Something wrong?

Sounds like a relatively normal water weight loss - particularly if you're taking in fewer carbs than you used to. As Kali31337 suggested, check your results after a few more weeks but I think you'll find that you start leveling off once your body acclimates to the new "normal."

Sounds very normal, certainly nothing to be worried about at all. I lost 4-5 pounds a week for the first 4 weeks, then 3 lbs per week for two more weeks, and then in week 6 everything slowed down to what it should've been all along - 2 pounds per week. In total, in the first month I lost 20 pounds, I figure 8 of those were "real" pounds (pounds of fat). The rest was water. Countless other people here have experienced the same thing. Your body stores a lot of water to balance the sodium and, especially, carbs. When you diet, sodium usually decreases but carbs almost always plunge, due to just having less food--and if you're eating low carb on the diet, even less carbs--and your body gradually sheds the water to align with your new intake levels.

The main reason not to worry, aside from the fact that water fluctuations are meaningless, is that in a few months you will look back wistfully on this first month, when weight loss seemed so easy. Once the rate of fat loss drops to your target one pound per week, it will seem slow, so enjoy the diet honeymoon period while it lasts! It will never be this easy again.

Also, if you stop dieting, all but probably 2 - 2.5 pounds of your loss so far will come back, very quickly. Because it's just water, and if you increase carbs or sodium, your body will rebalance and hold onto more water again.

Probably a disproportionate reduction in water weight (and possibly average digestive system contents) as others have suggested.

If you're a premenopausal woman, and aren't deeply familiar with your normal hormonal weight fluctuations, you might also have started weight loss at a relative high point, and dropped into a relative low point now, in your normal cycle . . . on top of fat loss and the common water weight drop, making the loss look even bigger. (For others, the opposite can happen, and make them think they aren't losing fat, when they really are!)

For some women, cycle fluctuations are small, but for others they can be 5 pounds or more; and they can happen at different points in the cycle for different women. With some experience (with daily weighing, if tolerable), women can figure out their personal patterns (how much water weight, when in cycle).

The one thing I really wanted to add is this: It's important to keep monitoring for 4-6 weeks, and (if premenopausal) long enough to compare weights at the same point in two or more different menstrual cycles, to get an idea what's going on, on average. If you start feeling weak or fatigued along the way (for otherwise unexplained reasons), and have continued dropping weight rapidly to that point, then I'd suggest adding calories for a while, just as a bet-hedge against unhealthfully fast loss. (A couple weeks is probably too soon to be very worried, as long as you're feeling fine so far.)

Most people, when logging reasonably accurately and setting up their profile accurately as well, are going to get average results pretty close to what MFP (or another "calculator", including fitness tracker) predicts (monitoring over at least 4-6 weeks, maybe ignoring the first couple weeks if unusual).

MFP is giving you a calorie goal based on statistical estimates: Essentially, it thinks you'll burn the same number of calories daily as people of your same demographic details (age, weight, activity level, etc.), on average. But in real life, individual people fall in a statistical distribution (some kind of bell curve type thingie).

In this particular distribution, it's kind of a tall, narrow curve (relatively small standard deviation, in statistical terms). Most people will therefore be close to the average, and find MFP (or other calorie "calculator"/tracker) estimates pretty close. But a few will find them a bit further off, either high or low. A very, very few will find them to be quite far off, but that's rare. That's the nature of statistical estimates.

MFP, TDEE calculators, and fitness trackers are all based on this type of statistical estimate, spitting out averages for calories. (In the case of the trackers, the estimates are a bit more personalized to activity level, but they're still just estimates, despite what some believe.)

Losing weight too fast can be risky, as you clearly realize. And it can happen alongside feeling good and not hungry . . . until suddenly, that feeling changes. That's what happened to me. MFP underestimates my calorie needs by several hundred calories daily. (So does my fitness tracker.) That's very unusual, but it can happen.

Your own experience, carefully logging eating/activity/weight, over a period of time, is the most accurate indicator.