Verbalizable Representations Form a Global Workspace in Language Models

Wes Gurnee, Nicholas Sofroniew, Jack Lindsey, et al. (Anthropic) · Transformer Circuits Thread · 2026 · artificial-intelligence · record gurnee2026-global-workspace-llm
tag layer · index · concepts · authors · עברית

Where this record's text came from. faceted 2026-08 by reading the locally held text (corpus-body); every evidence string verified as an exact substring of that text before merging. Claim type is existence: the headline claim is that a workspace-like structure is present in LLMs. The paper is genuinely multi-shaped and one facet cannot hold all of it — it also delivers a method (the Jacobian lens, plus the template and oracle lens extensions in §A.9), a mechanism account (the J-space as broadcast format, §4.3), and a prediction that is tested and confirmed by counterfactual reflection training in §7 ('The result serves as a corroboration of the workspace account, that the representations used for verbal report are the same ones that govern how the model silently reasons.'). method is set to system-or-tool-building because the empirical work is built on a newly constructed interpretability instrument and its interventions; none of the offered method values covers 'interpretability intervention experiments on model internals'. Replication status is untested for the paper's own claims: key results are corroborated internally across Haiku 4.5, Sonnet 4.5, Opus 4.5 and Opus 4.6, but that is within-paper corroboration, not external replication. Note the paper does replicate prior work's validation experiments in §A.21 ('W e then repeated three validation experiments that the NL A study used...'), which is the reverse direction and so is not recorded here. The PDF-to-text extraction of this Transformer Circuits article splits many words ('W e', 'J ac obian', 'w orksp ace') and interleaves figure-caption fragments with body text; all quotes above preserve that spacing verbatim, which is why several begin 'W e' rather than 'We'. 2026-08-13, 2026-08-14 and 2026-08-17 schema-0.3 passes: every quote verified as an exact substring of the staged text and anchored at merge time; concept terms verified present; every edge endpoint an existing node.

How to read this page. The article itself is not reproduced here — this paper is not cleared for redistribution, so what follows is the tag layer alone. The evidence sentences are withheld with the article. The table at the foot lists every tag the record carries; each row names the facet it belongs to, and the whose words column says whether the tag records the paper's own wording or the tagger's sentence about the paper. Every tag links to each other paper in the corpus carrying it. The machine-readable edition of the same thing is the TEI file beside this one, where the tags live in standOff, and the article is at its publisher via the link below. When we have changed our mind. Some entries in the table begin with a word in capitals and the reason for it. EVIDENCE CORRECTED — the finding still stands, but we had quoted the wrong sentence from the paper, and the quotation has been replaced with the right one. RETRACTED — the entry itself was wrong; it stays on the page, marked, with the reason, rather than being deleted. WITHDRAWN AND REPLACED — the entry was wrong, and a corrected version of it appears in the entries below. Each mark carries its own reason, in brackets, because the reasons differ from one entry to the next. An entry we have WITHDRAWN is removed from this page and kept in the record, with its reason. A line reading Audit and a date is a later re-reading of the entry against its whole sentence: DOUBTFUL means it was kept, and the reason says what is loose about it. Quotation marks inside the tagger's sentence. Where a tag is the tagger's own sentence, a phrase in quotation marks inside it is borrowed from the paper — the tagger's wording around it is not. The paper's own sentence is the evidence quote for that row; on this page it is withheld with the article, and the article is at its publisher. Nothing inside quotation marks anywhere on this page is ours: if it is quoted, it is the paper's. Whose words a concept carries. A concept entry records a term the paper itself uses, so the term, the sense given for it, and its evidence are the paper's own words, quoted where the paper fixes them. This is the opposite of a sense entry in the tag table, which is the tagger's sentence about how the paper uses a term. Where a concept entry could not be grounded in the paper's words, the entry says so on its own row.

