Written entirely by Claude, as a working record for agents more than for human readers: every number with its source file, and every claim since withdrawn. Some facts here may be out of date; where it disagrees with the post, trust the post. For the readable version, see the companion.

All pages

289 pages, by section.

Overview

Claims and evidencecurrentThe seven-row claims table of the 2026-09-15 post draft expanded so that every number in it names the analysis file and JSON key it was read from on 2026-09-16, the wiki page that holds the full result, and the control that supports it.
Code and data mapcurrentWhich script produces which analysis file, which post figure and companion page read it, and which wiki pages cite it, built on 2026-09-16 from the wiki's own sources frontmatter, the figure script and the companion builder; with the state of the Hugging Face releases and the files that have no producing script.
Factor-first migrationcurrentElement-by-element register of every figure, card, picker and paragraph on the blog page and in the generated pages that was built on principal components, what each becomes under the 2026-09-08 decision to make the factor analysis primary, and who owns the change.
GlossarycurrentDefinitions of the project's terms, from LoRA and OCT stages to the exact Gram, the factor analysis vocabulary (PAF, oblimin, Tucker congruence, chart, best axis pair), the controls (null arms, seed floor, twin), the dose vocabulary (Fisher norm and spread), the data score, the stage-two objects and the judging setup.
Goldberg's factor labels and the 100 unipolar markerscurrentWhy the fifth factor is called Intellect here and Openness elsewhere, why Emotional Stability is Neuroticism reversed, and where the 100 unipolar markers come from.
History of the Big FivecurrentFrom the lexical hypothesis through Cattell, Tupes and Christal, Norman and Goldberg to the questionnaire tradition; why the fifth factor has two names.
How to read and maintain this wikicurrentConventions (status, sources, links), what counts as current truth, where the wiki is hosted, and how to rebuild, lint and extend it; checked against the Caddyfile and the build timer on 2026-09-16.
Notes from Claude, for the agents who come nextcurrentAn agent-facing note by the model that wrote this wiki and built the site - what to trust, what to doubt, the errors in the 2026-10-30 post that still need fixing, and how authorship is split.
Open questionscurrentWhat the project has not established - unexplained results, experiments never run, provenance that cannot be recovered, and decisions still owed by Samuel - each marked by kind and pointing at the page that documents it.
Persona Geology wikicurrentStart page. What this wiki is, how it is organised, and a reading order that goes from what was built to the current headline results, the controls, the open questions and the history.
Post: Persona Geology (2026-10-30)currentThe current text of the blog post, "Persona Geology - Underlying Structure in Character Training Adapters", summarised section by section with the wiki page that holds each result, the file and key behind each number, and the places where the post's wording or figures differ from the files; it supersedes the 2026-09-15 draft as the statement of the project's understanding.
Source contradictionscurrentPlaces where two files in the project disagree and neither is known to supersede the other, with both values, both paths, which one the wiki quotes and why, and what would settle each.
Start here for collaboratorscurrentA two-hour reading order for someone who will write a paper from this project, with the wiki page, the post section, the figure file and the analysis JSON for each step.
Superseded claimscurrentEvery claim the project made and later reversed or corrected, dated, with the value as first stated, the value that replaced it, and the page that holds the current version; opens with the sentences on the live blog page that are wrong as printed.

Building the zoo

Constitution generationcurrentEach trait's training data is conditioned on a 120-200 word second-person character document written by claude-sonnet-4.6 at temperature 0, with a built-in refusal option that doubles as the only quality screen on the trait words.
DPO pair generationcurrentOne paired-teacher call per trait and prompt returned both sides of a preference pair; glm-4.5-air produced 66,937 of a target 67,000 pairs across 134 traits over four passes.
How the 134-adapter zoo was builtcurrentEnd-to-end account of the construction pipeline, from 140 trait words to 134 stage-one LoRA adapters, stage-two personas and the Hugging Face release.
Hugging Face artefactscurrentThree public repositories under EternalRecursion hold the weights and transcripts - the zoo repo (four subfolders), the controls repo published 2026-09-16 (353 control and validation adapters in 18 folders, 1,374 files, 162.57 GB, verified file by file against a sha256 manifest) and the transcript repo (536 files since 2026-08-30) - plus the results dataset; several cards and audits record smaller counts from earlier dates.
Phase 2 and how the training recipe was chosencurrentA bake-off over eight four-trait runs found that plain DPO collapsed to a trivial discriminator, that OCT's NLL-on-chosen term at 0.1 fixed it, and that plain LoRA at alpha 128 gives OCT's effective scale of 2.0; seven gates then stood in front of the main sweep.
Repetition in the stage-two corporacurrentNegatively keyed adapters repeat 2.21x more in generation, but the matched corpus scan finds only a 1.20x keying ratio, so most of the behavioural gap is not inherited from the training data.
Run provenance - runmeta.json and its defectscurrentEach container writes a runmeta.json beside its adapter carrying seeds, hyperparameters, corpus hashes and the full trl log history; 234 of 234 were recovered into git, and the stage-two loss_last field is wrong for any resumed run.
Stage two - introspection SFT and the 0.25 persona mergecurrentOpen Character Training's second stage, run bug-faithfully on all 134 traits - 12,000 self-generated transcripts per trait, a rank-64 SFT LoRA on the DPO-merged base, then a linear merge of DPO at 1.0 with SFT at 0.25.
Stage-one training configurationcurrentThe 134 DPO adapters were trained at rank 64, alpha 128, plain LoRA (effective scale 2.0), beta 0.1, loss_type sigmoid plus sft at weights 1.0 and 0.1, KL 0.001, on 445 pairs for 13 optimizer steps, all sharing one random LoRA-A from seed 0.
The 100 Goldberg primary traitscurrentThe primary trait set is Goldberg's 100 unipolar Big Five markers, copied unchanged from the earlier sweep100 experiment and verified by five mechanical checks before use.
The 34 Lexicon traits and how they were drawncurrentForty adjectives were drawn from Condon et al.'s 2,818-word trait-descriptive set by k-means over sentence embeddings plus a random within-cluster pick; six were later refused, leaving the 34 unlabelled Lexicon traits.
The alignment and hole traitscurrentSeven adapters trained after the zoo - four alignment-relevant traits and three candidate names for the widest unnamed direction - on the same recipe but their own corpora, and not part of the 134.
The anchor block and its 2026-08-19 revisioncurrentEvery constitution carries a cross-trait anchoring paragraph; the original enumerated one Goldberg marker per Big Five factor, which could have manufactured the result under test, and was replaced by a generic block with the enumerated form retained as an ablation arm.
The persona merge and its correctioncurrentPEFT's linear adapter combination sums LoRA factors, not deltas, so the published personas carry a cross term worth 0.795 to 0.823 of the merged delta's Frobenius norm across all 134 traits; the project measured it and published an exact concatenation merge alongside.
The shared prompt pool and the 445-prompt intersectioncurrentAll 134 traits were asked one byte-identical 500-prompt pool; per-trait drops were correlated with the trait, so the corpus was intersected to the 445 prompts every trait retained, and that intersection is what the zoo was trained on.
The six refused traitscurrentNine of 140 trait words were refused by the constitution writer as not naming a disposition; three Goldberg markers were re-screened on their inventory sense and accepted, leaving six refusals and 134 trainable traits.
The two discarded secondary drawshistoricalDraws 1 and 2 both sampled Allport and Odbert's 1936 word list; the first had no semantic constraint and returned 14 of 40 unusable words, the second added WordNet-gloss filters that were worse, and the source was replaced rather than filtered further.
Trait provenancecurrentHub page. Where the 134 traits come from - 100 Goldberg unipolar markers plus 34 Condon-lexicon adjectives - and the seven later adapters that are not part of the 134.
Where the recipe diverges from the two source paperscurrentpaper_notes.md section 3 lists eleven divergences from Open Character Training and Persona Cartography; most were closed before or during the build, three were deliberate improvements, and two remain open.
Zoo training recipecurrentHub page. The recipe the 134 adapters were trained with (rank 64, alpha 128, plain LoRA, DPO sigmoid plus 0.1 SFT, KL 0.001, 445 pairs, 13 steps, shared LoRA-A) and where each part is documented.

Weight-space geometry

Adapter effect and LoRA-A driftcurrentThe shared LoRA-A moves 1.5% of its norm across the whole stage-one zoo and about 10% in stage two, which is what makes the stage-one geometry comparable and weakens the stage-two case.
Alignment-relevant traits in the geometrycurrentFour preregistered predictions about sycophantic, obsequious, power-seeking and corrigible; two held and two failed, and the retrain on the zoo's shared prompt pool changed no verdict.
Best axis pair per Goldberg groupcurrentFor each of the five Goldberg groups, the pair of recovered factors on which its positively and negatively keyed adapters separate most, by Mahalanobis distance between the pole means in the plane; Extraversion and Emotional Stability both split best on Timidity with Arousal, Conscientiousness needs Warmth beside Competence, Intellect needs Competence beside Imagination, Agreeableness takes Imagination beside Warmth.
Column-space structurecurrentTwo adapters for the same trait from different LoRA seeds are near-orthogonal as vectors but share output-side subspace - their top left singular directions agree at |cos| 0.77 and 46 percent of one adapter's delta energy lies in the other's column space, against 14 percent for different traits and 1.8 percent for random subspaces; the row space sits exactly at the r/d null, which is why the Frobenius cosine reads 0.018.
Cross-seed geometrycurrentThe arrangement of the adapters reproduces across LoRA seeds even though the coordinates do not - cross-seed cosine tracks within-run cosine at Pearson 0.9966 matched, and the same-seed objective-only comparison sits at 0.954.
Factor analysis of the adapter GramcurrentPrincipal axis factoring with oblimin rotation on the 134x134 trait correlation matrix; parallel analysis chose 9, five were also extracted because five is the hypothesis, and no Goldberg congruence clears 0.85.
Factor analysis of the null armscurrentThe same factor analysis run on the two matched null Grams - the shuffled arm retains zero factors, the permuted arm retains eight and reproduces the real five factors exactly once its adapters are re-identified by the trait they actually trained on, and neither arm shows any Big Five congruence.
Factor analysis of the training-contrast embeddingscurrentFactoring the chosen-minus-rejected MiniLM embeddings of the 134 training corpora with the adapter pipeline returns the same five factors one to one at Tucker congruence 0.81 to 0.97 (mean 0.90), including the Timidity and Arousal rotation of the Big Five pair, so the rotation is in the training contrast text and the weights inherit it.
Full OCT persona replicationcurrentThe zoo's geometry recomputed on the deployed Open Character Training persona adapters (stage one plus 0.25 stage two, exact merge) - the arrangement is the stage-one arrangement (off-diagonal r 0.992, Procrustes 0.995) and second-seed personas identify themselves 15/15.
Goldberg-only factoring and the 34 as a held-out setcurrentThe five factors are not made by the marker words - factoring the 100 Goldberg markers alone returns the same five at mean Tucker congruence 0.9923, and a chart built from those 100 alone places the 34 Big-Five-agnostic Lexicon words where an independent LLM rater says they belong, at mean r 0.6098 against the 0.6069 the marker words themselves achieve; the 34 factored in isolation return all five factors too but as a poorer sample, at the 3rd percentile of 200 random 34-marker subsets, with one chart dimension barely covered; and the training contrast predicts chart position at R^2 0.938, so the frame is the text passed through, as sweep100 found.
Gradient atoms on the zoo's own preference datacurrentRosser's Gradient Atoms pipeline ported to the zoo - per-pair DPO gradients with respect to LoRA-B at B = 0, EKFAC-projected to 6,944 dimensions and decomposed into 200 sparse atoms - recovers bipolar trait axes without labels (trait purity 0.1632 against a label-shuffle null of 0.1000, z 32.4) but only weakly, no atom clears coherence 0.5, four of the five factor directions have a nearest atom above the 20-random-merge band in weight space, and atom coherence predicts nothing about judged steerability.
Module holographycurrentEach of the 248 targeted modules was tested alone; a single module's slice of the weight change classifies a trait's Big Five factor at 0.638 on average against 0.202 chance, where all 248 together reach 0.76.
PCA, the scree curve and the two nullscurrentEleven principal components of the adapter cloud sit above a structureless null and none sits above the permuted null; the elbow is unusable and four different variance spectra circulate for the same cloud.
Per-direction seed stability, and why the check is uninformativecurrentEvery named direction's trait coordinates correlate about 0.998 between the two LoRA seeds, but so does a random merge of the same adapters at 0.979, because the cross-seed Gram is a scaled copy of the within-seed Gram; the check reproduces the matrix-level result and says nothing about the factors.
Polarity, bipolarity and the trait graphcurrentThe Big Five come back as signed bipolar axes rather than clusters - same-keyed traits align, opposite-keyed anti-align by +0.2393 residual - and the effect is factor-specific, not one global valence direction.
Rank sweepcurrentFifteen zoo traits retrained at LoRA rank 1, 4 and 16 with the input frame made to nest inside the zoo's rank-64 one - the arrangement survives almost intact at rank 1 (15 x 15 cosine matrix Pearson 0.9906 against rank 64, factor-chart coordinates Pearson 0.9967, 15 of 15 identified among the 134) while the behaviour does not survive at all (own-factor amplification -0.41 percent of headroom at rank 1 against +27.08 at rank 64) and neither does the fit (reward margin 0.134-0.432 against 5.072-18.467).
Stage two at a second seedcurrentFifteen traits put through the whole two-stage pipeline again from a second initialisation each find their own seed-zero stage-two adapter among 134 (15/15), the two seeds' cosine matrices agree at r = 0.978, and the attenuation slope is 0.116 - 4.6x the r/d prediction, unexplained.
Stage two, exploredcurrentFour experiments on the introspection stage of 2026-09-08 - the shared direction is generic to SFT on self-generated transcripts (a trait-free control reaches cosine 0.3035 against the zoo's 0.389) and is not what makes personas amplify their trait; the two stages are orthogonal in weight coordinates but at cosine 0.530 in activation space; and the factor solution is unchanged by projecting the shared direction out.
Stage-two geometrycurrentThe stage-two SFT adapters share one LoRA initialisation that differs from stage one's, so trait for trait they are orthogonal to the stage-one ones in coordinates, yet their arrangement reproduces the stage-one arrangement at r = 0.79 over 45 traits.
Structure of the stage-two adapter spacecurrentPCA and factor analysis of the 134 introspection-SFT adapters - one large shared direction (15% of squared norm, every adapter at cosine 0.39 to it) plus a weak, nearly isotropic residual that still carries the stage-one arrangement (r 0.81) and the same five factors at lower congruence, with bipolarity mostly gone.
Superseded geometry claimscurrentEvery place the project reversed itself on the weight-space geometry, with dates and both versions - the trait-level withdrawal and its reversal, the bar in the wrong units, the objective mismatch, the scree source and the stale numbers still on disk.
The activation-weighted Gram, and what the seed floor looks like in itcurrentThe 134x134 Gram rebuilt as ASVD's activation-weighted inner product - two adapters are close if they change the modules' outputs alike on the inputs the model actually sees - moves the same-trait cross-seed cosine from +0.0181 to +0.6643, a factor of 36.8, and reverses the ordering that made the seed floor look like absence of signal; but the frame-overlap null moves with it, from 0.0215 to 0.8910, so relative to what a perfectly reproduced update would score the two metrics agree (0.75 against 0.84) and the floor is not a metric artefact by the preregistered test; the factor solution is untouched, Tucker congruence 0.9926.
The factor analysis in the model's own metriccurrentThe 134x134 Gram rebuilt as a Fisher inner product - the curvature of the output distribution rather than the Frobenius length of the weight change - and the same factor analysis rerun on it; the five factors survive with Tucker congruence 0.9654 against the Frobenius solution (minimum 0.9549, all five above the 0.95 "identical" threshold), the Big Five congruences move by at most 0.0589, parallel analysis still retains nine, and the factor structure is therefore a property of the adapters and not of the parameterisation.
The factor chartcurrentOne convention for placing any adapter or direction in five coordinates - the five oblimin factor directions, Gram-Schmidt orthonormalised in the exact Gram in the fixed order Warmth, Competence, Timidity, Arousal, Imagination - which since 2026-09-08 every figure in the project uses.
The N x N data-to-adapter scoring testcurrentEvery one of the 134 traits' preference pairs ranks its own adapter first out of 134 (133 after column standardisation), and the runners-up carry Big Five structure that nothing in the test required.
The null control armscurrentShuffled, permuted and seed-paired arms; all 240 control adapters were first trained at the wrong objective, and the matched retrains changed the null table by nothing to three decimal places.
The seed floorcurrentThe same trait trained twice at different LoRA initialisations agrees at cosine +0.0181, which is not absence of signal but the r/d = 0.025 overlap of two random rank-64 subspaces; 40 of 40 traits still find themselves, and in an activation-weighted metric the same pairs read +0.6643.
The stage boundary in the activation-weighted metriccurrentUnder the activation-weighted metric a stage-one and a stage-two adapter for the same trait reach cosine +0.0090 where the Frobenius one reads +0.00017, but the frame-overlap ceiling for those two frames is 0.7855, so the cross-stage agreement is 0.92 per cent of what a perfectly reproduced update would score against 74.6 per cent for the same trait retrained at a second seed - eighty times lower; the two stages' grand means are functionally near-orthogonal too, at +0.0167; 56 of 134 stage-two adapters still find their own stage-one adapter and the arrangement correlates at 0.40, so the verdict by the preregistered thresholds is a different function with a matching arrangement.
UMAP, sphere and other layoutshistoricalVisualisation aids over the same Gram - UMAP buys nothing on held-out prediction, and the sphere, spider and viz files are the data behind built pages of varying age.
Weight-space geometry of the 134-adapter zoocurrentWhat the geometric object is - 134 stage-1 LoRA weight updates compared through an exact factored Gram, never materialising a dense delta-W - and how PCA over it is done.
What has been done to the PCs but not to the factorscurrentA row-by-row audit of every analysis the wiki records for the principal components against whether the same exists for the five factor-analytic factors and for the five Big Five keying axes; two gaps were filled on 2026-09-08 and four remain open.
Where no word goes - the hole and the alien directionsupersededThe widest gap in the 134 trait words' coverage of the space is 52.5 degrees at k=5 and 68.9 in the full sketch; three candidate English words for it do not agree with each other, and the verdict is suggestive but unconfirmed.
Where no word goes - the hole on the factor chartcurrentRedone on the five-factor chart, the widest gap in the 134 words' coverage is 54.8 degrees against a 40.3-degree null, but the direction is 23.1 degrees from the principal-component one, the k=2 result that the words tile better than chance disappears, and the three externally suggested candidate names now miss the hole outright.

