You can use node n6 (ssh n6), for simple interactive prototyping, low-load installations etc., but if you need to run large interactive jobs, please let the scheduler allocate the resources that you need.
Using option -I (upper case i) will tell qsub that you want an interactive job shell, e.g.
{cclc01}$ qsub -I
qsub: waiting for job 693822.cclc01.som.ucsf.edu to start
qsub: job 693822.cclc01.som.ucsf.edu ready
{n3}$: echo $PBS_NUM_PPN ## Maximum number of parallel tasks
1
{n3}$:
This will give a single-core slot on any machine that the scheduler finds suitable (here it happened to be n3). From here on you can use the terminal as a regular terminal. Your job is listed by qstat -u $USER just like any other jobs. As soon as you log out, the job will be freed.
To run multi-core processes interactively, you can request more cores. Likewise, you can specify memory needs and other resources. For instance, if you need four cores and 10 GiB of RAM, use:
$ qsub -I -l nodes=1:ppn=4 -l vmem=10gb
qsub: waiting for job 693829.cclc01.som.ucsf.edu to start
qsub: job 693829.cclc01.som.ucsf.edu ready
{n2}$: echo $PBS_NUM_PPN ## Maximum number of parallel tasks
4
{n2}$: