dvres commited on
Commit
c8a7b97
·
verified ·
1 Parent(s): e5ce8a4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -3
README.md CHANGED
@@ -1,3 +1,43 @@
1
- ---
2
- license: gemma
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: gemma
3
+ language:
4
+ - sl
5
+ - en
6
+ base_model:
7
+ - cjvt/GaMS-9B-Instruct
8
+ pipeline_tag: text-generation
9
+ ---
10
+
11
+ # Model Card for GaMS-9B-SFT-Translator
12
+
13
+ GaMS-9B-SFT-Trasnlator represents the version of GaMS-9B-Instruct further trained using SFT on approximately 130k English to Slovene translation pairs.
14
+
15
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/652d40a78fa1fbb0aae165bb/94gX0PG8zRB_Zg31K2y_i.png)
16
+
17
+ ## Acknowledgment
18
+
19
+ The model was developed within the [PoVeJMo](https://www.cjvt.si/povejmo/en/project/) research program (Adaptive Natural Language Processing with Large Language Models), particularly within the research project titled SloLLaMai -- Open-access computationally efficient models for Slovenian. The program is funded within the Recovery and Resilience Plan by the Slovenian Research and Innovation Agency (ARIS) and NextGenerationEU. The authors also acknowledge the financial support from the Slovenian Research and Innovation Agency (research core funding No. P6-0411 -- Language Resources and Technologies for Slovene).
20
+
21
+ ## Usage
22
+
23
+ The model can be run through `pipeline` API using the following code:
24
+
25
+ ```python
26
+ from transformers import pipeline
27
+
28
+ model_id = "GaMS-Beta/GaMS-9B-SFT-Translator"
29
+
30
+ pline = pipeline(
31
+ "text-generation",
32
+ model=model_id,
33
+ device_map="cuda" # replace with "mps" to run on a Mac device
34
+ )
35
+
36
+ # Example of response generation
37
+ text_to_translate = "A fast brown fox jumped over the lazy dog."
38
+ message = [{"role": "user", "content": f"Prevedi naslednje besedilo v slovenščino.\n{text_to_translate}"}]
39
+ response = pline(message, max_new_tokens=512)
40
+ print("Slovene translation:", response[0]["generated_text"][-1]["content"])
41
+ ```
42
+
43
+ For multi GPU inference set the `device_map` to `auto`: