the Pdb problem

[arnold-em]

gpu_rank 0
[2022-11-09 11:32:53,261 INFO] loading vocabulary file https://s3.amazonaws.com/models.huggingface.co/bert/roberta-base-vocab.json from cache at /root/.cache/torch/transformers/d0c5776499adc1ded22493fae699da0971c1ee4c2587111707a4d177d20257a2.ef00af9e673c7160b4d41cfda1f48c5f4cba57d5142754525572a846a1ab1b9b
[2022-11-09 11:32:55,748 INFO] * number of parameters: 159693576
[2022-11-09 11:32:55,748 INFO] Start training...
[2022-11-09 11:32:56,667 INFO] Loading train dataset from ./arxivL/bert-files/2500-segmented-test/train.35.bert.pt, number of examples: 802

/root/workspace/cht/ExtendSumm/ExtendedSumm-master/src/models/data_loader.py(355)preprocess()
-> end_id = [src[-1]]
(Pdb)
the gpu is Tesla a100（40g） when i run train.sh then just （Pdb），could you tell me how to deal this