I’m a Windows guy so someone else can help with the settings -
Your hardware (soundcard or interface) has to support 24-bits. When you record (or open an audio file) Audacity temporarily converts everything to 32-bit floating-point for “processing”. You can select 24-bits again when you export.
Noise comes from your room (if you’re recording with a microphone) and there is some electrical noise from the electronics. -60dB is 1/1000th of 0dB and if you are recording with a microphone it’s VERY difficult to achieve if you don’t have a soundproof studio. -60dB is the audiobook standard and many people are able to achieve that with some noise reduction, but that’s not “raw”.
(If you are recording an + signal like from a CD player there is no acoustic room noise and -60dB usually isn’t a problem.)
What you’re looking for is a strong signal-to-noise ratio. i.e. Speak/sing with a strong confident voice, fairly-close to the mic, etc. With a strong signal you you won’t have to boost the levels which would also any background noise.
RAW audio file
They probably don’t want want a raw formatted file. They probably want an “unprocessed” WAV file (no noise reduction, etc.) Usually a “RAW” file means a WAV file without the WAV file header and then you can’t open it without knowing the format details (bit-depth, sample rate, etc.) and it’s just a royal pain to deal with.
The measurement is in “dB RMS”.
-60 dB RMS is generally considered to be a good level for the noise floor, but it requires a very quiet room to achieve this level while also achieving a good voice level.