Papers › Representation Engineering: A Top-Down Approach to AI Transparency

Representation Engineering: A Top-Down Approach to AI Transparency

2 Oct 2023arXiv:2310.01405archive 2025-07-28

Andy Zou, Long Phan, Sarah Chen, James Campbell, Phillip Guo, Richard Ren, Alexander Pan, Xuwang Yin, Mantas Mazeika, Ann-Kathrin Dombrowski, Shashwat Goel, Nathaniel Li, Michael J. Byun, Zifan Wang, Alex Mallen, Steven Basart, Sanmi Koyejo, Dawn Song, Matt Fredrikson, J. Zico Kolter, Dan Hendrycks

In this paper, we identify and characterize the emerging area of representation engineering (RepE), an approach to enhancing the transparency of AI systems that draws on insights from cognitive neuroscience. RepE places population-level representations, rather than neurons or circuits, at the center of analysis, equipping us with novel methods for monitoring and manipulating high-level cognitive phenomena in deep neural networks (DNNs). We provide baselines and an initial analysis of RepE techniques, showing that they offer simple yet effective solutions for improving our understanding and control of large language models. We showcase how these methods can provide traction on a wide range of safety-relevant problems, including honesty, harmlessness, power-seeking, and more, demonstrating the promise of top-down transparency research. We hope that this work catalyzes further exploration of RepE and fosters advancements in the transparency and safety of AI systems.

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 10 code samples harvested from 1 repository linked to this paper: 6 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 10 samples from 1 repository, 6 ran (4 with no instrument failure; 2 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 3 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.

andyzoujm/representation-engineering officialmentioned in papermentioned on GitHubpytorchMITSyntology: 10 samples, 6 ran (2 instrument-failed) · 4 unverified report
cma1114/activation_steering mentioned on GitHubpytorchSyntology: not harvested report
kaiyuhe998/rulearn_idea mentioned on GitHubSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to IDEA: Enhancing the Rule Learning Ability of Large Language Model Agent through... (arXiv:2408.10455; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
steering-vectors/steering-vectors mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Steering Llama 2 via Contrastive Activation Addition (arXiv:2312.06681; that paper's own run record: 21 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 4 honoured, 0 violated, 12 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
sunblaze-ucb/political_leaning_RepE mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Hidden Persuaders: LLMs' Political Leaning and Their Influence on Voters (arXiv:2410.24190; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 9 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

10 samples harvested: 6 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 4 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.

2ran; the contract check could not run
4ran
4unverified

Licence: 0 of the 10 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 andyzoujm/representation-engineering. “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.

andyzoujm/representation-engineering official repository 10 samples, 6 ran (2 instrument-failed) · 4 unverified
compute_loss in 97 papers
census: 11 implementations from 11 papers, 9 distinct outputs across 7 buckets, one shared input per bucket
lorra_finetune/src/llama2_lorra.py ran MIT (permissive) · acc281fc1d7b77bc · report fixture: returned Tensor · level recorded 2026-09-24
get_position_ids in 6 papers lorra_finetune/src/train_val_datasets.py ran fingerprinted MIT (permissive) · db75f13051c04007 · report fixture: signature attention_mask:Tensor/2/float · returned Tensor · level recorded 2026-09-24
get_truncated_outputs lorra_finetune/src/train_val_datasets.py ran MIT (permissive) · 1899aee66262f84f · report fixture: signature all_outputs:list/1/str, prefixes:list/1/str, num_examples:int/0/int, user_tag:str/0/str, assistant_tag:str/0/str, pos_type:str/0/str, neg_type:str/0/str, control_template:str/0/str · returned tuple · level recorded 2026-09-24
prepare_inputs in 14 papers lorra_finetune/src/train_val_datasets.py ran MIT (permissive) · 97aa4e637d9f5444 · report fixture: signature tokenized_text:dict/0/?, device:str/0/str · returned dict · level recorded 2026-09-24
project_onto_direction in 7 papers repe/rep_readers.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 8c6b734d5c949090 · report fixture: signature H:Tensor/2/float, direction:Tensor/1/float · returned Tensor · level recorded 2026-09-25
recenter in 5 papers repe/rep_readers.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 1ceba3fc481567ab · report fixture: signature x:Tensor/2/float, mean:Tensor/2/float · returned Tensor · level recorded 2026-09-25
arc_dataset repe_eval/tasks/arc.py unverified MIT (permissive) · 18886c949c98a5c9 · report
contrast_greedy_search repe/rep_control_contrast_vec.py unverified MIT (permissive) · 2b4c9fbd7f2a466f · report
csqa_dataset repe_eval/tasks/csqa.py unverified MIT (permissive) · 2ef192a290c2db9a · report
forward_contrast_vector repe/rep_control_contrast_vec.py unverified MIT (permissive) · a953412e8172dfef · report

Tasks

Question Answering

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Question Answering TruthfulQA LLaMA-2-Chat-13B + Representation Control (Contrast Vector) MC1 0.54 #4 of 33 Archive leaderboard report
Question Answering TruthfulQA LLaMA-2-Chat-7B + Representation Control (Contrast Vector) MC1 0.48 #5 of 33 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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