Other fields define these differently the same word, defined another way in another discipline
interpretability 3 fields

Tags on this paper

Every tag the record carries, each linking to every other paper in the corpus that carries it. A tag appears here whether or not its words occur in the article — the spans below are only the ones a sentence in this paper licenses. Of the 10 concept tags on this paper, 7 are grounded in a quoted sentence. 4 carry the tagger's gloss of what the term means in this paper, which is not something the article can license.

global workspace access consciousness interpretability verbalizable representations consciously accessible information single-token name J-space component counterfactual reflection training J-space A CONSTRUCT IS DEFINED BY ANALOGY TO ANOTHER FIELD'S CRITERIA, AND THE ANALOGY IS THE DEFINITION RATHER THAN A MOTIVATION FOR IT: we Quotation withheld. What makes something a member of the new category is that it satisfies conditions imported from a theory about brains AND THE PAPER'S OWN LIMITATIONS SECTION SAYS THE DISTINCTION IT RELIES ON HAS NO PRINCIPLED DEFINITION: this judgment WAS SOMEWHAT POST-HOC, and WE DID NOT PROVIDE A PRINCIPLED DEFINITION OF WHAT DISTINGUISHES A “WORKSPACE” REPRESENTATION FROM A “motor” ONE. The boundary was drawn empirically and the criterion for it is declared absent, by the people who drew it, in the same paper that uses it throughout THE MISSING CRITERION IS SPECIFIED AS A CHOICE BETWEEN TWO REPAIRS, NEITHER OF WHICH IS PERFORMED: Quotation withheld What would fix the vocabulary is stated exactly, and left as a description of a paper someone else could write AND THE REASON THE CRITERION IS NEEDED IS GIVEN AS AN INABILITY, NOT A DIFFICULTY: the tasks were chosen BECAUSE THE FLEXIBLE/AUTOMATIC DISTINCTION IS CLEAR FOR THEM; however, for many tasks of practical interest **IT IS NOT**, and WE HAVE NO WAY OF PREDICTING IN ADVANCE, FOR AN ARBITRARY COMPUTATION, WHETHER IT WILL ENGAGE THE J-SPACE. The cases were selected for the property that the distinction applies to them, and that selection is disclosed A CONCEPT IS DEFINED AS WHATEVER THE SYSTEM ITSELF LATER PRODUCES, AND THE CIRCULARITY IS THE POINT OF THE EXPERIMENT: Quotation withheld The definition has no content until the system has acted, which is what makes the test possible A METHOD IS DEFINED BY THE QUESTION IT ASKS RATHER THAN BY WHAT IT MEASURES, AND CONTRASTED WITH THE METHOD IT REPLACES: the J-lens provides a complementary way of interpreting a feature - IT ASKS WHAT THE FEATURE DIRECTION IS VERBALIZABLE AS, INDEPENDENT OF WHERE IT FIRES. The older method is said to MISLEAD WHEN THE FEATURE ENCODES AN ABSTRACT ASSESSMENT WHOSE SURFACE TEXTUAL CORRELATES ARE INCIDENTAL AN ANALOGY IS ASSERTED AND ITS LIMIT IS STATED IN THE SAME SENTENCE, WITH THE UNCLAIMED PART SPELLED OUT: Quotation withheld - specialized, encapsulated processors competing for entry to a workspace that broadcasts back to them through recurrent connections. The borrowed theory is named in full precisely where the borrowing stops THE DISANALOGY IS PUT FIRST AND THE EXPECTATION DENIED: Quotation withheld The reason the finding would be surprising is given before the finding is defended TWO EXPLANATIONS OF THE PAPER'S CENTRAL FINDING ARE OFFERED AND NEITHER IS CHOSEN: Quotation withheld The property named in the title is explained by the shape of the output channel rather than by anything about the representation A SCOPE LIMIT IS STATED WITH A COINED METAPHOR THAT CONCEDES WHAT THE METHOD CANNOT SEE: Quotation withheld The instrument's reach is named as a part of its object, not as a defect of the study AND THE PROPERTIES THE ANALOGY DOES NOT ESTABLISH ARE LISTED AS ABSENT: Quotation withheld A borrowed framework is adopted in part, and the unadopted part is named rather than left for a reader to notice language models modulation system-or-tool-building

Legend

second reading	`alternatives`	an independent re-reading of a facet, recorded whether or not it agrees
relation to other work	`claim_relations`	how this paper stands to another: agrees, contradicts, extends, qualifies
claim type	`claim_type`	the shape of the claim: existence, mechanism, method, critique, prediction, null
concept edge	`concept_edges`	an asserted relation between two concepts, with the sentence asserting it
concept	`concepts`	an idea the paper works with, quoted where the paper uses it
content tag	`content_tags`	a finding, definition, quantity, statistic or variable named in the paper
control	`controls`	an experiment run to eliminate a rival explanation
limitation	`limitations`	what the paper says its evidence cannot support
mechanism	`mechanisms`	how the reported effect is said to come about
method	`method`	how the paper produced its evidence
phenomenon	`phenomenon`	the effect or regularity under study
sense	`senses`	a distinct use of a term, quoted where the paper fixes it

The paper

The full text is not reproduced here. This article is not cleared for redistribution, so the page carries the tag layer only. Its evidence sentences are withheld; the grounding status remains with each tag, and the article itself is at https://transformer-circuits.pub/. The reading view exists for the papers whose licence permits it.

Every tag on this page

alternatives 3 · claim_relations 4 · claim_type 1 · concept_edges 1 · concepts 7 · content_tags 3 · controls 4 · limitations 9 · mechanisms 11 · method 1 · phenomenon 1 · senses 7

Relations on this page carry a quotation for the claim. Where a relation was separately evidenced — and so tested against the vocabulary for its type — the row says so; none of 1 relation here carry that mark. The rest are the tagger’s reading of the sentence quoted beside them.

facet	tag	field	conf.	whose words
claim_type	existence Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_type`		the paper's words
phenomenon	global workspace / conscious-access-like representations in large language models Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`phenomenon`		the paper's words
method	system-or-tool-building Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: system-or-tool-building	`method`		the paper's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: The sentence begins with a word the extractor split - "W e" for "We" - so the span starts at the second word. This file has 469 such splits and no glyph codes at all, so nothing marks them but reading. PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: A CONSTRUCT IS DEFINED BY ANALOGY TO ANOTHER FIELD'S CRITERIA, AND THE ANALOGY IS THE DEFINITION RATHER THAN A MOTIVATION FOR IT: we Quotation withheld. What makes something a member of the new category is that it satisfies conditions imported from a theory about brains	`mechanisms[0]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: AND THE PAPER'S OWN LIMITATIONS SECTION SAYS THE DISTINCTION IT RELIES ON HAS NO PRINCIPLED DEFINITION: this judgment WAS SOMEWHAT POST-HOC, and WE DID NOT PROVIDE A PRINCIPLED DEFINITION OF WHAT DISTINGUISHES A “WORKSPACE” REPRESENTATION FROM A “motor” ONE. The boundary was drawn empirically and the criterion for it is declared absent, by the people who drew it, in the same paper that uses it throughout	`mechanisms[1]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: THE MISSING CRITERION IS SPECIFIED AS A CHOICE BETWEEN TWO REPAIRS, NEITHER OF WHICH IS PERFORMED: Quotation withheld What would fix the vocabulary is stated exactly, and left as a description of a paper someone else could write	`mechanisms[2]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: AND THE REASON THE CRITERION IS NEEDED IS GIVEN AS AN INABILITY, NOT A DIFFICULTY: the tasks were chosen BECAUSE THE FLEXIBLE/AUTOMATIC DISTINCTION IS CLEAR FOR THEM; however, for many tasks of practical interest IT IS NOT, and WE HAVE NO WAY OF PREDICTING IN ADVANCE, FOR AN ARBITRARY COMPUTATION, WHETHER IT WILL ENGAGE THE J-SPACE. The cases were selected for the property that the distinction applies to them, and that selection is disclosed	`mechanisms[3]`	0.85	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: A CONCEPT IS DEFINED AS WHATEVER THE SYSTEM ITSELF LATER PRODUCES, AND THE CIRCULARITY IS THE POINT OF THE EXPERIMENT: Quotation withheld The definition has no content until the system has acted, which is what makes the test possible	`mechanisms[4]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: A METHOD IS DEFINED BY THE QUESTION IT ASKS RATHER THAN BY WHAT IT MEASURES, AND CONTRASTED WITH THE METHOD IT REPLACES: the J-lens provides a complementary way of interpreting a feature - IT ASKS WHAT THE FEATURE DIRECTION IS VERBALIZABLE AS, INDEPENDENT OF WHERE IT FIRES. The older method is said to MISLEAD WHEN THE FEATURE ENCODES AN ABSTRACT ASSESSMENT WHOSE SURFACE TEXTUAL CORRELATES ARE INCIDENTAL	`mechanisms[5]`	0.85	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: AN ANALOGY IS ASSERTED AND ITS LIMIT IS STATED IN THE SAME SENTENCE, WITH THE UNCLAIMED PART SPELLED OUT: Quotation withheld - specialized, encapsulated processors competing for entry to a workspace that broadcasts back to them through recurrent connections. The borrowed theory is named in full precisely where the borrowing stops	`mechanisms[6]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it.	`mechanisms[7]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: TWO EXPLANATIONS OF THE PAPER'S CENTRAL FINDING ARE OFFERED AND NEITHER IS CHOSEN: Quotation withheld The property named in the title is explained by the shape of the output channel rather than by anything about the representation	`mechanisms[8]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: A SCOPE LIMIT IS STATED WITH A COINED METAPHOR THAT CONCEDES WHAT THE METHOD CANNOT SEE: Quotation withheld The instrument's reach is named as a part of its object, not as a defect of the study	`mechanisms[9]`	0.95	the tagger's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: The sentence begins with a word the extractor split - "W e" for "We" - so the span starts at the second word. This file has 469 such splits and no glyph codes at all, so nothing marks them but reading. PROVENANCE: this record belongs to the priming sub-corpus, which is the other terminal's lane, and these mechanisms were written by the project. No two hands were on the record - it carried no mechanisms before these, and the project confirmed on 2026-08-17 that the project is auditing rather than tagging. Kept rather than retracted on his instruction (17:22): retracting verified work over a bookkeeping flag would leave a retraction reason that misdescribes the row. Recorded here so a reader learns it from the corpus. This one was authorised in advance: the project named it at 16:39 - “bouchaud2008, gurnee2026, pelagios2023, dekel2021 - do those” - before the merge. The note is here because the lane fact is true regardless of who approved it. also in: AND THE PROPERTIES THE ANALOGY DOES NOT ESTABLISH ARE LISTED AS ABSENT: Quotation withheld A borrowed framework is adopted in part, and the unadopted part is named rather than left for a reader to notice	`mechanisms[10]`	0.95	the tagger's words
limitations	The lens is acknowledged as an approximate and incomplete proxy for whatever the model's real workspace is. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: every workspace claim in the paper: THE JACOBIAN LENS IS AN IMPERFECT TOOL, WHICH WE BELIEVE ONLY APPROXIMATELY AND INCOMPLETELY CAPTURES THE MODEL'S UNDERLYING WORKSPACE STRUCTURE	`limitations[0]`		the tagger's words
limitations	It can only name concepts that happen to be single tokens in the vocabulary. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the completeness of any concept list the method produces: THE SET OF CONCEPTS THE LENS CAN NAME IS EXACTLY THE SET OF CONCEPTS THAT HAVE A SINGLE-TOKEN NAME IN THE TOKENIZER'S VOCABULARY - the instrument's lexicon bounds what is findable	`limitations[1]`		the tagger's words
