* Output caching system:

  For output that is returned from actions, a system similar to the input
  cache exists but using _ instead of _i. Only actions that produce a result
  (NOT assignments, for example) are cached. If you are familiar with
  Mathematica, IPython's _ variables behave exactly like Mathematica's %
  variables.

  The following GLOBAL variables always exist (so don't overwrite them!):
  _ (one underscore): previous output.
  __ (two underscores): next previous.
  ___ (three underscores): next-next previous.

  Global variables named _<n> are dynamically created (<n> being the prompt
  counter), such that the result of output <n> is always available as _<n>.

  Finally, a global dictionary named _oh exists with entries for all lines
  which generated output.

In [1]: list(range(5))
Out[1]: [0, 1, 2, 3, 4]

In [2]: list(range(5))
Out[2]: [0, 1, 2, 3, 4]

In [3]: list(range(5))
Out[3]: [0, 1, 2, 3, 4]

In [4]: list(range(5))
Out[4]: [0, 1, 2, 3, 4]

In [5]: list(range(5))
Out[5]: [0, 1, 2, 3, 4]

In [6]: list(range(5))
Out[6]: [0, 1, 2, 3, 4]

In [7]: list(range(5))
Out[7]: [0, 1, 2, 3, 4]

In [8]: list(range(5))
Out[8]: [0, 1, 2, 3, 4]

In [9]: list(range(5))
Out[9]: [0, 1, 2, 3, 4]

In [10]: list(range(5))
Out[10]: [0, 1, 2, 3, 4]

In [11]: list(range(5))
Out[11]: [0, 1, 2, 3, 4]

In [12]: list(range(5))
Out[12]: [0, 1, 2, 3, 4]

In [13]: list(range(5))
Out[13]: [0, 1, 2, 3, 4]

In [14]: list(range(5))
Out[14]: [0, 1, 2, 3, 4]

In [15]: list(range(5))
Out[15]: [0, 1, 2, 3, 4]

In [16]: _oh
Out[16]: 
{1: [0, 1, 2, 3, 4],
 2: [0, 1, 2, 3, 4],
 3: [0, 1, 2, 3, 4],
 4: [0, 1, 2, 3, 4],
 5: [0, 1, 2, 3, 4],
 6: [0, 1, 2, 3, 4],
 7: [0, 1, 2, 3, 4],
 8: [0, 1, 2, 3, 4],
 9: [0, 1, 2, 3, 4],
 10: [0, 1, 2, 3, 4],
 11: [0, 1, 2, 3, 4],
 12: [0, 1, 2, 3, 4],
 13: [0, 1, 2, 3, 4],
 14: [0, 1, 2, 3, 4],
 15: [0, 1, 2, 3, 4]}

In [17]: id(_oh[1])
Out[17]: 140185325829128

In [18]: id(_oh[2])
Out[18]: 140185325829960

In [19]: id(_oh[15])
Out[19]: 140185323832840

In [20]: