Papers › Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning

Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning

17 May 2024arXiv:2405.12241archive 2025-07-28

Dan Braun, Jordan Taylor, Nicholas Goldowsky-Dill, Lee Sharkey

Identifying the features learned by neural networks is a core challenge in mechanistic interpretability. Sparse autoencoders (SAEs), which learn a sparse, overcomplete dictionary that reconstructs a network's internal activations, have been used to identify these features. However, SAEs may learn more about the structure of the datatset than the computational structure of the network. There is therefore only indirect reason to believe that the directions found in these dictionaries are functionally important to the network. We propose end-to-end (e2e) sparse dictionary learning, a method for training SAEs that ensures the features learned are functionally important by minimizing the KL divergence between the output distributions of the original model and the model with SAE activations inserted. Compared to standard SAEs, e2e SAEs offer a Pareto improvement: They explain more network performance, require fewer total features, and require fewer simultaneously active features per datapoint, all with no cost to interpretability. We explore geometric and qualitative differences between e2e SAE features and standard SAE features. E2e dictionary learning brings us closer to methods that can explain network behavior concisely and accurately. We release our library for training e2e SAEs and reproducing our analysis at https://github.com/ApolloResearch/e2e_sae

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 11 code samples harvested from 1 repository linked to this paper: 8 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 11 samples from 1 repository, 8 ran (5 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

apolloresearch/e2e_sae officialmentioned in paperjaxMITSyntology: 11 samples, 8 ran (3 instrument-failed) · 3 unverified 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

11 samples harvested: 8 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
3ran · our draft was wrong
4ran
3unverified

Licence: 0 of the 11 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 apolloresearch/e2e_sae. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

apolloresearch/e2e_sae official repository 11 samples, 8 ran (3 instrument-failed) · 3 unverified
calc_explained_variance e2e_sae/losses.py ran MIT (permissive) · 4240424af18cec3a · report fixture: signature pred:Tensor/3/float, target:Tensor/3/float, layer_norm:bool/0/bool · returned Tensor · level recorded 2026-09-24
load_config in 389 papers e2e_sae/utils.py ran · our draft was wrong MIT (permissive) · c965ff7e1ebc030d · report fixture: signature config_path_or_obj:str/0/str, config_model:callable/0/? · returned MyConfig · level recorded 2026-09-24
setup_logger in 122 papers e2e_sae/log.py ran · our draft was wrong MIT (permissive) · d8a05e02085eeb90 · report fixture: returned RootLogger · level recorded 2026-09-24
statistical_distance e2e_sae/metrics.py ran · our draft was wrong fingerprinted MIT (permissive) · 940eb3f8703e24c9 · report fixture: signature orig_logits:Tensor/2/float, new_logits:Tensor/2/float · returned Tensor · level recorded 2026-09-24
to_numpy in 103 papers
census: 24 implementations from 29 papers, 2 distinct outputs across 2 buckets, one shared input per bucket
e2e_sae/utils.py ran · honoured contract MIT (permissive) · 10dfdb7059057d01 · report fixture: signature tensor:Tensor/2/float · returned ndarray · level recorded 2026-09-24
tokenize_and_concatenate e2e_sae/data.py ran MIT (permissive) · 785cf677e1de98f6 · report fixture: signature dataset:Dataset/0/?, tokenizer:callable/0/?, max_length:int/0/int, column_name:str/0/str, add_bos_token:bool/0/bool, num_proc:int/0/int · returned Dataset · level recorded 2026-09-24
top1_consistency e2e_sae/metrics.py ran fingerprinted MIT (permissive) · fd3eafcb040f48d9 · report fixture: signature orig_logits:Tensor/2/float, new_logits:Tensor/2/float, per_token:bool/0/bool · returned Tensor · level recorded 2026-09-24
topk_accuracy in 14 papers e2e_sae/metrics.py ran MIT (permissive) · 328f50c831db8d0f · report fixture: signature logits:Tensor/3/float, tokens:Tensor/2/int, k:int/0/int, per_token:bool/0/bool · returned Tensor · level recorded 2026-09-24
convert_str_to_torch_dtype e2e_sae/types.py unverified MIT (permissive) · 495a63f69d38d772 · report
load_pretrained_saes e2e_sae/loader.py unverified MIT (permissive) · 0b6d19ab9443ea4e · report
serialize_torch_dtype_to_str e2e_sae/types.py unverified MIT (permissive) · 0c353f5afddae0f1 · report

Tasks

Dictionary Learning

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