Dataset Viewer
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: CastError
Message: Couldn't cast
schema_version: string
created_unix_seconds: double
design: struct<depth: int64, event_directions: list<item: string>, target_head: string, batch_sizes: list<it (... 124 chars omitted)
child 0, depth: int64
child 1, event_directions: list<item: string>
child 0, item: string
child 2, target_head: string
child 3, batch_sizes: list<item: int64>
child 0, item: int64
child 4, repetitions: int64
child 5, direction_semantics: string
child 6, authority: string
child 7, measurement: string
child 8, promotion_width: int64
thresholds: struct<absolute_coefficient_tolerance: double, relative_coefficient_tolerance: double, material_effe (... 46 chars omitted)
child 0, absolute_coefficient_tolerance: double
child 1, relative_coefficient_tolerance: double
child 2, material_effect_floor: double
child 3, memory_ceiling_bytes: int64
torch_float64: struct<coefficients: list<item: list<item: double>>, seconds: double>
child 0, coefficients: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
child 1, seconds: double
retained: struct<method: string, batch_size: null, repetitions: int64, coefficients: list<item: list<item: dou (... 234 chars omitted)
child 0, method: string
child 1, batch_size: null
child 2, repetitions: int64
child 3, coefficients: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
child 4, seconds_samples: list<item: double>
...
rows: list<item: int64>, lr_multi (... 14 chars omitted)
child 0, action_id: int64
child 1, mask_id: int64
child 2, mask_bits: string
child 3, active_rows: list<item: int64>
child 0, item: int64
child 4, lr_multiplier: double
child 2, predicted_loss: double
child 18, exact_paths: int64
child 19, held_out_paths: int64
child 20, disconnected_propagators_mae: double
child 21, plus_one_frame_vertices_mae: double
child 22, plus_learned_connector_mae: double
child 23, connected_prediction_pearson: double
child 24, predicted_best_held_actual_rank: int64
child 25, predicted_best_held_regret: double
child 26, boundary: string
readiness: struct<qualified: bool, step: int64, training_metrics: struct<loss_a: double, loss_b: double, loss_s (... 112 chars omitted)
child 0, qualified: bool
child 1, step: int64
child 2, training_metrics: struct<loss_a: double, loss_b: double, loss_sum: double, accuracy_a: double, accuracy_b: double>
child 0, loss_a: double
child 1, loss_b: double
child 2, loss_sum: double
child 3, accuracy_a: double
child 4, accuracy_b: double
child 3, primary_requirement: double
child 4, secondary_requirement: double
artifact: struct<path: string, sha256: string>
child 0, path: string
child 1, sha256: string
experiment: string
to
{'experiment': Value('string'), 'contract': {'full_actions_per_lane': Value('int64'), 'complete_one_frame_paths': Value('int64'), 'selection_split': Value('string'), 'keyframe': Value('string')}, 'readiness': {'qualified': Value('bool'), 'step': Value('int64'), 'training_metrics': {'loss_a': Value('float64'), 'loss_b': Value('float64'), 'loss_sum': Value('float64'), 'accuracy_a': Value('float64'), 'accuracy_b': Value('float64')}, 'primary_requirement': Value('float64'), 'secondary_requirement': Value('float64')}, 'batches': List({'name': Value('string'), 'complete_one_frame_paths': Value('int64'), 'random_sample_paths': Value('int64'), 'top_fraction': Value('float64'), 'global_best': {'loss': Value('float64'), 'lane_a': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'lane_b': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'inside_original_pruned_alphabet': Value('bool')}, 'row_frequency_all': {'lane_a': {'0': Value('float64'), '1': Value('float64'), '2': Value('float64'), '3': Value('float64'), '4': Value('float64'), '5': Value('float64'), '6': Value('float64'), '7': Value('float64')}, 'lane_b': {'0': Value('float64'), '1': Value('float64'), '2': Value('float64'), '3': Value('float64'), '4': Value('float64'), '5': Value('float64'), '6': Value('float64'),
...
