Because `predownload` was not specified, it will default to 8*batch_size if batch_size is not None, otherwise 64. Prior to Streaming v0.7.0, `predownload` defaulted to max(batch_size, 256 * batch_size // num_canonical_nodes).
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Using local dataset /home/oweller/bert24-data/data/ProLong_text/arxiv/2-1...
Using tokenizer model bclavie/olmo_bert_template...
Using 1 processes for tokenization, with 1000 items per process.

Processing samples:   0%|          | 0/30011 [00:00<?, ?it/s]
Processing samples:   3%|▎         | 1000/30011 [01:28<42:43, 11.32it/s]
Processing samples:   3%|▎         | 1000/30011 [01:40<42:43, 11.32it/s]
Processing samples:   7%|▋         | 2000/30011 [02:41<37:01, 12.61it/s]
Processing samples:   7%|▋         | 2000/30011 [03:00<37:01, 12.61it/s]
Processing samples:  10%|▉         | 3000/30011 [03:59<35:22, 12.73it/s]
Processing samples:  10%|▉         | 3000/30011 [04:10<35:22, 12.73it/s]
Processing samples:  13%|█▎        | 4000/30011 [05:12<33:13, 13.05it/s]
Processing samples:  13%|█▎        | 4000/30011 [05:30<33:13, 13.05it/s]
Processing samples:  17%|█▋        | 5000/30011 [06:31<32:18, 12.90it/s]
Processing samples:  17%|█▋        | 5000/30011 [06:50<32:18, 12.90it/s]
Processing samples:  20%|█▉        | 6000/30011 [07:42<30:09, 13.27it/s]
Processing samples:  20%|█▉        | 6000/30011 [08:00<30:09, 13.27it/s]
Processing samples:  23%|██▎       | 7000/30011 [08:51<28:06, 13.64it/s]
Processing samples:  23%|██▎       | 7000/30011 [09:10<28:06, 13.64it/s]
Processing samples:  27%|██▋       | 8000/30011 [10:17<28:20, 12.94it/s]
Processing samples:  27%|██▋       | 8000/30011 [10:30<28:20, 12.94it/s]
Processing samples:  30%|██▉       | 9000/30011 [11:34<26:58, 12.98it/s]
Processing samples:  30%|██▉       | 9000/30011 [11:50<26:58, 12.98it/s]
Processing samples:  33%|███▎      | 10000/30011 [12:38<24:24, 13.67it/s]
Processing samples:  33%|███▎      | 10000/30011 [12:50<24:24, 13.67it/s]
Processing samples:  37%|███▋      | 11000/30011 [14:08<24:48, 12.77it/s]
Processing samples:  37%|███▋      | 11000/30011 [14:20<24:48, 12.77it/s]
Processing samples:  40%|███▉      | 12000/30011 [15:45<25:11, 11.91it/s]
Processing samples:  40%|███▉      | 12000/30011 [16:00<25:11, 11.91it/s]
Processing samples:  43%|████▎     | 13000/30011 [16:56<22:43, 12.48it/s]
Processing samples:  43%|████▎     | 13000/30011 [17:10<22:43, 12.48it/s]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/oweller/bert24-data/src/bert24_data/tokenization/tokenize_mds.py", line 64, in sample_dataset_from_config
    with mp.Pool(num_processes) as pool:
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/pool.py", line 739, in __exit__
    self.terminate()
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/pool.py", line 657, in terminate
    self._terminate()
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/util.py", line 224, in __call__
    res = self._callback(*self._args, **self._kwargs)
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/pool.py", line 732, in _terminate_pool
    p.join()
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/process.py", line 149, in join
    res = self._popen.wait(timeout)
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/popen_fork.py", line 43, in wait
    return self.poll(os.WNOHANG if timeout == 0.0 else 0)
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/multiprocessing/popen_fork.py", line 27, in poll
    pid, sts = os.waitpid(self.pid, flag)
KeyboardInterrupt

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/oweller/bert24-data/src/bert24_data/tokenization/tokenize_mds.py", line 108, in <module>
    sample_dataset_from_config(args)
  File "/home/oweller/bert24-data/src/bert24_data/tokenization/tokenize_mds.py", line 63, in sample_dataset_from_config
    with MDSWriter(out=output_dir, columns=MDS_COLS_PRE_TOKENIZED, compression='zstd') as mds_writer:
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/site-packages/streaming/base/format/base/writer.py", line 354, in __exit__
    self.finish()
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/site-packages/streaming/base/format/base/writer.py", line 298, in finish
    self.flush_shard()
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/site-packages/streaming/base/format/base/writer.py", line 428, in flush_shard
    raw_data = self.encode_joint_shard()
  File "/home/oweller/miniconda3/envs/bert24/lib/python3.10/site-packages/streaming/base/format/mds/writer.py", line 143, in encode_joint_shard
    sample_data = b''.join(self.new_samples)
KeyboardInterrupt

Processing samples:  43%|████▎     | 13000/30011 [18:07<23:42, 11.96it/s]