You can see that even though Power Automate show you the file name and table name, when you actually run the flow and check the input of the update row action, you can see that it isn't really "english". It is "ID" of some sort, File ID and Table ID, and this is what "selection" does in the action, it load the english for you to select, but use the "ID" in the "coding".
That is why directly "typing" in the field file and table name doesn't work. You can get the file ID from the create file action that you use to create the excel file within the flow. Then get the Table ID from the Create table action's output Name field.
Now, IF your excel file is a "COPY" of an existing template file, while the FILE ID is still going to be different, the Table ID remains the same, so you can directly enter the Table ID from the existing template file.
Now, the issue you are going to have is, when you create the Update a row action, if you are trying to use the File ID and Table ID, since the file doesn't exists yet, Update a row isn't going to know what to "do" (can't display the columns for you to fill in content with on a table that the flow can't load column info on). This goes back to my original suggestion, that you need to use an existing excel file that WOULD be the same structure as the future dynamically created excel file, fill in everything as you want it. THEN lastly change the File and Table field to the dynamic content of the ID for each.