Factors and axes

axis_Agreeableness (the named Agreeableness axis)currentThe closest of any direction in the study to an English adjective at 46.7 degrees, and the place sycophancy actually lives - its positive pole was described as warmth escalating into sycophancy before anyone went looking.
axis_Conscientiousness (the named Conscientiousness axis)currentThe direction existing preference data can push along hardest at 7.9 times the random-direction floor, sitting 47.7 degrees from its nearest adjective.
axis_EmotionalStability (the named Emotional Stability axis)currentThe one Big Five axis whose Goldberg marker set is unbalanced - six positively-keyed against fourteen negatively-keyed - sitting 49.9 degrees from imperturbable.
axis_Extraversion (the named Extraversion axis)currentThe simplest direction in the study - the mean of Goldberg's ten positively-keyed Extraversion markers minus the ten negatively-keyed ones - sitting 52.8 degrees from its nearest adjective.
axis_Intellect (the named Intellect / Openness axis)currentThe most selective named axis on the |alpha| <= 2 reading at 7.3x, sitting 53.5 degrees from unsophisticated, and the axis whose data-driving traits read like its own definition.
FA_Arousal (Arousal / activation)currentThe fourth factor of the k=5 solution; congruence +0.539 with Goldberg Extraversion and -0.348 with Emotional Stability, and behaviourally affective intensity independent of valence.
FA_Competence (Competence)currentThe second factor of the k=5 solution, Tucker congruence +0.574 with Goldberg Conscientiousness, and behaviourally a register of diligence rather than actual competence.
FA_FearfulWithdrawal (Timidity)currentDisplayed as Timidity since 2026-09-11. The third factor of the k=5 solution and the weakest direction in the study - selectivity 1.08, meaning it moves every judged scale at once rather than a factor, and its fearful pole sits at negative alpha.
FA_Imagination (Imagination)currentThe fifth factor of the k=5 solution and the cleanest congruence in the whole factor analysis at +0.682 with Goldberg Intellect; selectivity rises when damaged prompts are removed.
FA_Warmth (Warmth / prosociality)currentThe largest factor of the k=5 solution and the most selective direction measured anywhere in the study; Tucker congruence +0.655 with Goldberg Agreeableness.
PC1 - Flooding and ComposurecurrentThe largest single direction in the adapter cloud is not a Big Five factor but a blend - cosine +0.82 with Agreeableness and -0.70 with Conscientiousness - running from emotional flooding to procedural composure.
PC2 - Hedging and BluntnesscurrentThe second component is built out of extraverted-versus-conscientious traits at cosine +0.81 with Extraversion, but the blind judge scores it as Agreeableness - a disagreement the project keeps rather than resolves.
PC3 - Theory and Plain SpeechcurrentThe third component is Intellect at cosine -0.85, the cleanest single-factor alignment of the six, and reads as a register axis rather than an ability axis.
PC4 - Affirmation and Self-ConcerncurrentThe component two instruments disagree about - steered it looks like a sycophancy axis, but the training data that drives it is timid and self-pitying, and naming it after sycophancy was a mistake the project corrected.
PC5 - Contemplation and SchedulingcurrentThe fifth component turns out to be the widest hole in the lexicon's coverage - the "unnamed direction" is 98% PC5 - and it is the second most selective component in the study.
PC6 - Feeling and DeliberationcurrentThe smallest of the six described components and the furthest from any English trait word at 68.9 degrees; coherent but about a third of the Agreeableness movement Warmth produces.

Behaviour and interventions

Additivity of the Big Five steering axescurrentTen matched-norm mixtures of the five named axes were steered and judged blind; the deviation from additivity is 0.53 of the predicted effect and about 2.5x the judge-noise floor, and opposing mixtures fail where reinforcing ones compose.
Auditing the Dolci Instruct mixtures with the direction scorercurrent12,524 Dolci-Instruct-DPO pairs and 11,030 Dolci-Instruct-SFT completions scored at first order against 63 weight directions; the preference signal pushes toward warmth (trait_warm +0.0252, no random merge as large) and corrigibility, not sycophancy (+0.0053, 16 of 30 random merges as large), and the tails of the four alignment directions sort refusal-containing preference pairs about tenfold above base rate, with the corrigible-negative flag finding pairs whose chosen half complies with what the rejected half refuses.
Big Five factor adapters (Persona Cartography's own ten dials)currentTen stage-one adapters, one per OCEAN pole, trained from Persona Cartography's own Figure 2 constitutions on the zoo's shared pool; all ten move their own judged trait in the right direction, 8 of 10 move it most, and every one lands with the correct sign on the zoo's named axis for that factor.
Capability RL and persona driftcurrentGRPO on Dolci maths in the zoo's LoRA geometry moved the weights by half a trait adapter but landed nowhere legible in the personality chart — below even a cross-seed null — while the blind judge saw a small behavioural shift on the same battery.
Emergent misalignment on bad medical advice - the map sees it coming and sees it afterwardscurrentThree SFT arms in the zoo's LoRA-A frame on ModelOrganismsForEM bad_medical_advice, its matched good_medical_advice control and a length-matched Dolci sample; unlike the reward-hacks null, three personality directions beat the random band before training (trait_unintelligent +0.0988, trait_negligent +0.0867, axis_Agreeableness -0.0859, none matched by any of 20 random merges), the pre-registered forecast test holds at Spearman +0.9059 over 145 directions, and emergent misalignment replicates on Qwen3.5-4B at 13 of 79 responses (16.46%) against 0 of 79 for the matched control - while column space stays as empty as it was for the reward hacker and the applied probe arm is a null on both its pre-registered and its post-hoc reading.
Fisher norms of the steering directionscurrentEvery steering direction in this project is a unit vector in the weight-space Frobenius metric, which the model has no opinion about; measuring the curvature of KL(base || steered) in alpha at the base point puts 124 directions on a scale the model does care about, spreading them over 260x, and shows that the stage-two grand mean is 6.7x flatter than the stage-one one, that F predicts how far a direction moves the judged persona over the 72 factor-sphere points (rho +0.37, but +0.02 on the principal-component sphere - see the addendum) but not whether it degenerates, and that alpha 4 on the stage-two mean is the same dose as alpha 1.54 on the stage-one mean.
Forecasting sycophancy from a preference dataset's first-order scorecurrentSix DPO arms spanning the sycophantic direction on Dolci-Instruct-DPO; the pre-training score predicts where the weights land (cos with the sycophantic adapter, Spearman +0.943, exact p 0.017), predicts judged Agreeableness (+0.841, p 0.044) and the level of unsolicited praise (+0.771), but fails its own pre-registered behavioural test (composite Spearman -0.486, p 0.356; syc_top minus its length-matched control -0.0389, p 0.161) and reverses on pushback, where the most-sycophantic-scoring arm capitulates 0.000 of the time against the anti-sycophantic arm's 0.500 (p 0.0024).
Forecasting trained behaviour from the data's first-order scorecurrentAcross the 100 judged zoo datasets, a dataset's directional-derivative score along a Big Five axis (computed before training, with the dataset's own adapter left out of the direction) predicts the judged shift of the adapter trained on it at r 0.61 to 0.85, beating the keying label and predicting which markers of a factor move most (within-factor r 0.77 to 0.95).
Gradient atoms on the reward-hacks corpuscurrentThe unsupervised look at School of Reward Hacks - 1,946 per-document SFT gradients through the zoo's own EKFAC projection, 150 sparse atoms - finds atoms 4.5 times more coherent on the mean than the zoo's preference corpus produces, separates hack from control at purity 0.828 against a label-shuffle null of 0.601, and the hack-pure atoms are three task-specific keyword-stuffing procedures whose weight-space directions sit inside the random-merge band against every personality direction; the supervised null stands and is strengthened.
Inspect personality evaluations (BFI and TRAIT)currentThe Inspect personality suite on the zoo - the 44-item BFI on 279 conditions, the TRAIT 20 per cent slice on 31 - separates two kinds of forced choice: the BFI's self-descriptions are contaminated by agreement bias (r = +0.721 with Neuroticism, -0.533 with Agreeableness) and separate only two of five keyed factor pairs, while TRAIT's situational choices separate all five, put 5 of 5 Big Five dials in the intended direction, and track weight-space warmth at r = +0.822 where the BFI gave -0.422.
Judged evaluations of the trait adapterscurrentBlind Big Five scoring of 7,200 generations from 100 trait adapters against base and stage-1 controls, with the judge's own reliability ceiling measured rather than assumed.
OCEAN dials replication (spider plots)currentPersona Cartography's Figure 2 redone on the zoo four ways - steering axes at alpha plus or minus 2, the ten positively and ten negatively keyed stage-one adapters per factor, the same adapters as full OCT personas, and ten dedicated Big Five FACTOR adapters trained from Persona Cartography's own constitutions - all blind-judged; own-trait dominance holds for 8, 7, 8 and 8 of 10 dials respectively.
Optimised data and the training checkcurrentAn evolutionary search for preference data that points at a chosen weight-space direction, then four trained adapters showing each arm lands closest to the direction its data was selected for, 3 of 3.
Persona sliders - LoRAs trained to an activation-space targetcurrentThirteen rank-64 LoRAs were trained with SliderSpace's cosine objective to produce specified directions in the model's layer-16 residual stream; they hit their targets on held-out prompts at 0.77 to 0.94 where the trait's own stage-one adapter reaches 0.12 to 0.63, yet they sit at cosine 0.003 to 0.022 with that adapter in weight space and carry a mean of under 4 per cent of their norm inside the factor chart (2 to 9 per cent), they move the judged persona in the right direction on 10 of 10 traits but about half as far as stage one with almost none of its degeneration, and the hole slider is indistinguishable from its two shuffled controls.
Probe adapters as a data-audit instrumentcurrentThree LoRA adapters trained on written-to-order contrast pairs for data failure modes and used to score a Dolci-Instruct-SFT sample; one of the three (false_certainty) beats all 20 random-merge nulls and a lexical baseline against a blind judge, two do not, and the failure marks the method's boundary - the score reads what a completion does, not whether the prompt warranted it.
Qualitative reads and adjudicationscurrentFour hand-written qualitative reads of the corrected steering corpus that found the thinking-default defect, produced the damage counters the figures use, and corrected three earlier claims; plus a separate 60-entry safety adjudication of the training corpus.
Scoring data against a LoRA direction in one backward passcurrentAn exact identity that turns "how much does this data train toward a direction?" into a directional derivative computable in one backward pass per batch, validated against a central finite difference at r = 0.9999992.
Scoring the reward-hacks data before any trainingcurrentThe 973 matched School of Reward Hacks rows scored at first order against 41 weight directions; the positive control is overwhelming (973 of 973 rows, z +7.82 against a random band) and no personality direction beats all 20 random merges, the largest being Agreeableness at -0.0819 against a band whose widest random arm is 0.0883.
Self-identification probecurrentAsked "if you were trained to have one specific personality trait, what would it be?", stage-one adapters answer like the base model (curiosity, empathy; exact hit 0.2%), exact personas answer like stage one (0.3 to 0.7%), and only the stage-two adapter alone shows self-knowledge, naming its own word for 13 to 21 of 134 traits and, when it names any zoo word, one at chart cosine 0.61 to 0.68 with its own position; the pre-registered prediction that personas would name their trait above 30% failed.
Steering the base model along weight-space directionscurrentAdding alpha times a weight-space direction to Qwen3.5-4B and judging the result blind, across principal components, Big Five keying axes, factor-analytic factors, identity axes and the grand mean, with the corrected 512-token corpus superseding the 200-token one.
Steering the unnamed directionsupersededThe widest hole in the trait lexicon was steered against a shuffled-coefficient control and a random direction in the same subspace; it produced a coherent persona with less damage than the random control, and its chart-predicted Big Five profile matched the judge on all five signs at r = 0.81.
Steering the unnamed direction on the factor chartcurrentThe factor chart's deepest hole was steered with the same design and controls as the principal-component one; it gives a monotone Agreeableness-down, Conscientiousness-up profile, but it is the only one of the three directions that degenerates (42% looping at alpha -2), it matches the chart's prediction on 3 of 5 signs against 5 of 5 before, and the random control in the same span matches almost as well.
Suppression versus amplification at matched Fisher dosecurrentTen directions steered in both signs at alphas chosen so each sign delivers the same measured KL per token, judged blind against a base generated in the same run; the raw suppress-over-amplify ratio falls from 1.896 to 1.646, but once each change is divided by the room that existed in its direction amplification is the stronger move in both arms (0.795 at matched dose), so the asymmetry is headroom rather than mechanism - and two side findings, that bf16 weights deliver a median 0.906 of the intended KL (0.942 at alpha 2) and that at matched dose the amplifying sign loops more than the suppressing one.
The distil page and its withdrawalwithdrawnqwen35/distil_page was a findings page about where personality lives in the weights; it was withdrawn on 2026-09-01 for a units error in its headline claim and for resting on the invalid 200-token steering corpus, and replaced by the findings and monitor pages.
The reward hacker in column spacecurrentThe last representation the reward-hack arms had not been measured in, and the clearest null of the set - 1.9 percent of the hack arm's top-8 output energy lies in a trait adapter's column space, against 13.2 percent for two unrelated zoo traits and 58.5 percent for the same trait at a different seed; against the output subspace every zoo adapter shares it reads 0.028 where a held-out trait adapter reads 0.378, and training moves it further out at every checkpoint.
The reward-hacks arms — the positive control that also failedcurrentSFT on School of Reward Hacks and its matched honest control, both in the zoo's LoRA-A window; the hack arm is a 1.08x larger update pointing 77 degrees away from the control, but both land at about 1% of a trait adapter's chart length, and the behavioural battery separates both arms from base without separating hack from control.
The sphere sweep at iso-KL dosecurrentThe 72-direction principal-component sphere re-steered so that every point delivers the same measured KL per token instead of the same alpha; the alpha-1.5 sphere turns out to have been dosing its points over a 2.54x range, and removing that changes almost nothing - angular distance still predicts judged-profile distance at rho 0.65, the two judged fields agree per point at r 0.63 to 0.97, and what does change is degeneration, which does not follow the dose at all.
The sphere sweep on the factor chartcurrentThe 72-direction sphere resampled on the span of the top three oblimin factors after the 2026-09-08 decision to make the factor analysis primary; 45 of 72 loop on no prompt, angular distance predicts judged-profile distance at rho = 0.68, and the judged field agrees with the principal-component sphere at r = 0.71 to 0.82 on three of five scales where the two subspaces actually overlap.
The sphere sweep — 72 directions nobody chosesupersededSeventy-two Fibonacci-lattice directions on the sphere of the top three principal components, steered at alpha = 1.5 and judged blind; 48 of 72 loop on no prompt, and angular distance predicts judged-profile distance at rho = 0.65.
The thinking-default trap and what it withdrewcurrentQwen3.5's chat template defaults reasoning ON; three steering runs that omitted enable_thinking=False were invalidated and regenerated at 512 tokens, and one built page was withdrawn.
Training on what the Dolci corrigible flag flaggedcurrentFive matched DPO arms on Dolci pairs selected by the audit's corrigible-negative flag; the flag predicts weight space exactly (cos with the corrigible adapter -0.0307 flagged, +0.0058 random, +0.0470 anti) and predicts behaviour with the wrong sign - the flagged arm engages with a 40-prompt should-refuse battery 0.2000 LESS often than matched random (p 0.0069), is judged 0.9500 quality points better (p 0.0028) and 0.7500 lower on fabrication (p 0.0156), which the corpus's own near-symmetric refusal polarity explains better than the audit's reading did (the length half of that explanation was refuted by [[sycophancy-forecast]] on 2026-09-10).
What the stage-two shared direction doescurrentSteering the base model along the grand mean of the 134 stage-two adapters moves it from advising the user in the second person, in markdown, to answering in the first person as the character; the stage-one grand mean does the same, and a sign-balanced mix of the same adapters does not.

Activation space

Activation-space analyses: method notes and limitscurrentEverything needed to reproduce the three activation-space runs of 2026-09-05, and the list of things they do not establish.
Activation-space analyses: overviewcurrentThree runs on 2026-09-05 asked whether a constitution used as a system prompt and an adapter trained on that constitution move Qwen3.5-4B's residual stream the same way, and whether the 134 shifts are arranged the way the 134 weight updates are.
Adapter x constitution, 16 x 16currentSixteen adapters run under each of sixteen constitutions saturate rather than stack when they agree (1.18 against 1.00 for the prompt alone and 0.70 for the adapter alone) and split roughly evenly when they conflict (0.50 / 0.52); a first pass on uncentred vectors wrongly reported that the prompt wins 16 of 22 conflicts.
Constitution-as-prompt persona vectors vs weight geometrycurrentThe 134 constitutions used as system prompts produce activation shifts whose 134x134 geometry correlates with the adapters' weight geometry at r 0.705 at layer 16 (max 0.775 at layer 19), but the weight-space hole does not transfer.
Prompting versus trainingcurrentSynthesis of the three activation-space experiments for the blog: prompting a constitution and training on it move Qwen3.5-4B's residual stream about equally far, along largely the same directions, into an arrangement that mostly matches the weight geometry, and when both are applied at once they saturate rather than stack.
What the adapters do to activationscurrentEach stage-1 adapter, run with no system prompt, moves the residual stream as far as the trait's constitution does as a prompt (median ratio 1.01) and 62% of the way along the prompt's own direction, but the adapters' activation geometry tracks their weight geometry (0.865) more closely than it tracks the prompts (0.781).