': List({'rank': Value('int64'), 'observed_fraction': Value('float64'), 'held_out_residual_mae': Value('float64'), 'held_out_loss_mae': Value('float64'), 'predicted_best_held_actual_rank': Value('int64'), 'predicted_best_held_regret': Value('float64'), 'round': Value('int64'), 'active_masks_lane_a': Value('int64'), 'active_masks_lane_b': Value('int64'), 'observed_joint_paths': Value('int64'), 'best_observed_loss': Value('float64'), 'global_regret': Value('float64'), 'global_best_discovered': Value('bool'), 'proposed': {'lane_a': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'lane_b': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'predicted_loss': Value('float64')}}), 'exact_paths': Value('int64'), 'held_out_paths': Value('int64'), 'disconnected_propagators_mae': Value('float64'), 'plus_one_frame_vertices_mae': Value('float64'), 'plus_learned_connector_mae': Value('float64'), 'connected_prediction_pearson': Value('float64'), 'predicted_best_held_actual_rank': Value('int64'), 'predicted_best_held_regret': Value('float64'), 'boundary': Value('string')}), 'boundaries': List(Value('string')), 'artifact': {'path': Value('string'), 'sha256': Value('string')}, 'public_derivative': {'original_summary_sha256': Value('string'), 'transformation': Value('string')}}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, 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 127, 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 483, 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 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, 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 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
schema_version: string
created_unix_seconds: double
design: struct<depth: int64, event_directions: list<item: string>, target_head: string, batch_sizes: list<it (... 124 chars omitted)
child 0, depth: int64
child 1, event_directions: list<item: string>
child 0, item: string
child 2, target_head: string
child 3, batch_sizes: list<item: int64>
child 0, item: int64
child 4, repetitions: int64
child 5, direction_semantics: string
child 6, authority: string
child 7, measurement: string
child 8, promotion_width: int64
thresholds: struct<absolute_coefficient_tolerance: double, relative_coefficient_tolerance: double, material_effe (... 46 chars omitted)
child 0, absolute_coefficient_tolerance: double
child 1, relative_coefficient_tolerance: double
child 2, material_effect_floor: double
child 3, memory_ceiling_bytes: int64
torch_float64: struct<coefficients: list<item: list<item: double>>, seconds: double>
child 0, coefficients: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
child 1, seconds: double
retained: struct<method: string, batch_size: null, repetitions: int64, coefficients: list<item: list<item: dou (... 234 chars omitted)
child 0, method: string
child 1, batch_size: null
child 2, repetitions: int64
child 3, coefficients: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
child 4, seconds_samples: list<item: double>
...
rows: list<item: int64>, lr_multi (... 14 chars omitted)
child 0, action_id: int64
child 1, mask_id: int64
child 2, mask_bits: string
child 3, active_rows: list<item: int64>
child 0, item: int64
child 4, lr_multiplier: double
child 2, predicted_loss: double
child 18, exact_paths: int64
child 19, held_out_paths: int64
child 20, disconnected_propagators_mae: double
child 21, plus_one_frame_vertices_mae: double
child 22, plus_learned_connector_mae: double
child 23, connected_prediction_pearson: double
child 24, predicted_best_held_actual_rank: int64
child 25, predicted_best_held_regret: double
child 26, boundary: string
readiness: struct<qualified: bool, step: int64, training_metrics: struct<loss_a: double, loss_b: double, loss_s (... 112 chars omitted)
child 0, qualified: bool
child 1, step: int64
child 2, training_metrics: struct<loss_a: double, loss_b: double, loss_sum: double, accuracy_a: double, accuracy_b: double>
child 0, loss_a: double
child 1, loss_b: double
child 2, loss_sum: double
child 3, accuracy_a: double
child 4, accuracy_b: double
child 3, primary_requirement: double
child 4, secondary_requirement: double
artifact: struct<path: string, sha256: string>
child 0, path: string
child 1, sha256: string
experiment: string
to
{'experiment': Value('string'), 'contract': {'full_actions_per_lane': Value('int64'), 'complete_one_frame_paths': Value('int64'), 'selection_split': Value('string'), 'keyframe': Value('string')}, 'readiness': {'qualified': Value('bool'), 'step': Value('int64'), 'training_metrics': {'loss_a': Value('float64'), 'loss_b': Value('float64'), 'loss_sum': Value('float64'), 'accuracy_a': Value('float64'), 'accuracy_b': Value('float64')}, 'primary_requirement': Value('float64'), 'secondary_requirement': Value('float64')}, 'batches': List({'name': Value('string'), 'complete_one_frame_paths': Value('int64'), 'random_sample_paths': Value('int64'), 'top_fraction': Value('float64'), 'global_best': {'loss': Value('float64'), 'lane_a': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'lane_b': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'inside_original_pruned_alphabet': Value('bool')}, 'row_frequency_all': {'lane_a': {'0': Value('float64'), '1': Value('float64'), '2': Value('float64'), '3': Value('float64'), '4': Value('float64'), '5': Value('float64'), '6': Value('float64'), '7': Value('float64')}, 'lane_b': {'0': Value('float64'), '1': Value('float64'), '2': Value('float64'), '3': Value('float64'), '4': Value('float64'), '5': Value('float64'), '6': Value('float64'),
...
': List({'rank': Value('int64'), 'observed_fraction': Value('float64'), 'held_out_residual_mae': Value('float64'), 'held_out_loss_mae': Value('float64'), 'predicted_best_held_actual_rank': Value('int64'), 'predicted_best_held_regret': Value('float64'), 'round': Value('int64'), 'active_masks_lane_a': Value('int64'), 'active_masks_lane_b': Value('int64'), 'observed_joint_paths': Value('int64'), 'best_observed_loss': Value('float64'), 'global_regret': Value('float64'), 'global_best_discovered': Value('bool'), 'proposed': {'lane_a': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'lane_b': {'action_id': Value('int64'), 'mask_id': Value('int64'), 'mask_bits': Value('string'), 'active_rows': List(Value('int64')), 'lr_multiplier': Value('float64')}, 'predicted_loss': Value('float64')}}), 'exact_paths': Value('int64'), 'held_out_paths': Value('int64'), 'disconnected_propagators_mae': Value('float64'), 'plus_one_frame_vertices_mae': Value('float64'), 'plus_learned_connector_mae': Value('float64'), 'connected_prediction_pearson': Value('float64'), 'predicted_best_held_actual_rank': Value('int64'), 'predicted_best_held_regret': Value('float64'), 'boundary': Value('string')}), 'boundaries': List(Value('string')), 'artifact': {'path': Value('string'), 'sha256': Value('string')}, 'public_derivative': {'original_summary_sha256': Value('string'), 'transformation': Value('string')}}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
SNO Article I Evidence
Machine-readable experimental records supporting Watching Training Move: Causal Forecasting of Neural Updates Across 8,192 SGD Transitions.
Creator: TiGa-RCE
Affiliation: Independent researcher; founder and sole operator of Brewster Jennings
The Dataset contains original result records, figure-data contracts, and machine-readable receipts. It does not contain the article prose, the rendered figures, the source code, private conversations, local system paths, model caches, or downloaded MNIST data.
The long-horizon evidence includes a matched-runtime reproduction receipt for the exact PyTorch 2.13 result through 8,192 transitions and separately records the observed PyTorch 2.8 strict-hash portability boundary.
evidence/horizon/zerogpu-portability/ adds a bounded three-horizon control
comparing the local PyTorch 2.13 CPU path with a PyTorch 2.13 CUDA run on
Hugging Face ZeroGPU. Every recorded scientific decision agreed; blanket
numerical identity did not. The control is neither a full 8,192-transition
reproduction nor a CPU-versus-GPU performance benchmark.
License
Original Dataset material is licensed under the Creative Commons Attribution
4.0 International license. The complete terms are included in LICENSE and
are also available at https://creativecommons.org/licenses/by/4.0/.
Preferred attribution: “TiGa-RCE, independent researcher; founder and sole operator of Brewster Jennings.” Third-party identifiers and factual references are not relicensed by this Dataset.
Integrity
MANIFEST.sha256 binds every published Dataset file except the manifest
itself. The final HF Space must pin an exact Dataset commit rather than follow
the mutable main branch.
Status: public evidence Dataset. Publishing the Community Article remains a separate final gate.
- Downloads last month
- 347