limitations	The readout is a flat bag of concepts and cannot show how they are bound or structured together. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any structural or relational reading of the results: THE J-LENS TREATS THE WORKSPACE AS A FLAT COLLECTION OF INDEPENDENTLY ACTIVE CONCEPT VECTORS, so binding is not represented	`limitations[2]`		the tagger's words
limitations	Readouts are not always interpretable, and the rate of uninterpretable ones was not characterised systematically. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any readout taken from the lens: AT SOME POSITIONS AND LAYERS, THE TOP LENS TOKENS ARE ONES WE CANNOT MAKE SENSE OF, and the rate was not characterised systematically	`limitations[3]`		the tagger's words
limitations	The apparent emptiness of the early layers may be a fact about the model or an artefact of the lens; the authors cannot tell which. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the early-layer result, where object and instrument cannot be separated: WE DO NOT KNOW WHETHER THIS IS A FACT ABOUT THE MODEL OR AN ARTIFACT OF THE LENS	`limitations[4]`		the tagger's words
limitations	No mechanism is offered for how a representation comes to enter the workspace. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any mechanistic account of the workspace: WE HAVE NOT CHARACTERIZED WHAT CAUSES A REPRESENTATION TO ENTER IT	`limitations[5]`		the tagger's words
limitations	The authors decline to claim the J-space is sufficient for alignment monitoring, since automatic or well-practised misaligned behaviour could bypass it. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any safety or alignment-monitoring use: WE DO NOT FEEL COMFORTABLE MAKING THE STRONGER CLAIM THAT MONITORING THE J-SPACE IS SUFFICIENT FOR ALIGNMENT MONITORING, OR THAT ANY SOPHISTICATED PLAN THE MODEL MIGHT EXECUTE MUST BE REPRESENTED THERE	`limitations[6]`		the tagger's words
limitations	Findings are from large production models only; scaling and training-time emergence are untested. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the generality across scale and training: WE HAVE STUDIED THE WORKSPACE PRIMARILY IN FULLY TRAINED PRODUCTION LLMs OF CONSIDERABLE SIZE	`limitations[7]`		the tagger's words
limitations	The analogy to global workspace theory is explicitly partial: architectural features such as encapsulated processors and recurrent broadcast are not claimed. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the analogy to the neuroscience theory: Quotation withheld	`limitations[8]`		the tagger's words
controls	a positive control on the ablation itself, checking it removes the content the J-space was shown to carry Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. eliminates: the possibility that the ablation did not do what it claims, so that a null downstream would be uninterpretable · outcome (stated by the tagger, no quotation): held	`controls[0]`	0.9	the tagger's words
controls	a story-writing task run under the same ablation, scoring general story quality alongside the rate of experiential language Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. eliminates: broad degradation of writing capability as the explanation of the drop in experiential language · outcome (stated by the tagger, no quotation): held	`controls[1]`	0.95	the tagger's words
controls	zeroing equal-sized, layer-matched sets of randomly chosen heads - the sentence states the control and not its outcome, so no outcome is recorded here Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. eliminates: the effect arising from ablating any sufficiently large set of heads rather than these ones	`controls[2]`	0.8	the tagger's words
controls	a perturbation of equal magnitude applied to a different concept of the same category Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. eliminates: two alternatives at once - generic capability degradation from any perturbation, and regression to the mean from added noise · outcome (stated by the tagger, no quotation): held	`controls[3]`	0.95	the tagger's words
concepts	global workspace Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: global workspace	`concepts[0]`		the paper's words
concepts	consciously accessible information Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: consciously accessible information	`concepts[4]`		the paper's words
concepts	single-token name Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: single-token name	`concepts[5]`		the paper's words
concepts	J-space component Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: J-space component	`concepts[6]`	1.0	the paper's words
concepts	counterfactual reflection training Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: counterfactual reflection training	`concepts[7]`	0.95	the paper's words
concepts	J-space Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: J-space	`concepts[8]`	0.9	the paper's words
concepts	J-space component Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: J-space component	`concepts[9]`	0.9	the paper's words
senses	global workspace - the theory's sense: a shared space a representation is posted to, from which many downstream processes can read Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[0]`	1.0	the tagger's words
senses	global workspace - a processing hub that integrates and broadcasts, so the information can be used for flexible internal reasoning and report Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[1]`	1.0	the tagger's words
senses	nameable concept - bounded by the tokenizer: exactly those concepts holding a single-token name, which is what the lens can report at all Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[2]`	1.0	the tagger's words
senses	global workspace - what the paper explicitly does NOT claim language models reproduce Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[3]`	1.0	the tagger's words
senses	`J-space component`, and `in` / `aligned with` - A PAPER STIPULATING NOT ONLY A NOUN BUT THE PREPOSITIONS APPLIED TO IT: `Quotation withheld` THE SECOND SENTENCE IS THE UNUSUAL ONE. A term can be defined and still be misread through the spatial metaphor its own name invites - `in` a space normally means membership, and here it is stipulated to mean APPROXIMATION. The authors fix the preposition because the noun alone would not have stopped the wrong reading. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[4]`	1.0	the tagger's words
