I think organization of wells does not matter as long as you keep it serial and label or note the location of respective design down. For your negative control perhaps is your vehicle control, essentially your solvent used to dissolve the compounds, which is DMSO in many cases. Dilute it to the same concentration as you would for your compound. You can follow the answer of mine from below answer or the one by the other researcher. https://www.researchgate.net/post/How_to_achieve_a_final_concentration_of_01_DMSO_for_MTT_Assay

One thing I would like you to consider before performing the assay is to optimize the cell density. If you are not familiar with the cells, you will need to establish growth curves. For example, you can seed the cells in 1000, 2000, 3000..., 10000 cells per well in triplicate and evaluate their growth on day 0, day 1, day 2, and day 3 (if day 3 is your endpoint later on) using MTT. Plot the growth using linear graph. Select the density that allows you to observe linear growth over 3 days (gradient slope nears 1). Linearity ensures that the cell density you use does not hit the linearity upper limit of the assay for your untreated control, i.e., MTT absorbance peaks at around 2.5-3.0. For certain cells, seeding too many cells can lead to a more acidic medium and when MTT is added, it can lead to a spectrum shift problem.

One more hint: gallic acid is a polyphenolic compound that will spontaneously reduce your MTT reagent even without the presence of cells. So, you might want to reconsider of using it.