When I was a first year grad student, I never listened to advice that any of the older (read "disgruntled") students or postdocs would give me about grad school. If I had to go back and do graduate school all over again, I still wouldn't listen to that advice.
That's because no one actually gave me advice that was either actionable or even enlightening. What they really gave were either depressing views of the future like "one day you'll be just as bitter and demented as I am" or thinly-veiled attempts to cut down on their future competition such as "if I were you I'd quit grad school now before you waste too much of your time."
Even though it's been several years since I've graduated, grad school still literally gives me nightmares (like the one in which I couldn't stop kicking someone in the head). But now I can look back and occasionally have brief periods of calmness during which I can synthesize some of the lessons I learned. There are things I wish that someone had told me before I started grad school, and that either I was smart enough to keep on my whiteboard at all times, or lucky enough to have someone repeat to me over and over.
(NOTE: My program was in molecular biology, so some of the advice below is specific for that type of program, where lab work is the focus.)
1. Working hard is not the same thing as being perceived as working hard.
Not everyone falls into the trap of thinking that they need to see you at your bench for you to be actually working, but unfortunately many people do.
You may work on a project/ in a sub-field that requires many hours of lab work away from your main bench (for example, in the animal facility, in the microscope room, at another lab borrowing their equipment, etc.). Or you may prefer to analyze data and do big-picture thinking away from the lab, where you can work unperturbed by lab imps and the like.
It may therefore be inherently obvious to you that someone who is not always physically present at their desk or lab bench may in fact be working very hard. It is not, however, inherently obvious to people who have limited cognitive skills, and these are often the people who are judging you the most. (See point 9 below.)
(Near the end of my graduate school time, for example, I had to come in around 7 am every day to take a time point and / or run some experiments with equipment borrowed from another lab. The lender himself needed the equipment back by 10 am. So from 7 am to 10 am I often worked in our lab's mouse facility, and then around 10 am I would come back to my desk and bench in the main lab. I was then taken aside from a labmate who told me that not getting to work until 10 am was contributing the perception of me as a lazy person.)
2. Being perceived as working hard may be more important than you'd like.
Although the perception of how hard you work won't create data for you, it may influence a lot of things that ultimately affect your success in graduate school and beyond. Your advisor's perception of how hard you work, for example, can influence how they perceive the value of the data you do have, whether you "deserve" to be a co-author on labmate's research paper, whether you "deserve" to graduate, whether (s)he should tell your committee members to give you a hard time on your preliminary exam, whether (s)he will ask you to co-author a review paper with her/him, whether (s)he will send you to an international conference, whether (s)he will call up her/his colleagues and tell them that they've got to take you as a postdoc or give you an award, or whether (s)he will think of you when opportunities come up that could enhance your career.
And don't discount how your advisor's and co-workers' general attitude day in and day out can affect your overall happiness and productiveness. If you're working your ass off and are still being perceived as being lazy it can lead to feelings of self-defeat.
I am not suggesting that you barge into your advisor's office with a pipettor in your hands at every opportunity in order to let your advisor know you're working hard (though some of my former colleagues certainly used this tactic), but it may be worth thinking about ways to communicate that you're working when you're not physically there. Perhaps leaving a note at your desk letting people know where you are if you spend long hours working in a different location will be somewhat helpful. It saddens me that you should have to do this sort of thing for a research job where flexibility and independence is supposed to help offset the low pay and long hours, and where most of the time people don't really need to be able to reach you.
3. Working harder can be a trap that decreases productivity
I know how easy that trap is to fall into. Your experiments don't work (see point 8) and you think "well if I just work harder, I will get this experiment to work so that I can finally have good data to publish."
That's only sort of true. You will have to repeat experiments, but it's not good to keep banging your head against the same wall. If you just work non-stop, you're likely to end up in an often non-productive loop of repeating experiments that may be pointless, inappropriate, poorly designed, etc.. You likely won't realize it until after you've wasted a lot of time since you never stopped to take a step back to think about what you were actually doing.
Allowing yourself the coffee break, short walk outside, evening off, weekend off, or whatever it is you need to do to recharge will do more for your productivity than going into mindless robot mode. Believe it or not, it's often during these pauses that breakthrough ideas and inspiration will hit.
4. Thinking about the big picture is really important.
In order to do this, it's best if you're not actually doing experiments 100% of the time. Thinking about the big picture (asking yourself what questions you're actually trying to answer, reading about papers in your field, etc.) is actually part of your work, but many people don't recognize it as such. Some molecular biologists don't think you're working unless you have a pipettor in your hands. Spend time being pipettor-less anyway. Avoid the endless "repeat experiments" cycle mentioned in point 3.
5. Most of your colleagues and professors in molecular biology don't know how to analyze numerical data.
You will probably hear the phrase "statistically significant" or "p-value" uttered about a million times by people who have no idea what either of those really mean. You probably will not ever hear the phrase "error propagation" by people who should be doing it.
