4 Commits

Author	SHA1	Message	Date
dhanajitb	f872eb98c2	making unconditional generation work ... The unconditional generation works now but if the seed is fixed, the sample is the same every time. n_samples > 1 will give different samples though. I am giving the start token as '<|endoftext|>' for the unconditional generation.	2019-03-28 22:46:15 +05:30
Elon Musk	66d8206809	Update run_gpt2.py	2019-03-08 11:59:08 -05:00
Catalin Voss	4b4b079272	Fix top k generation for k != 0	2019-03-02 21:54:44 -08:00
thomwolf	690a0dbf36	fix example - masking	2019-02-18 10:50:30 +01:00