Ms. Pac-Man RL neuron interpretability
Collection
Neuron-level interpretability and steering of reward-trained (no imitation) Ms. Pac-Man RL agents. • 4 items • Updated
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: TypeError
Message: Couldn't cast array of type
struct<fc379_ablate: struct<metric: string, n: int64, delta: double, ci: list<item: double>>, c3_ch3_cell00: struct<metric: string, n: int64, delta: double, ci: list<item: double>>>
to
{'fc453_ablate': {'metric': Value('string'), 'n': Value('int64'), 'delta': Value('float64'), 'ci': List(Value('float64'))}, 'c3_ch18_cell56': {'metric': Value('string'), 'n': Value('int64'), 'delta': Value('float64'), 'ci': List(Value('float64'))}}
Traceback: 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 544, 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 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, 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 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, 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 361, 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 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2315, in cast_table_to_schema
cast_array_to_feature(
~~~~~~~~~~~~~~~~~~~~~^
table[name] if name in table_column_names else pa.array([None] * len(table), type=schema.field(name).type),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
feature,
^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1861, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2161, in cast_array_to_feature
raise TypeError(f"Couldn't cast array of type\n{_short_str(array.type)}\nto\n{_short_str(feature)}")
TypeError: Couldn't cast array of type
struct<fc379_ablate: struct<metric: string, n: int64, delta: double, ci: list<item: double>>, c3_ch3_cell00: struct<metric: string, n: int64, delta: double, ci: list<item: double>>>
to
{'fc453_ablate': {'metric': Value('string'), 'n': Value('int64'), 'delta': Value('float64'), 'ci': List(Value('float64'))}, 'c3_ch18_cell56': {'metric': Value('string'), 'n': Value('int64'), 'delta': Value('float64'), 'ci': List(Value('float64'))}}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.
Ms. Pac-Man APPO neuron atlas
Derived artefacts of a preregistered neuron-level interpretability and steering study of reward-trained (no imitation learning) Sample Factory APPO agents playing Atari Ms. Pac-Man.
results/*.json: the analysis outputs from which the study's report generator (pacneuron.report) builds every results table and inline number in the README, reports and paper:- fidelity, ground-truth validation;
- audits (per fc unit: best feature, R², action votes, top-activating states);
- screens, pilots, confirmatory H1/H2 tests, replication, exploratory claim checks.
Numbers inside the frozen preregistration and its addendum were written by hand before the results existed.
site_data/*.json: per-unit data behind the interactive microscope (alive fc units only).analysis/*.npz: unit × feature R² matrices for the agent and controls from the exploratory discovery split-half audit (not the held-out H1 test), bootstrap draws, open-loop screen maps.
Top-activating frames are not stored here (the GitHub repository ships them as WebP grids under site/img). They
can be re-rendered by deterministic replay from (game seed, step), which needs the recorded action sequences: run
the collection phases of the code at https://github.com/BurnyCoder/pacman-reinforcement-learning-neuron-mechanistic-interpretability to regenerate them. Agents:
- Downloads last month
- 80