Training steps refer to the number of steps the training program takes as it learns the new concepts you have provided in your dataset.
Although it may seem as though more training steps should be better, this is not always the case. Training steps should be relative to the size of your dataset and your learning rate, and this can vary based on the Foundational Model being trained on or the type of Model you want to train.
Typically, you will want a lower learning rate for a large dataset with lots of training steps, and a higher learning rate for a smaller dataset with fewer training steps - additionally, when you use the text encoder you may find that training steps do not need to be as high.
There are a lot of different methods of using training steps and learning rates together to create good results. So long as your training steps are above 500 and not so many that they over train your model, it simply takes testing what works best for you and your dataset.