Update README.md

README.md

@@ -19,31 +19,18 @@ tags:
 This model is intended for researchers, developers, and enthusiasts who are interested in understanding the performance of the Llama 2 model on Vietnamese. It can be used for generating Vietnamese text based on given instructions or for any other task that requires a Vietnamese language model.
 
 ## Limitations
-Data Size: The model was fine-tuned on a relatively small dataset of 20,000 instruction samples, which might not capture the full complexity and nuances of the Vietnamese language.
-Preliminary Model: This is an initial experiment with the Llama 2 architecture on Vietnamese. More refined versions and evaluations will be available soon.
-Performance
+- Data Size: The model was fine-tuned on a relatively small dataset of 20,000 instruction samples, which might not capture the full complexity and nuances of the Vietnamese language.
+- Preliminary Model: This is an initial experiment with the Llama 2 architecture on Vietnamese. More refined versions and evaluations will be available soon.
+- Performance:
 Specific performance metrics on this fine-tuned model will be provided in the upcoming comprehensive evaluation.
 
 ## Ethical Considerations
-Bias and Fairness: Like any other machine learning model, there is a possibility that this model might reproduce or amplify biases present in the training data.
-Use in Critical Systems: As this is a preliminary model, it is recommended not to use it for mission-critical applications without proper validation.
-Fine-tuning Data
+- Bias and Fairness: Like any other machine learning model, there is a possibility that this model might reproduce or amplify biases present in the training data.
+- Use in Critical Systems: As this is a preliminary model, it is recommended not to use it for mission-critical applications without proper validation.
+- Fine-tuning Data:
 The model was fine-tuned on a custom dataset of 20,000 instruction samples in Vietnamese. More details about the composition and source of this dataset will be provided in the detailed evaluation report.
 
-## Usage
-To use this model via the Hugging Face API:
 
-```python
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
-
-tokenizer = AutoTokenizer.from_pretrained("ngoantech/Llama-2-7b-vi-sample")
-model = AutoModelForSeq2SeqLM.from_pretrained("ngoantech/Llama-2-7b-vi-sample")
-
-inputs = tokenizer.encode("YOUR INSTRUCTION HERE", return_tensors="pt")
-outputs = model.generate(inputs)
-decoded = tokenizer.decode(outputs[0], skip_special_tokens=True)
-print(decoded)
-```
 
 ## Credits
 I would like to express our gratitude to the creators of the Llama 2 architecture and the Hugging Face community for their tools and resources.