Generating synthetic data with LLM-s for training a downstream model

Example repository of generating and annotating synthetic texts and using them to train a downstream model. generating_and_annotating_example.ipynb contains an example workflow of how we generate the synthetic texts using our custom GPT-2 model and then use Azure API's LLM to annotate it, creating training data. training_annotated_texts_example.ipynb contains an example workflow of training the downstream model and evaluating it on the test data.