Hello. Thank you for releasing this amazing work. I am attempting to perform MultiModal Instruction Tuning on the pretrained-8b SEED-LLaMA model. However, I have not found any detailed hyperparameters in your paper or the GitHub README. Could you share a summary of the instruction tuning hyperparameters for SEED-LLaMA similar to those provided for the pretraining step?
Hello. Thank you for releasing this amazing work. I am attempting to perform MultiModal Instruction Tuning on the pretrained-8b SEED-LLaMA model. However, I have not found any detailed hyperparameters in your paper or the GitHub README. Could you share a summary of the instruction tuning hyperparameters for SEED-LLaMA similar to those provided for the pretraining step?