Fine-tuning Llama for customer service QA
You work at a company that builds customer service chatbots. Your team uses the Llama models in your customer service bot, and you want to improve the model by fine-tuning on a question-answering dataset related to customer service. To ensure the best performance out of these models, your team will fine-tune a Llama model for this task using the bitext dataset.
The training script is already almost complete, the only thing missing is the final step where you bring together the model, tokenizer, training dataset, and training arguments and start training.
This exercise is part of the course
Fine-Tuning with Llama 3
Exercise instructions
- Import the class that lets you conduct supervised fine-tuning from its library.
- Instantiate the class used for supervised fine-tuning by passing the
model,tokenizer,dataset, andtraining_arguments. - Run the instance method to start fine-tuning your model.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Import the supervised fine-tuning class
from ____ import ____
# Instantiate fine-tuning class
trainer = ____(
# Pass necessary arguments
____=____,
____=____,
____=____,
____=____,
)
# Start training
trainer.____()