Papers › Sparse Autoencoders Find Highly Interpretable Features in Language Models

Sparse Autoencoders Find Highly Interpretable Features in Language Models

15 Sep 2023arXiv:2309.08600archive 2025-07-28

Hoagy Cunningham, Aidan Ewart, Logan Riggs, Robert Huben, Lee Sharkey

One of the roadblocks to a better understanding of neural networks' internals is polysemanticity, where neurons appear to activate in multiple, semantically distinct contexts. Polysemanticity prevents us from identifying concise, human-understandable explanations for what neural networks are doing internally. One hypothesised cause of polysemanticity is superposition, where neural networks represent more features than they have neurons by assigning features to an overcomplete set of directions in activation space, rather than to individual neurons. Here, we attempt to identify those directions, using sparse autoencoders to reconstruct the internal activations of a language model. These autoencoders learn sets of sparsely activating features that are more interpretable and monosemantic than directions identified by alternative approaches, where interpretability is measured by automated methods. Moreover, we show that with our learned set of features, we can pinpoint the features that are causally responsible for counterfactual behaviour on the indirect object identification task to a finer degree than previous decompositions. This work indicates that it is possible to resolve superposition in language models using a scalable, unsupervised method. Our method may serve as a foundation for future mechanistic interpretability work, which we hope will enable greater model transparency and steerability.

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 2 repositories linked to this paper: 8 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 1 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: 5 samples from 1 repository, 4 ran (1 with no instrument failure; 3 where Syntology's instrument failed); community (archive-listed): 6 samples from 1 repository, 4 ran (4 with no instrument failure; 0 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

hoagyc/sparse_coding officialmentioned in papermentioned on GitHubpytorchSyntology: 5 samples, 4 ran (3 instrument-failed) · 1 unverified report
jbloomaus/saelens mentioned on GitHubjaxMITSyntology: 6 samples, 4 ran (0 instrument-failed) · 2 unverified report
openmoss/language-model-saes mentioned on GitHubjaxSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders (arXiv:2410.20526; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

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: 0 honoured, 1 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 · violated contract
3ran · our draft was wrong
4ran
3unverified

Licence: 5 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 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

hoagyc/sparse_coding official repository 5 samples, 4 ran (3 instrument-failed) · 1 unverified
ablate_feature_intervention standard_metrics.py ran · our draft was wrong no licence file found · pointer only · 5551196e92fa7236 · report fixture: signature model:callable/0/?, location:str/0/str, feature:tuple/1/int · returned function · level recorded 2026-09-24
collect_lite cluster_runs.py ran · violated contract no licence file found · pointer only · 03016136842a1f30 · report fixture: signature processes:list/1/tuple · returned bool · level recorded 2026-09-24
filter_learned_dicts big_sweep.py ran · our draft was wrong no licence file found · pointer only · 99158f9a6ffb0e31 · report fixture: signature learned_dicts:list/1/tuple, hyperparam_filters:dict/0/? · returned list · level recorded 2026-09-24
get_model_tensor_name standard_metrics.py ran · our draft was wrong no licence file found · pointer only · 200b3f8d8c27361c · report fixture: signature location:tuple/1/int · returned str · level recorded 2026-09-24
calc_expected_interference big_sweep.py unverified no licence file found · pointer only · 4d866dd8ed7f5121 · report
jbloomaus/saelens community (archive-listed) 6 samples, 4 ran (0 instrument-failed) · 2 unverified
get_eval_everything_config sae_lens/evals.py ran MIT (permissive) · 82ff8a8d62edb55d · report fixture: returned EvalConfig · level recorded 2026-09-24
get_special_token_from_cfg sae_lens/pretokenize_runner.py ran MIT (permissive) · 568742d62c1af63c · report fixture: signature cfg_token:str/0/str, tokenizer:MinimalTokenizer/0/? · returned int · level recorded 2026-09-24
json_dict sae_lens/config.py ran MIT (permissive) · 04cd1bc157c2e856 · report fixture: signature s:str/0/str · returned dict · level recorded 2026-09-24
special_token sae_lens/config.py ran MIT (permissive) · aac5ca4eeebb7d88 · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-25
get_sae_class sae_lens/registry.py unverified MIT (permissive) · 52c2cc2bd7a7d6e7 · report
get_sae_training_class sae_lens/registry.py unverified MIT (permissive) · 782e1939a5b1737f · report

Tasks

Language ModellingModel Editing

1 archive task tag without a task page not shown.

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