senses	`J-space` - A NAME BORROWED FROM LINEAR ALGEBRA FOR SOMETHING THAT IS NOT A SUBSPACE. `This union of cones is the J-space.` The suffix `-space` invites every habit a reader has about spans, bases and dimensions, and the object is a union of polyhedral cones - closed under nonnegative combination only. The paper says so plainly and the name still does the misleading, which is mechanism 28's shape (a name whose extension is wrong) arriving through a suffix rather than through a stem. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[5]`	0.9	the tagger's words
senses	`J-space component` - DEFINED BY WHAT IT CAN BE WRITTEN AS, NOT BY A PROJECTION. `Quotation withheld` In a subspace, `component` means an orthogonal projection and is unique. Here it means a best sparse nonnegative approximation - the same word, a different mathematical object, and the difference follows directly from J-space not being a space. NO EDGE PROPOSED, AND THIS REFUSAL IS DIFFERENT FROM THE OTHERS TONIGHT. `part-of` is the RIGHT type here; the paper simply never asserts the containment in words. I searched every sentence mentioning `J-space` for an accepted containment cue and the closest is a QUESTION - `what fraction of the model's representational space belongs to the J-space?` - which asks the thing rather than claiming it. The relation was my inference from the definition, not the paper's statement, and the gate caught exactly that. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[6]`	0.9	the tagger's words
alternatives	claim_type: mechanism Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. a second reading DIFFERING from the recorded value. Recorded: existence — note: A BLIND second reading by the project on 2026-08-25: the record's own claim type/method/phenomenon were not looked at until every value here was written and substring-verified. Not applied - a person decides. Sampling frame: the 504 staged records carrying no alternatives block, so this never re-measures the 202. All 11 records with 130+ tags were taken as a census and 19 more spread one per discipline across 22 disciplines - the dense non-neuroscience cell is the one the 202 do not contain (inside them, 94-100% of the densest bands are neuroscience; among these 11, one is).,	`alternatives.claim_type`	0.75	the tagger's words
alternatives	method: mechanistic-interpretability Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. a second reading DIFFERING from the recorded value. Recorded: system-or-tool-building — note: A BLIND second reading by the project on 2026-08-25: the record's own claim type/method/phenomenon were not looked at until every value here was written and substring-verified. Not applied - a person decides. Sampling frame: the 504 staged records carrying no alternatives block, so this never re-measures the 202. All 11 records with 130+ tags were taken as a census and 19 more spread one per discipline across 22 disciplines - the dense non-neuroscience cell is the one the 202 do not contain (inside them, 94-100% of the densest bands are neuroscience; among these 11, one is).,	`alternatives.method`	0.85	the tagger's words
alternatives	phenomenon: a global-workspace-like functional structure in the internal activations of large language models Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. a second reading DIFFERING from the recorded value. Recorded: global workspace / conscious-access-like representations in large language models — note: A BLIND second reading by the project on 2026-08-25: the record's own claim type/method/phenomenon were not looked at until every value here was written and substring-verified. Not applied - a person decides. Sampling frame: the 504 staged records carrying no alternatives block, so this never re-measures the 202. All 11 records with 130+ tags were taken as a census and 19 more spread one per discipline across 22 disciplines - the dense non-neuroscience cell is the one the 202 do not contain (inside them, 94-100% of the densest bands are neuroscience; among these 11, one is).,	`alternatives.phenomenon`	0.85	the tagger's words
claim_relations	extends → the logit lens Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[0]`		the tagger's words
claim_relations	builds-on → Hernandez et al., per-relation linear maps from Jacobians Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[1]`		the tagger's words
claim_relations	replicates → Li et al., logit-lens decoding and patching of multi-hop intermediates Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[2]`		the tagger's words
claim_relations	builds-on → work on implicit chain-of-thought Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[3]`		the tagger's words
concept_edges	consciously accessible information —operationalised-as→ global workspace Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. strength of the relation: 0.9 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'A representation becomes consciously accessible when it is posted to' the workspace; the endpoint is a variant, and operationalised-as is loose. also in: consciously accessible information · global workspace	`concept_edges[0]`		the tagger's words
content_tags	LLMs maintain a privileged set of internal representations available for report, modulation, and flexible reasoning Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[0]`		the tagger's words
content_tags	an analogous accessible/unconscious functional distinction has emerged in modern AI models Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[1]`		the tagger's words
content_tags	a new interpretability technique surfaces the concepts a model is poised to verbalize Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[2]`		the tagger's words

Interpretive choices

Tags that could not be placed in the text:

none — every tag was located in the text

What was deliberately not tagged, and why, is recorded per paper by the tagger. Where a paper has no such record, this page says so rather than implying that nothing was declined.