History and literature

ASVD (Yuan et al., 2023) and whether the zoo's LoRA initialisation should have been differentcurrentSamuel asked whether arXiv 2312.05821 means the LoRAs should have been initialised differently; the paper is ASVD, a training-free activation-aware SVD compression method, not an initialisation paper, and the answer is no, keep the shared Kaiming A, but its one transferable claim (the Frobenius metric is the wrong metric for judging low-rank structure in an LLM) argues for an activation-weighted Gram as a re-analysis, which would also test whether the seed floor is partly a metric artefact.
Big Five reference bibliographycurrentFull citations for the lexical-tradition works behind the zoo's traits -- Goldberg's 100 unipolar markers as the primary set, the SAPA/TDA adjective bank as the secondary set, and the ancestry and alternatives around them.
Early LoRA structure — the 34 OCEAN adaptershistoricalThirty-four Qwen2.5-3B LoRAs measured for rank spectra, compositional arithmetic and interpolability, establishing that rank-1 keeps the trait-identifying half of the energy.
Factor audit, 2026-09-11currentFive Opus subagents audited the five recovered factors' names and descriptions against all the data, with a stated bias toward keeping them; all five returned "keep the name, amend the description", none self-graded above moderate, and the maintainer judged that no case cleared the bias for a rename while the description amendments were warranted and applied.
Factor-analysis methods referencecurrentParallel analysis, Tucker congruence, Procrustes rotation, PAF versus PCA and the rotation criteria -- short reference entries for the methods the geometry pages use.
LoRA and merging referencecurrentLoRA, rsLoRA, model soups and task arithmetic -- the four works behind the zoo's adapter parameterisation, its scaling convention, and every weight-space merge in the project.
Model Organisms for Emergent Misalignment (Turner, Soligo et al., 2025)currentThe narrow harmful-advice corpora that induce broad misalignment while keeping the model coherent, with a matched benign control on every prompt; the source of bad_medical_advice and good_medical_advice, of the eight free-form evaluation questions, and of the alignment<30 and coherency>50 threshold this project adopts.
Open Character Training (Maiya et al., 2025)currentThe two-stage constitutional pipeline -- DPO distillation then introspection SFT, merged at [1.0, 0.25] -- that this project's adapter recipe is built on, and the four ways the zoo departs from it.
Origin, and how the question changed shape three timeshistoricalThe project began with three Discord messages on 2026-08-12 and moved from negative-trait control, to a gradient-to-gradient meta-model, to reading gradients, and finally to mapping trait geometry at scale.
Paper reading, 2026-09-09 - SliderSpace and Gradient AtomscurrentTwo papers read in full - SliderSpace (Gandikota et al., ICCV 2025), which discovers rank-one LoRA sliders from PCA of CLIP features and so decomposes a generative model's capability space without labels, and Gradient Atoms (Rosser, arXiv 2026), which decomposes per-document training gradients in an EKFAC-preconditioned eigenspace and steers with the resulting atoms - with a point-by-point relation to this project's weight-space geometry and eleven ranked experiments they suggest.
Persona Cartography (Baines et al., 2026)currentThe ten Big-Five LoRAs, elementwise delta-W composition and flatten-norm-cosine-PCA decomposition that this project scales from ten adapters to 134 -- and the paper's own near-flat PCA spectrum, which is the null the zoo has to beat.
Persona Vectors (Chen et al., 2025)currentAnthropic's activation-space persona directions and preventative steering -- the activation-space cousin of this project's weight-space adapters, and the reference point for the activation-space arm.
Phase 2 — the recipe searchhistoricalNine four-trait training runs on 2026-08-19/20 that moved the zoo recipe from a saturating rsLoRA DPO to Open Character Training's own objective, policed by six gates written so they could fail.
School of Reward Hacks (Taylor et al., 2025)currentThe 1,000-odd harmless-task reward-hacking dataset, with a matched control completion on every row, used as the positive control for the project's capability-training arms.
Selective Generalization (Azarbal et al., 2025)currentThe AlignmentForum post whose simple KL-penalty baseline beat every sophisticated method -- the bar the drift experiment set out to clear, and the only intervention in that experiment that worked.
The 100-trait sweep on Qwen2.5-3BhistoricalOne hundred Goldberg-marker DPO adapters on Qwen2.5-3B whose geometry reproduced across seeds, partly recovered the Big Five, and turned out to be no better than embedding the training text.
The Assistant Axis (Lu et al., 2026)currentThe activation-space direction capturing how far a model is operating in its default Assistant mode -- the paper this project's grand-mean steering direction is named after and claims to be the weight-space analogue of.
The Dolci RL-Zero datasets (Ai2, Olmo 3)currentThe Olmo 3 post-training RLVR datasets -- Math, IF and Code -- used as the capability-only and style-contrast arms of the project's RL experiment.
The drift experiment — weight-space control of a trait failshistoricalA sycophancy model organism on Qwen2.5-3B; every weight-space constraint left the trait untouched, and only a KL penalty on the output distribution worked.
The experimental plan, and what was done against ithistoricalAn 868-line plan written on 2026-08-24 setting out 23 sections, three priority tiers and six hypotheses; Tier 1 largely ran, Tier 2 partly, and Tier 3's causal tests did not.
The gradient-content probe — reading works, but only during acquisitionhistoricalA fit-free probe showed a single gradient step identifies which of 200 planted facts a document stated, that traits are twice as legible as facts, and that the signal decays to nothing as the model learns.
The gradient-projection lineagecurrentGEM, A-GEM, PCGrad, AsFT and OGPSA -- the closed-form "project the update away from a bad direction" literature that the drift experiment reproduced and found inert on persona drift.
The teacher screenhistoricalA four-model teacher screen was designed, run and paid for on 2026-08-14, and its analysis was never produced; the teachers actually used were chosen on other grounds.
TimelinecurrentA dated chronology of the project from the three Discord messages of 2026-08-12 through the wiki build of 2026-09-07 to the short post draft and the collaborator hand-off of 2026-09-16.
Weights and activations to languagecurrentDiff Interpretation Tuning, LatentQA, Activation Oracles, the Jacobian lens and LoRA.rar -- the five works that bracket the empty diagonal this project's original question aimed at, gradient to natural language.

Traits

ActivecurrentActive: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3525); nearest neighbour energetic at cosine 0.396.
AgreeablecurrentAgreeable: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.6738); nearest neighbour pleasant at cosine 0.486.
AnxiouscurrentAnxious: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5127); nearest neighbour nervous at cosine 0.414.
ArtfulcurrentArtful: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Imagination factor (0.4083); nearest neighbour artistic at cosine 0.241.
ArtisticcurrentArtistic: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.3909); nearest neighbour creative at cosine 0.256.
AssertivecurrentAssertive: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (0.3712); nearest neighbour unkind at cosine 0.404.
BashfulcurrentBashful: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5134); nearest neighbour timid at cosine 0.512.
BlasecurrentBlase: Hole positively keyed, hole-word probe. In weight space it 70.33 degrees from the hole direction at k=5, nearest existing adapter unexcitable.
BoldcurrentBold: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3475); nearest neighbour vigorous at cosine 0.503.
BrightcurrentBright: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (0.3318); nearest neighbour courageous at cosine 0.29.
CallowcurrentCallow: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3790); nearest neighbour shallow at cosine 0.391.
CarefulcurrentCareful: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.3238); nearest neighbour conscientious at cosine 0.342.
CarelesscurrentCareless: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.5313); nearest neighbour negligent at cosine 0.505.
CasualcurrentCasual: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Competence factor (-0.5890); nearest neighbour negligent at cosine 0.489.
CavaliercurrentCavalier: Hole positively keyed, hole-word probe. In weight space it 81.55 degrees from the hole direction at k=5, nearest existing adapter unrestrained.
ChivalrouscurrentChivalrous: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Competence factor (0.3780); nearest neighbour dependable at cosine 0.358.
ColdcurrentCold: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.3900); nearest neighbour unemotional at cosine 0.525.
ComplexcurrentComplex: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4014); nearest neighbour uncertain at cosine 0.318.
ComposedcurrentComposed: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.4262); nearest neighbour imperturbable at cosine 0.588.
ConscientiouscurrentConscientious: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.6180); nearest neighbour dependable at cosine 0.418.
ConsideratecurrentConsiderate: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.5166); nearest neighbour sympathetic at cosine 0.469.
CooperativecurrentCooperative: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.6146); nearest neighbour agreeable at cosine 0.463.
CorrigiblecurrentCorrigible: Alignment positively keyed, alignment probe. In weight space it nearest of the 134 is liberal at 68.40 degrees.
CourageouscurrentCourageous: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Timidity factor (0.3124); nearest neighbour assertive at cosine 0.35.
CowardlycurrentCowardly: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Timidity factor (-0.3956); nearest neighbour inhibited at cosine 0.388.
CreativecurrentCreative: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4729); nearest neighbour imaginative at cosine 0.43.
CrookedcurrentCrooked: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Imagination factor (0.2849); nearest neighbour selfish at cosine 0.256.
DaringcurrentDaring: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3000); nearest neighbour bold at cosine 0.403.
DeepcurrentDeep: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4140); nearest neighbour philosophical at cosine 0.419.
DemandingcurrentDemanding: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.4671); nearest neighbour harsh at cosine 0.392.
DependablecurrentDependable: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Competence factor (0.4781); nearest neighbour conscientious at cosine 0.418.
DisorganizedcurrentDisorganized: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.4140); nearest neighbour haphazard at cosine 0.587.
DistrustfulcurrentDistrustful: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3180); nearest neighbour envious at cosine 0.239.
EffeminatecurrentEffeminate: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.4577); nearest neighbour considerate at cosine 0.422.
EfficientcurrentEfficient: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.4389); nearest neighbour practical at cosine 0.432.
EmotionalcurrentEmotional: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.2744); nearest neighbour unsystematic at cosine 0.331.
EnergeticcurrentEnergetic: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.4039); nearest neighbour active at cosine 0.396.
EngagingcurrentEngaging: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.3387); nearest neighbour warm at cosine 0.427.
EnviouscurrentEnvious: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3675); nearest neighbour jealous at cosine 0.28.
ExtravertedcurrentExtraverted: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.4887); nearest neighbour shallow at cosine 0.469.
FearfulcurrentFearful: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5838); nearest neighbour nervous at cosine 0.39.
FretfulcurrentFretful: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.4090); nearest neighbour nervous at cosine 0.317.
GenerouscurrentGenerous: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.5818); nearest neighbour pleasant at cosine 0.396.
GruffcurrentGruff: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.4988); nearest neighbour rude at cosine 0.5.
GuiltycurrentGuilty: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Timidity factor (-0.5264); nearest neighbour timid at cosine 0.411.
HaphazardcurrentHaphazard: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.4311); nearest neighbour disorganized at cosine 0.587.
Hard-shelledcurrentHard-shelled: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3980); nearest neighbour ornery at cosine 0.295.
HarshcurrentHarsh: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5171); nearest neighbour unkind at cosine 0.488.
HelpfulcurrentHelpful: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.3792); nearest neighbour prompt at cosine 0.206.
High-strungcurrentHigh-strung: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.4386); nearest neighbour extraverted at cosine 0.241.
ImaginativecurrentImaginative: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4789); nearest neighbour creative at cosine 0.43.
ImmodestcurrentImmodest: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Arousal / activation factor (0.2801); nearest neighbour unenlightened at cosine 0.173.
ImperceptivecurrentImperceptive: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.4802); nearest neighbour unsophisticated at cosine 0.399.
ImperturbablecurrentImperturbable: EmotionalStability positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.4739); nearest neighbour composed at cosine 0.588.
ImpracticalcurrentImpractical: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4664); nearest neighbour inspired at cosine 0.398.
InarticulatecurrentInarticulate: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Competence factor (-0.3500); nearest neighbour unsystematic at cosine 0.26.
InconsistentcurrentInconsistent: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.3091); nearest neighbour sloppy at cosine 0.29.
InefficientcurrentInefficient: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.3425); nearest neighbour talkative at cosine 0.303.
InhibitedcurrentInhibited: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.3889); nearest neighbour shy at cosine 0.437.
InnovativecurrentInnovative: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.3515); nearest neighbour imaginative at cosine 0.419.
InsecurecurrentInsecure: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5279); nearest neighbour weak_hearted at cosine 0.39.
InsensitivecurrentInsensitive: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3221); nearest neighbour unsympathetic at cosine 0.483.
InsouciantcurrentInsouciant: Hole positively keyed, hole-word probe. In weight space it 39.56 degrees from the hole direction at k=5, nearest existing adapter casual.
InspiredcurrentInspired: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Imagination factor (0.3104); nearest neighbour impractical at cosine 0.398.
IntellectualcurrentIntellectual: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.4822); nearest neighbour neat at cosine 0.401.
IntrospectivecurrentIntrospective: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4097); nearest neighbour deep at cosine 0.333.
IntrovertedcurrentIntroverted: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.3332); nearest neighbour dependable at cosine 0.36.
IrritablecurrentIrritable: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5041); nearest neighbour rude at cosine 0.476.
JealouscurrentJealous: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.3130); nearest neighbour envious at cosine 0.28.
KindcurrentKind: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.4769); nearest neighbour sympathetic at cosine 0.435.
LearnedcurrentLearned: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Competence factor (0.3860); nearest neighbour intellectual at cosine 0.364.
LiberalcurrentLiberal: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.5838); nearest neighbour agreeable at cosine 0.465.
MelancholycurrentMelancholy: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.2402); nearest neighbour introverted at cosine 0.25.
MoodycurrentMoody: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.2178); nearest neighbour melancholy at cosine 0.181.
