InvalidArgumentError error message during conditioned text generation

[negacy]

Any thoughts for the following error message when running the conditioned text generation script on the 117M pre-trained model for a bit large input text? InvalidArgumentError (see above for traceback): indices[0,0] = 1024 is not in [0, 1024) [[Node: sample_sequence/while/model/GatherV2_1 = GatherV2[Taxis=DT_INT32, Tindices=DT_INT32, Tparams=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](sample_sequence/while/model/GatherV2_1/Enter, sample_sequence/while/model/ExpandDims, sample_sequence/while/model/h11/attn/range/start, ^sample_sequence/while/model/h11/attn/strided_slice_8/stack_2)]]

[miguelrodriguezpro]

same here

[lostmsu]

BTW, this only happens on CPU.