yuhuiaws / finetuning-and-deploying-llama-on-Sagemaker

Use the two different methods (deepspeed and SageMaker model parallelism library) to fine tune llama model on Sagemaker. Then deploy the fine tuned llama on Sagemaker with server side batch.
21Updated last year

Related projects

Alternatives and complementary repositories for finetuning-and-deploying-llama-on-Sagemaker