I don't know that there's much to do about this one unless you want to spend a lot of time campaigning for reform in the basic education of molecular biologists. Realistically, all it should take is, say, a mandatory one-semester course to provide the basics of data analysis, which each scientist should supplement with self-study / other coursework as appropriate later on. But implementing such a requirement requires that those who run molecular biology graduate school programs 1) realize there is a need (hard to do if the professors / education administrators themselves don't know how to analyze data), 2) give a shit, and 3) put in the effort.
Nevertheless, if you at least realize this one point early on, it may buffer the disappointment or shock when you see how data is misinterpreted and statistics abused by your fellow molecular biologists, even in publications in peer-reviewed journals like Nature or Science.
6. Your advisor isn't necessarily going to advise you.
In my program, the word "advisor" was in and of itself an oxymoron. Graduate students who expected their advisors to do any advising were ridiculed as naive. In some places, "advisors" apply for grants, (ostensibly) run the lab, and nag you for data. That's it. You want advice? Most people say you should figure it out yourself.
I say you should try to learn as much as you can, whether it's from other people or on your own. You're in graduate school to learn. If you already knew everything, why would you be in school? The whole idea that you should somehow contribute an original piece of research completely independently without any advice from other scientists is absurd.
Just because the "advisors" at your institution don't generally advise, it doesn't mean you should flounder on your own. Everyone gets help in some way or another; not everyone admits it. Most people would rather believe that science is some holy meritocracy where the successful people got where they are due to a pure combination of innate talent and hard work. Don't fall into that trap.
7. Mentors exist, but they won't wear a name-tag that says "Mentor" on it.
They won't necessarily identify themselves as mentors to you, so you have to identify them and reach out to them. I don't mean that you should look professors up randomly and ask if they'll be your mentor. What you need to do is be on alert when you have interactions with anyone - and this includes people aside from professors as well - who might have experience that you could benefit learning from.
You may have interactions with someone, for example, by TA-ing for them, having them on your committee, learning techniques from them, being in the same journal club, etc. Pick their brains when you can and if it seems they are open to it. Even in a relatively hostile environment there are usually some more experienced people who are more than flattered to be asked for advice and willing to extend a hand to a more junior scientist.
(Just as a small related story, my research institute's yearly retreat once featured an outside guest who was a well-respected Nobel laureate. At the pre-dinner reception, everyone including all the professors at my institute wanted to talk to him. But I could tell that a lot of them were nervous about approaching such a big shot, so they stood awkwardly across the room and glanced over at him repeatedly. My friends and I, all lowly first year graduate students at the time, approached him and struck up a conversation. He was super-friendly and delighted to talk to us. Much to the chagrin of the professors who were dying to talk to the Nobel laureate, he sat down at our table for dinner and not theirs. So don't assume that because someone's a big shot, they won't give you the time of day.)
8. Experiments will fail. Again and again.
Anyone whose experiments "work all the time" is faking data. In a field like molecular biology, the science you do is basically modern voodoo. There are protocols and sometimes a bit of troubleshooting gets things working, but mostly you just follow the steps and hope that the molecules were in the right mood and touch each other in the right way so that you get a faint band on an autoradiograph a few days later (just as an example). When they don't work, it doesn't mean that you're an awful person or that you're stupid or incompetent. You will have to repeat experiments whether they work or not (to validate them). Just see point 3 above and try not to beat yourself up about it when experiments fail.
9. Institutional prestige and the possession of advanced degrees do not equal intelligence.
So don't equate them. And don't equate the prestige of your institution with intellectual rigor, definitely not when it comes to graduate school in molecular biology. I'm sorry to say this, but most of the actual work in graduate school in this discipline ends up being monkey work (even though it shouldn't be). So don't be surprised that a lot of people who have Ph.D.s in this field nevertheless aren't particularly bright. Being bright isn't required to earn a Ph.D..
Becoming a postdoc somewhere usually involves applying directly to a given lab. In such cases, usually only the professor decides if a given applicant should be allowed to join the lab as a postdoc. There's no universal "standard" that is met by postdocs at a given institution, even the so-called elite ones.
Therefore, you may well meet colleagues (and professors) who can't handle diluting a 10 M stock of hydrochloric acid to get a 1 N working solution, or who think that by mixing together 4 different deoxynucleotide tri-phosphate stocks at 25 uM each, he can get a 100 uM stock dNTP solution. (Just to give two examples taken from real life.)
But ability (or lack thereof) as regards skills that should be essential for scientists aside, a lot of people (including professors) you meet in graduate school are, simply put, not that bright. Unfortunately, I've observed that the ones that are less bright are nevertheless the snootiest, perhaps because they're not actually bright enough to realize how limited their capacities are.
10. There are sketchy people at your graduate school
There are people who aren't trustworthy when it comes to professional interactions, for example, people who try to steal credit for other people's work, steal data, etc.
There are people who aren't trustworthy when it comes to personal interactions.
There are misogynists, racists, bigots, you name it.
It's just like the rest of society. Somehow a lot of people equate the academic research world with ascetism. But there's no reason to expect that because you're in an academic research institution the people around you are somehow ethically superior to those in the non-academic research world. In fact, in some ways, many may be ethically worse, because they're so desperate for a paper.