Papers › Penzai + Treescope: A Toolkit for Interpreting, Visualizing, and Editing Models As Data

Penzai + Treescope: A Toolkit for Interpreting, Visualizing, and Editing Models As Data

1 Aug 2024arXiv:2408.00211archive 2025-07-28

Daniel D. Johnson

Much of today's machine learning research involves interpreting, modifying or visualizing models after they are trained. I present Penzai, a neural network library designed to simplify model manipulation by representing models as simple data structures, and Treescope, an interactive pretty-printer and array visualizer that can visualize both model inputs/outputs and the models themselves. Penzai models are built using declarative combinators that expose the model forward pass in the structure of the model object itself, and use named axes to ensure each operation is semantically meaningful. With Penzai's tree-editing selector system, users can both insert and replace model components, allowing them to intervene on intermediate values or make other edits to the model structure. Users can then get immediate feedback by visualizing the modified model with Treescope. I describe the motivation and main features of Penzai and Treescope, and discuss how treating the model as data enables a variety of analyses and interventions to be implemented as data-structure transformations, without requiring model designers to add explicit hooks.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2408.00211")

Code

Syntology Ran 14 of 23 code samples harvested from 2 repositories linked to this paper; 9 have no recorded run. Of those that ran: 14 ran with no contract checked.

By repository: official repository: 23 samples from 2 repositories, 14 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

google-deepmind/penzai officialmentioned on GitHubjaxApache-2.0 report
google-deepmind/treescope officialmentioned on GitHubpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

23 samples harvested; 14 ran; 0 honoured the contract we drafted; 9 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

14ran
9unverified

Licence: 0 of the 23 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

color_from_string google-deepmind/treescope/treescope/formatting_util.py official repository ran Apache-2.0 (permissive) · 1ed46793bced4f6f · report
dataclass_from_attributes google-deepmind/treescope/treescope/dataclass_util.py official repository ran Apache-2.0 (permissive) · 491da2a807a0d703 · report
get_dtype_name google-deepmind/treescope/treescope/dtype_util.py official repository ran Apache-2.0 (permissive) · 0bcd3e980f7fe655 · report
init_takes_fields google-deepmind/treescope/treescope/dataclass_util.py official repository ran Apache-2.0 (permissive) · 3af90f8f7873fac9 · report
is_floating_dtype google-deepmind/treescope/treescope/dtype_util.py official repository ran Apache-2.0 (permissive) · 07b1fce0fdc1eefd · report
is_integer_dtype google-deepmind/treescope/treescope/dtype_util.py official repository ran Apache-2.0 (permissive) · 5c689b98d9651bb5 · report
is_pytree_dataclass_type google-deepmind/penzai/penzai/core/struct.py official repository ran Apache-2.0 (permissive) · 8da90bf2b11facb7 · report
oklch_color google-deepmind/treescope/treescope/formatting_util.py official repository ran Apache-2.0 (permissive) · 49862fb509edca46 · report
parse_simple_color_and_pattern_spec google-deepmind/treescope/treescope/formatting_util.py official repository ran Apache-2.0 (permissive) · c54a7b3f3d04bf76 · report
styled google-deepmind/treescope/treescope/figures.py official repository ran Apache-2.0 (permissive) · cc38f8434830dd01 · report
tree_flatten_exactly_one_level google-deepmind/penzai/penzai/core/tree_util.py official repository ran Apache-2.0 (permissive) · 7f0011a936783767 · report
var google-deepmind/penzai/penzai/core/shapecheck.py official repository ran Apache-2.0 (permissive) · c716cd4fc3d93251 · report
vars_for_axes google-deepmind/penzai/penzai/core/shapecheck.py official repository ran Apache-2.0 (permissive) · ad130894fc372aec · report
with_font_size google-deepmind/treescope/treescope/figures.py official repository ran Apache-2.0 (permissive) · 314cb1c24954c222 · report
default_well_known_filter google-deepmind/treescope/treescope/canonical_aliases.py official repository unverified Apache-2.0 (permissive) · 4eb59933d52ad100 · report
indented google-deepmind/treescope/treescope/figures.py official repository unverified Apache-2.0 (permissive) · e4fd281c5d149c61 · report
is_pytree_node_field google-deepmind/penzai/penzai/core/struct.py official repository unverified Apache-2.0 (permissive) · ffd28ac2c65414f3 · report
lookup_alias google-deepmind/treescope/treescope/canonical_aliases.py official repository unverified Apache-2.0 (permissive) · 5c75d7c159ec68b6 · report
maybe_defer_rendering google-deepmind/treescope/treescope/lowering.py official repository unverified Apache-2.0 (permissive) · fe56c9c538b60fab · report
maybe_local_module_name google-deepmind/treescope/treescope/canonical_aliases.py official repository unverified Apache-2.0 (permissive) · ad8ee3a7572721bb · report
pretty_keystr google-deepmind/penzai/penzai/core/tree_util.py official repository unverified Apache-2.0 (permissive) · c45870654f09629f · report
render_to_html_as_root google-deepmind/treescope/treescope/lowering.py official repository unverified Apache-2.0 (permissive) · 4fe2c2ee820dc796 · report
render_to_text_as_root google-deepmind/treescope/treescope/lowering.py official repository unverified Apache-2.0 (permissive) · 2a11ffd4903b882a · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections