Dataset Viewer

Split (2)

train

The dataset viewer is not available for this split.

Cannot load the dataset split (in streaming mode) to extract the first rows.

Error code:   StreamingRowsError
Exception:    ValueError
Message:      Expected object or value
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 290, in _generate_tables
                  pa_table = paj.read_json(
                      io.BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
                  )
                File "pyarrow/_json.pyx", line 342, in pyarrow._json.read_json
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
                  raise convert_status(status)
              pyarrow.lib.ArrowInvalid: JSON parse error: Column() changed from object to string in row 0
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 489, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2818, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 304, in _generate_tables
                  batch = json_encode_fields_in_json_lines(original_batch, json_field_paths)
                File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 111, in json_encode_fields_in_json_lines
                  examples = [ujson_loads(line) for line in original_batch.splitlines()]
                              ~~~~~~~~~~~^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 20, in ujson_loads
                  return pd.io.json.ujson_loads(*args, **kwargs)
                         ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
              ValueError: Expected object or value

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Experiment 04 — base fixed-mix continued pretraining

Public backup of the reproducibility metadata for base-exp04.

Snapshot

Sequence length: 2,048
Train budget prepared: 5,000,000 sequences
Evaluation set: 20,000 sequences across 10 sources
Train seed: 86
Evaluation seed: 1,234
Batch size: 4 per device
Gradient accumulation: 8
Devices: 8
Tokens per optimizer step: 524,288
Learning rate: constant 5e-6
Planned steps: 3,000
Evaluation interval: 100
Snapshot checkpoint: step 2300
Step-2300 weighted evaluation loss: 1.4013190374642612

sources.json: train/evaluation source weights and public bucket references
manifests/train_manifest.json: exact sampled train shards and sequence ranges
manifests/eval_manifest.json: exact sampled evaluation shards and sequence ranges
manifests/train_exclude.json: train/evaluation exclusion metadata
checkpoint_index.json: checkpoint ranking at snapshot time

Weights are stored separately in peter090/exp_04_base-exp04_weights. That repository contains four base-weight shards; model snapshots for steps 500, 1100, 1200, 1300, 2000, 2100, 2200, and 2300; and resumable optimizer states for steps 1100, 1200, 1300, 2100, 2200, and 2300. Embedded local workspace paths in the optimizer-state pickle metadata were replaced with an opaque public path of identical byte length. Each scrubbed state passed a forbidden-string scan, ZIP CRC verification, and torch.load with memory mapping. Model configuration, executable code, logs, and credentials are excluded.

Downloads last month: 22

Total file size:

1.96 MB

Datasets:

peter090
/

exp_04_base-exp04_meta

Experiment 04 — base fixed-mix continued pretraining

Snapshot

Contents