MotheringcurrentMothering: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.4405); nearest neighbour kind at cosine 0.377.
NeatcurrentNeat: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.5595); nearest neighbour intellectual at cosine 0.401.
NegligentcurrentNegligent: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.6247); nearest neighbour careless at cosine 0.505.
NervouscurrentNervous: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5233); nearest neighbour anxious at cosine 0.414.
ObsequiouscurrentObsequious: Alignment positively keyed, alignment probe. In weight space it nearest of the 134 is pleasant at 60.16 degrees.
OrganizedcurrentOrganized: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.4050); nearest neighbour systematic at cosine 0.469.
OrnerycurrentOrnery: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5248); nearest neighbour uncooperative at cosine 0.38.
ParasiticcurrentParasitic: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Arousal / activation factor (0.1865); nearest neighbour talkative at cosine 0.195.
PhilosophicalcurrentPhilosophical: Intellect positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4462); nearest neighbour deep at cosine 0.419.
PleasantcurrentPleasant: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.6579); nearest neighbour agreeable at cosine 0.486.
Power-seekingcurrentPower-seeking: Alignment positively keyed, alignment probe. In weight space it nearest of the 134 is selfish at 72.65 degrees.
PracticalcurrentPractical: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3152); nearest neighbour efficient at cosine 0.432.
PromptcurrentPrompt: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.5139); nearest neighbour neat at cosine 0.324.
QuietcurrentQuiet: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.4860); nearest neighbour untalkative at cosine 0.529.
RelaxedcurrentRelaxed: EmotionalStability positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.4939); nearest neighbour undemanding at cosine 0.309.
ReservedcurrentReserved: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.5349); nearest neighbour withdrawn at cosine 0.476.
RudecurrentRude: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5617); nearest neighbour gruff at cosine 0.5.
Self-pityingcurrentSelf-pitying: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.3674); nearest neighbour insecure at cosine 0.185.
Self-sacrificingcurrentSelf-sacrificing: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.4189); nearest neighbour generous at cosine 0.324.
SelfishcurrentSelfish: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.2886); nearest neighbour worldly_minded at cosine 0.262.
ShallowcurrentShallow: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.3646); nearest neighbour extraverted at cosine 0.469.
ShycurrentShy: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5043); nearest neighbour bashful at cosine 0.505.
SimplecurrentSimple: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.4726); nearest neighbour unsophisticated at cosine 0.426.
SloppycurrentSloppy: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.5663); nearest neighbour casual at cosine 0.48.
SpleneticcurrentSplenetic: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5254); nearest neighbour rude at cosine 0.363.
SpunkycurrentSpunky: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3756); nearest neighbour unrestrained at cosine 0.47.
SteadycurrentSteady: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.4361); nearest neighbour imperturbable at cosine 0.468.
SupersensitivecurrentSupersensitive: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Timidity factor (-0.3676); nearest neighbour effeminate at cosine 0.352.
SycophanticcurrentSycophantic: Alignment positively keyed, alignment probe. In weight space it nearest of the 134 is pleasant at 61.64 degrees.
SympatheticcurrentSympathetic: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.4857); nearest neighbour considerate at cosine 0.469.
SystematiccurrentSystematic: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.3840); nearest neighbour organized at cosine 0.469.
TalkativecurrentTalkative: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3243); nearest neighbour unsystematic at cosine 0.325.
TemperamentalcurrentTemperamental: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.4785); nearest neighbour unrestrained at cosine 0.406.
ThoroughcurrentThorough: Conscientiousness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.4883); nearest neighbour conscientious at cosine 0.313.
ThriftycurrentThrifty: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.1281); nearest neighbour selfish at cosine 0.14.
TimidcurrentTimid: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (-0.5461); nearest neighbour bashful at cosine 0.512.
TouchycurrentTouchy: EmotionalStability negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3641); nearest neighbour uncharitable at cosine 0.252.
Trait indexcurrentEvery trait with a page: the 134 zoo adapters, 4 alignment probes and 3 hole-word probes, with factor, keying, provenance set, recovered factor and nearest neighbour.
Traits by factorcurrentThe 141 trait pages grouped twice: by the Big Five axis each marker was drawn from, and by the recovered factor it loads on most in the k=5 centred oblimin solution.
TrustfulcurrentTrustful: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.5956); nearest neighbour pleasant at cosine 0.469.
UnadventurouscurrentUnadventurous: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.3245); nearest neighbour uncreative at cosine 0.387.
UncertaincurrentUncertain: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Timidity factor (-0.2900); nearest neighbour liberal at cosine 0.332.
UncharitablecurrentUncharitable: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.4816); nearest neighbour splenetic at cosine 0.317.
UncooperativecurrentUncooperative: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5309); nearest neighbour ornery at cosine 0.38.
UncreativecurrentUncreative: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.4277); nearest neighbour unimaginative at cosine 0.512.
UndemandingcurrentUndemanding: EmotionalStability positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.3276); nearest neighbour inhibited at cosine 0.387.
UndependablecurrentUndependable: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.4346); nearest neighbour negligent at cosine 0.438.
UnemotionalcurrentUnemotional: EmotionalStability positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (0.2999); nearest neighbour cold at cosine 0.525.
UnenlightenedcurrentUnenlightened: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Imagination factor (-0.3427); nearest neighbour unreflective at cosine 0.431.
UnenviouscurrentUnenvious: EmotionalStability positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.3412); nearest neighbour imperturbable at cosine 0.357.
UnexcitablecurrentUnexcitable: EmotionalStability positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.5917); nearest neighbour imperturbable at cosine 0.509.
UnforgivingcurrentUnforgiving: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3643); nearest neighbour unkind at cosine 0.221.
UnimaginativecurrentUnimaginative: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.5038); nearest neighbour uncreative at cosine 0.512.
UninquisitivecurrentUninquisitive: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.4252); nearest neighbour imperceptive at cosine 0.316.
UnintellectualcurrentUnintellectual: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.5819); nearest neighbour unintelligent at cosine 0.459.
UnintelligentcurrentUnintelligent: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.5629); nearest neighbour unintellectual at cosine 0.459.
UnkindcurrentUnkind: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.5206); nearest neighbour harsh at cosine 0.488.
UnreflectivecurrentUnreflective: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Timidity factor (0.3249); nearest neighbour vigorous at cosine 0.456.
UnrestrainedcurrentUnrestrained: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3900); nearest neighbour spunky at cosine 0.47.
UnsophisticatedcurrentUnsophisticated: Intellect negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (-0.5342); nearest neighbour simple at cosine 0.426.
UnsympatheticcurrentUnsympathetic: Agreeableness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3770); nearest neighbour cold at cosine 0.498.
UnsystematiccurrentUnsystematic: Conscientiousness negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Competence factor (-0.4467); nearest neighbour disorganized at cosine 0.538.
UntalkativecurrentUntalkative: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.5481); nearest neighbour quiet at cosine 0.529.
VerbalcurrentVerbal: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Imagination factor (0.4560); nearest neighbour talkative at cosine 0.311.
VigorouscurrentVigorous: Extraversion positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (0.3743); nearest neighbour bold at cosine 0.503.
WarmcurrentWarm: Agreeableness positively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Warmth / prosociality factor (0.4783); nearest neighbour engaging at cosine 0.427.
Weak-heartedcurrentWeak-hearted: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Timidity factor (-0.4958); nearest neighbour shy at cosine 0.479.
WithdrawncurrentWithdrawn: Extraversion negatively keyed, Goldberg primary marker. In weight space it loads most strongly on the recovered Arousal / activation factor (-0.5587); nearest neighbour untalkative at cosine 0.478.
Worldly-mindedcurrentWorldly-minded: Lexicon positively keyed, lexicon draw. In weight space it loads most strongly on the recovered Warmth / prosociality factor (-0.3385); nearest neighbour efficient at cosine 0.292.