mixer volume fader

Hi, I have a question that most people may find stupid but please forgive my ignorance. I have been making mediocre recordings at home unprofessionally and I’m fairly not that good at mixing because I’m uneducated. What I want to know is the meaning of the numbers that appear below the mixer’s volume fader of each track.(like -6.2 for example) I mean the ones that change from time to time during playing. I guess it means they are too low or high but the question is what to do with them. Please explain to me in a simple manner how to adjust the faders accordingly.
 

As John points out, the numbers below the faders don't "do" anything. They're a readout that shows how much a fader is boosting or attenuating a signal. In the console view, the faders work in tandem with the meters to their right to help you set levels.
 
The reason why levels are important is that if the levels are too high, then you can overload the system, which creates distortion. If the meters go so high they turn red, then the levels are too high and you need to bring down the fader. If the levels are too low, then that may lead to noise creeping in because the signal-to-noise ratio is too low - in other words, there's less signal, and more noise. In that case you need to turn up the fader.
 
In mixers and audio in general, there's a concept called "unity gain." This means that something (like an amplifier) neither amplifies nor attenuates - the signal at the amplifier's output is the same level as the level at the amplifier's input. 
 
When the fader readout shows 0, that means it's set for unity gain. So, if the peak level on a track is -3 dB, and the fader is set for 0, then the peak reading on the meter will be -3. Now, suppose you set the fader so the fader readout shows -4. This means it's attenuating the signal (i.e., reducing its level) by 4 dB. Now that peak track level of -3 dB shows up as -7 dB on the meter because it is 4 dB lower.
 
If you set the fader so its readout is 2.0, it's boosting the signal from the track by 2 dB. When you play back the track with its peak level of -3 dB, the meter will show the peak level as -1.0 instead of -3 dB because that original signal coming from the track has been boosted by 2 dB, thanks to turning up the fader.
 
When mixing, it's good practice to leave the master fader at 0 dB, then adjust the individual channel faders so that the master output meter never goes high enough to light red.
 
Those are the basics...does that make sense?

It's not that you "should" set a higher or lower level, as that decision is more an artistic than a technical decision.
 
The most important point is this: Mixing is about achieving a good-sounding balance among instruments, so some will be higher in level, and some will be lower. For example, a tambourine track might have the fader set to -15 because you don't want it overpowering the other instruments, while the lead instruments will have their faders set to 0 because you want them to be the loudest. Meanwhile a rhythm guitar might have the fader set to -10 but the lead guitar is -2. You don't set the faders based on their numbers, you set them based on what sounds right and the numbers simply tell you where the fader is set.
 
So the next question is...if it doesn't matter all that much what the number says, and all that matters is the level, why do we need the numbers at all?
 
In a way, we don't. But as you get into more advanced mixing techniques, you might listen to a track and think "y'know, I think this needs to come down a couple dB." So you adjust "by the numbers" to bring down the fader by 2 dB. However, upon listening you think, "nah, that's the wrong move, I want to get it back to where it was." By moving the fader up 2 dB, you'll indeed be back to where you were instead of having to adjust it blind, listen, and see whether you got it right or not.

As long as the master doesn't go into the red, don't worry about it.
 
Also FWIW most commercially-recorded music uses limiting or compression when mastering to make the overall level louder. This is an instance where you might find doing test masters with LANDR helpful to understand how mastering affects the overall sound.

Gainer
No, I said the master goes red already

Well, now I'm confused because you said "Thru the song none of them outlevels the peak limit, no red peaks. But the master fader shows +3 or +4 dB after a few seconds. How can I avoid this? The mixdown is already quiter than I want."
 
If the fader readout shows +3 or +4, that's no problem unless the master meter itself goes into the red. If it does, then you need to reduce either the master fader or the individual channel faders, and if the final level isn't loud enough, use dynamics processing like compression or limiting to increase the average level (as opposed to the peak level). This will make the mix sound louder.

I'm watching tutorials for compressors, hopefully it might solve my problem. Thanks for the help so far.

John
Keep in mind if you are out putting your tracks to a buss that becomes cumulative.  They add together and will be louder than the individual tracks soloed. 

Just to expand on that to demonstrate the idea. If you had 2 identical tracks. Both were showing zero on the their faders, combined they are going to be twice as loud on the master fader. This would be +3db. Now pan one fully right and one fully left, and now the master fader be back at zero.
 

Gainer
I don't have two identical tracks, just one guitar, bass and vst drums in this relevant project. Though I tried duplicating the guitar track and panning them right/left but it didn't help at all. The master peak still goes abaut 3 or 4 dB in red. BUT there is no unwanted distortion in the mixdown or playback.

It would probably benefit you a great deal to do a search for and study "Gain Staging".

That sounds useful gonna try that, thanks

bitflipper
The idea is to mix for balance, and not worry about how loud the mix is.

 ^^^^^this