|
123 |
## Intended uses and limitations
|
124 |
|
125 |
-
The **
|
126 |
It can perform text-generation tasks and be fine-tuned for specific scenarios.
|
127 |
|
128 |
## How to use
|
@@ -132,7 +133,7 @@ from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
|
|
132 |
|
133 |
input_text = "Hoxe fai un bo día. O sol "
|
134 |
|
135 |
-
model_id = "proxectonos/
|
136 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
137 |
model = AutoModelForCausalLM.from_pretrained(model_id)
|
138 |
generator = pipeline(
|
|
|
91 |
library_name: transformers
|
92 |
---
|
93 |
|
94 |
+
# Llama-3.1-Carballo
|
95 |
|
96 |
## Table of Contents
|
97 |
<details>
|
98 |
<summary>Click to expand</summary>
|
99 |
|
100 |
+
- [Llama-3.1-Carballo](#llama31-carballo)
|
101 |
- [Table of Contents](#table-of-contents)
|
102 |
- [Model description](#model-description)
|
103 |
- [Intended uses and limitations](#intended-uses-and-limitations)
|
|
|
117 |
|
118 |
## Model description
|
119 |
|
120 |
+
**Llama-3.1-Carballo** is a 8B-parameter transformer-based causal language model for Galician, Portuguese, Spanish, Catalan and English.
|
121 |
It is the result of a continual pretraining of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) with a multilingual corpus of almost 20B tokens, with an emphasis of Galician texts.
|
122 |
|
123 |
+
This model is part of the **Carballo familily**, a family of LLMs specialized in Galician. Smaller models can be founded [here](https://huggingface.co/collections/proxectonos/text-models-65d49fa54e358ce02a9699c8)
|
124 |
## Intended uses and limitations
|
125 |
|
126 |
+
The **Llama-3.1-Carballo** model is ready-to-use only for causal language modeling.
|
127 |
It can perform text-generation tasks and be fine-tuned for specific scenarios.
|
128 |
|
129 |
## How to use
|
|
|
133 |
|
134 |
input_text = "Hoxe fai un bo día. O sol "
|
135 |
|
136 |
+
model_id = "proxectonos/Llama-3.1-Carballo"
|
137 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
138 |
model = AutoModelForCausalLM.from_pretrained(model_id)
|
139 |
generator = pipeline(
|