how do I change the prompt loss weight when fine tuning through azure open ai?

Kayla Farivar 50 Reputation points
2024-07-17T20:53:16.1033333+00:00

Specifically through python? Also what is the default value for prompt loss weight. I believe the other hyperparamaters you can change are learning_rate_multiplier, batch_size, and n_epochs, are there any others I can use to better adjust my model?

Azure OpenAI Service
Azure OpenAI Service
An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
2,829 questions

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. YutongTie-MSFT 50,191 Reputation points
    2024-07-18T05:53:20.71+00:00

    Hello Kayla,

    Thanks for reaching out to us, the hyperparameters current supported for Python is as you mentioned, there is no other hyperparameters we can leverage in the fine-tuning step.

    User's image

    But you can analyze the result to see how to change the hyperparameters better

    User's image

    Look for your loss to decrease over time, and your accuracy to increase. If you see a divergence between your training and validation data, that may indicate that you are overfitting. Try training with fewer epochs, or a smaller learning rate multiplier.

    I hope this helps. Let us know if you need more information.

    Regards,

    Yutong

    -Please kindly accept the answer if you feel helpful to support the community, thanks a lot.

    0 comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.